Re: Debian packages advertising non-free services

2017-08-08 Thread Andrey Rahmatullin
On Tue, Aug 08, 2017 at 01:47:29AM -0400, Scott Kitterman wrote:
> I think the general rule for software with a FOSS license that has no other 
> purpose than to interact with proprietary services is to put it in contrib.  
The usual example from main that gets mentioned in these discussions is
ICQ/MSN clients, but now we have more examples like already mentioned
s3cmd.

-- 
WBR, wRAR


signature.asc
Description: PGP signature


Re: Debian packages advertising non-free services

2017-08-07 Thread Charles Plessy
Le Tue, Aug 08, 2017 at 01:47:29AM -0400, Scott Kitterman a écrit :
> 
> I think the general rule for software with a FOSS license that has no other 
> purpose than to interact with proprietary services is to put it in contrib.  

Hi Scott and everybody,

I think that unfortunately there is no clear definition in Debian - hence no
consensus - of what is a "proprietary service".  In the case of s3cmd the
Amazon cloud runs proprietary software, however it is accessed by a public and
stable API, that was actually re-implemented in part by some Free software
projects.  So unless prefer to move packages back and forth to the contrib
area, we should perhaps focus on the API rather than on the remote
implementation ?

It would be a strong message to the World if Debian would manage to update its
foundation documents to recognise the fact that we increasingly rely on remote
services and that we need them to provide freedoms equivalent to what we
guarantee with DFSG.  The current discussion just scratches the surface...

Have a nice day,

-- 
Charles Plessy
Tsurumi, Kanagawa, Japan



Re: Debian packages advertising non-free services

2017-08-07 Thread Ben Finney
Vincent Bernat  writes:

> It was more to highlight that there is absolutely no consensus that
> interacting with proprietary services through the network is the same
> as combining work with a proprietary program.

Okay, and I think everyone here would agree those are not the same. So
I'm not sure why you'd highlight that.

Meanwhile, it's also true that the complaint is not merely “interacting
with proprietary services through the network”.

Instead, AIUI the complaint is packages which declare interaction with
non-free services *as their primary purpose*, to the point of
advertising that purpose in the package description.

-- 
 \   “Following fashion and the status quo is easy. Thinking about |
  `\your users' lives and creating something practical is much |
_o__)harder.” —Ryan Singer, 2008-07-09 |
Ben Finney



Re: Debian packages advertising non-free services

2017-08-07 Thread Scott Kitterman
On Tuesday, August 08, 2017 07:34:47 AM Vincent Bernat wrote:
>  ❦  8 août 2017 09:31 +1000, Ben Finney  :
> >> However, it is easy to find other packages interacting with
> >> proprietary services without a free implementation. For example, any
> >> package interacting with Google Cloud (golang-google-cloud package).
> > 
> > I am in agreement with Bas when he says: If other packages also violate
> > our principles or policy, that means they should be fixed, not that this
> > should be allowed.
> > 
> > So, “it is not hard to find other packages in Debian [with this
> > problem]” is not addressing the problem.
> 
> It was more to highlight that there is absolutely no consensus that
> interacting with proprietary services through the network is the same as
> combining work with a proprietary program.

I think the general rule for software with a FOSS license that has no other 
purpose than to interact with proprietary services is to put it in contrib.  
That's certainly what I've done.

I think there's a clear consensus that it's not at all the same (otherwise 
we'd put it in non-free).

Scott K

signature.asc
Description: This is a digitally signed message part.


Re: Debian packages advertising non-free services

2017-08-07 Thread Vincent Bernat
 ❦  8 août 2017 09:31 +1000, Ben Finney  :

>> However, it is easy to find other packages interacting with
>> proprietary services without a free implementation. For example, any
>> package interacting with Google Cloud (golang-google-cloud package).
>
> I am in agreement with Bas when he says: If other packages also violate
> our principles or policy, that means they should be fixed, not that this
> should be allowed.
>
> So, “it is not hard to find other packages in Debian [with this
> problem]” is not addressing the problem.

It was more to highlight that there is absolutely no consensus that
interacting with proprietary services through the network is the same as
combining work with a proprietary program.
-- 
In the Spring, I have counted 136 different kinds of weather inside of
24 hours.
-- Mark Twain, on New England weather


signature.asc
Description: PGP signature


Debian packages advertising non-free services (was: [pkg-go] Bug#856139: Bug#856139: certspotter: long description advertises commercial service)

2017-08-07 Thread Ben Finney
Vincent Bernat  writes:

>  ❦  7 août 2017 18:12 GMT, "Dr. Bas Wijnen"  :
>
> >> Example: [s3cmd]
> >
> > How is this not in contrib? This software is useless without the
> > non-free service (which is also software, and it is not in main)
> > from Amazon. Policy even mentions as an example for things in
> > contrib: wrapper packages or other sorts of free accessories for
> > non-free programs. That's exactly what this is.
>
> In this case, free S3 implementations exist (like Swift, available
> in Debian).

Do you think the free-software services are sufficiently brought to the
attention of the APT user, when considering the ‘s3cmd’ package? Or does
‘s3cmd’ rather give the impression that Amazon's non-free service is
needed?

> However, it is easy to find other packages interacting with
> proprietary services without a free implementation. For example, any
> package interacting with Google Cloud (golang-google-cloud package).

I am in agreement with Bas when he says: If other packages also violate
our principles or policy, that means they should be fixed, not that this
should be allowed.

So, “it is not hard to find other packages in Debian [with this
problem]” is not addressing the problem.

-- 
 \  “Yesterday I saw a subliminal advertising executive for just a |
  `\   second.” —Steven Wright |
_o__)  |
Ben Finney



Re: KickStarter for Debian packages - crowdfunding/donations for development

2013-06-17 Thread Gunnar Wolf
Philip Hands dijo [Sun, Jun 16, 2013 at 09:47:18AM +0100]:
> Manu Sporny  writes:
> ...
> That is an assumption that I happen to think is completely unfounded.
> 
> IBM tested various ways of incentivising coders decades ago -- almost
> (...)
> We tried DuncTank -- I'd contend that the net amount of productive work
> (...)
> It is bound to direct money to highly visible projects, regardless of
> (...)
> How do we determine a fair split between a couple of developers, one
> (...)
> I presume we'd be open about what people were being paid?  How about if
> (...)
> I'm not against people being paid for Free Software work -- that's what
> (...)
> If a developer and their customer negotiate a deal, nobody but the
> (...)
> In conclusion, I think this is a very dangerous idea, and that it would
> cause nothing but trouble.  The main underlying assumption is wrong.
> People work on Debian as amateurs, in the best sense of the word
> (i.e. motivated by the love of it, not for financial gain).  An influx
> of mercenaries would not be a net gain.
> 
> If it were needed or useful, Debian would not exist.
> 
> If it was a really good idea then we'd all be using something like
> Mandrake instead.

If I were more intelligent and better skilled in words, I wish I had
written Phil's mail. As things are, I can only subscribe to every word
he wrote.


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20130617234057.go54...@gwolf.org



Re: PaySwarm-based Debian donations (was: Re: KickStarter for Debian packages)

2013-06-17 Thread Charles Plessy
Le Sat, Jun 15, 2013 at 11:20:47PM -0400, Manu Sporny a écrit :
> 
> The files are composed together to suggest where donations should go to
> the sender. They are composed in this order:
> 
> 1. Upstream project's DONATE file.
> 2. Package maintainers DONATE file.
> 3. System's DONATE file.
> 
> So, when a benefactor types 'apt-donate apache2 $5', assuming 90% goes
> to ASF and 10% goes to Debian, they are provided with the following
> suggestion:
> 
> """
> Of your $5 donation:
>   1. $4.50 will go to the Apache Software Foundation
>   2. $0.50 will go to the Debian Project
> 
> If you would like to adjust the amounts, enter the number beside the
> amount that you'd like to adjust.
> 
> Do these amounts look good to you? (y/n)
> """

Dear Manu,

I like the idea of a DONATE file to facilitate donations to upstream projects.
At this point, I wonder what would be the role of apt.

 - If the goal is to donate for packages installed in the system, the DONATE
   files can be treated in a similar way to the FreeDesktop menu files:
   packages would install them in a given directory, and any donation system 
would
   parse them and detect additions and removals with Dpkg triggers.  One 
advantage
   is that software using the DONATE files to help the user to send money or
   bitcoins could be written independantly of the packaging system.

 - Apt would be useful if the goal is to gather the information in control
   files of the Debian archive (see DEP 11 for something a bit similar:
   http://wiki.debian.org/AppStreamDebianProposal).  But I think that this is 
not
   desirable, as it opens the risk of having conflicting settings when using
   third-party repositories.

Have a nice day,

-- 
Charles Plessy
Tsurumi, Kanagawa, Japan


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20130617120411.gg7...@falafel.plessy.net



Re: KickStarter for Debian packages - crowdfunding/donations for development

2013-06-16 Thread Manu Sporny
On 06/16/2013 04:47 AM, Philip Hands wrote:
> Manu Sporny  writes: ...
>> With respect to Debian-packaged software, if we address both 
>> issues, the benefit is that more resources can be directed toward 
>> Free Software development.
> 
> It is bound to direct money to highly visible projects, regardless of
> the effort required to package them, while people working on vital 
> but largely invisible infrastructure will get nothing much -- how 
> good is that going to be for the project?

Did you see the proposal to just send all donations to the Debian
project (or appropriate organization)? What do you think of that approach?

> I presume we'd be open about what people were being paid?  How about 
> if we end up publishing that we've given someone what amounts to a 
> fortune in their locale?

Donations would be between the benefactor and the entities receiving the
donation. It would be up to the people receiving the donation to be
transparent about it. This is the current state of affairs, btw. DDs
don't tell other people how much they're being paid to work on Debian.

> I'm not against people being paid for Free Software work -- that's 
> what pays my mortgage after all, and much of my income for the last 
> 20 years has been at least peripherally related to Debian.  I just 
> don't like the idea of Debian being the conduit for the money.

How is Debian being the conduit for the money in the proposal on the table?

> If a developer and their customer negotiate a deal, nobody but the 
> developer need worry if they think it's a fair deal, and nobody but 
> the developer's reputation is at risk.  Otherwise we'll start to see
>  complaints like: "I gave Debian $1000 and they don't even
> acknowledge my bug reports"

Why isn't this an issue w/ Debian donations today?

> If it were needed or useful, Debian would not exist.
> 
> If it was a really good idea then we'd all be using something like 
> Mandrake instead.

I don't follow your logic here, could you elaborate, please?

>> What do you think about the counter-argument to that statement 
>> posed by Martin Owens?
>> 
>> http://lists.debian.org/debian-project/2013/06/msg00031.html
> 
> The idea that it's currently impossible to fund Free Software is 
> nonsense.

Could you please point to where the argument presented makes that claim?

> Where payments to work on Debian make sense, removing friction is a 
> good thing for all involved, but that should all be done (far) 
> outside Debian.

What do you mean by this? Do you mean "If there was a program that made
it easier to donate money to software projects, I think that's fine and
I'd support such a program"?

-- manu

-- 
Manu Sporny (skype: msporny, twitter: manusporny, G+: ㆩ Sporny)
Founder/CEO - Digital Bazaar, Inc.
blog: Meritora - Web payments commercial launch
http://blog.meritora.com/launch/


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/51be6081.7050...@digitalbazaar.com



Re: KickStarter for Debian packages - crowdfunding/donations for development

2013-06-16 Thread Florian Weimer
* Manu Sporny:

> As an aside, PaySwarm is currency agnostic. The commercial
> implementation of it (Meritora) deals with USD today, has plans for Euro
> (and a few other national currencies) within a year, and Bitcoin shortly
> after that.

For the Euro, we already have the SEPA system, which is very hard to
beat in terms of usability.  I strongly suggest not to re-invent that
particular wheel.

(Few U.S. consumers have fully-fledged giro accounts, that's why
services like Paypal can get traction over there.)


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/87r4g2nlrd@mid.deneb.enyo.de



Re: KickStarter for Debian packages - crowdfunding/donations for development

2013-06-16 Thread Martin Owens
On Sun, 2013-06-16 at 13:25 +0100, Adam D. Barratt wrote:
> If you're seriously attempting to equate "I'll buy you a beer if you
> help me" with "corrupt bribery", then I suspect the net effect is
> going to be that people stop reading the rest of your argument.

It's not serious, it's absurdism. A rhetorical device used to point out
how daftly the opposing argument is being framed.

By 'people', you meant you right?

Martin,


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/1371386835.8589.4.camel@delen



Re: KickStarter for Debian packages - crowdfunding/donations for development

2013-06-16 Thread Adam D. Barratt
On Sun, 2013-06-16 at 07:59 -0400, Martin Owens wrote:
> On Sun, 2013-06-16 at 09:47 +0100, Philip Hands wrote:
> > The idea that it's currently impossible to fund Free Software is
> > nonsense. See IBM, HP, Canonical, my customers, anyone that's ever
> > said to a DD (or anyone else for that matter): "I'll buy you a beer if
> > you help me package this..."
> 
> The underlying assumption is wrong here. That only technical people,
> companies with technical people, or people that engage in corrupt
> bribery with alcohol are worthy of having their ideas and needs
> addressed in Free Software.

If you're seriously attempting to equate "I'll buy you a beer if you
help me" with "corrupt bribery", then I suspect the net effect is going
to be that people stop reading the rest of your argument.

Regards,

Adam


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/1371385552.4672.7.ca...@jacala.jungle.funky-badger.org



Re: KickStarter for Debian packages - crowdfunding/donations for development

2013-06-16 Thread Martin Owens
On Sun, 2013-06-16 at 09:47 +0100, Philip Hands wrote:
> The idea that it's currently impossible to fund Free Software is
> nonsense. See IBM, HP, Canonical, my customers, anyone that's ever
> said to a DD (or anyone else for that matter): "I'll buy you a beer if
> you help me package this..."

The underlying assumption is wrong here. That only technical people,
companies with technical people, or people that engage in corrupt
bribery with alcohol are worthy of having their ideas and needs
addressed in Free Software.

It's nice to think everything is hunky-dory, but all I see is a large
sea of users completely cut off from remedy and lots of developers
complaining they don't have enough time because they need to take jobs
either doing non-free software or not-software.

Now maybe packagers don't need to be involved in handling the money, it
might not fit structurally and well funded upstreams might be able to
fill in any holes anyway. But to deny upstreams the access to their
users because it's 'advertising' just makes me sad. Yes, Of course it's
advertising! Just like sticking their package in the repository and
indexing it is advertising. Debian already seeks to have a relationship
with upstreams and enfeebling upstream development is not a good way to
have that relationship play out.

I'm happy to see more research on the psychological effects of money in
software. But I will say that if there are problems and it's impossible
to involve economics, they are problems with human psychology and
Debian's social structure, not with money itself.

Thanks for the debate.

Martin Owens


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/1371383940.13856.25.camel@delen



Re: PaySwarm-based Debian donations (was: Re: KickStarter for Debian packages)

2013-06-16 Thread Stefano Zacchiroli
On Sun, Jun 16, 2013 at 10:05:48AM +0100, Lars Wirzenius wrote:
> You suggest that package maintainers get to suggest where donations go.
> There's two glaring problems there. First, it disregards all the great
> things people do to make Debian better that are _not_ about packaging
> at all.

Yeah, I agree with this concern as well. Everything which tries to
pay/tip Debian contributors and stays at the package level only is
doomed to fail, and bring with it some of the nasty consequences that
have been highlighted.

OTOH, I think it would be fine to have something at the package level to
pass on donations to our upstreams, as well as to ease donating to the
Debian project as a whole. See [1,2], already mentioned by Paul Wise in
his initial followup to this thread.

[1]: https://lists.debian.org/debian-www/2013/05/msg00025.html
[2]: http://upstream-metadata.debian.net/table/Donation

All the problems that have been highlighted are related to pay/tip
Debian contributors and the distorsions that institutionalizing that
process might introduce. But we already have at least 2 other kinds of
entities that actively seek donations (upstreams and Debian as a
project), and we do tolerate that. Making it easier for them to seek
donations doesn't seem problematic to me. On the contrary, there is an
active debate on how to sustain Free Software development while at the
same time keeping it away of companies and the need of turning a profit
(see for instance [3]).

[3]: https://lwn.net/Articles/511260/

Using Debian as a conduit to *ease* donation flows to FOSS that already
exist seems an useful thing to do to me. That is, after all, exactly
what [2] was meant for, maybe it can be improved and become more
successful adopting some of Manu's ideas.

Cheers.
-- 
Stefano Zacchiroli  . . . . . . .  z...@upsilon.cc . . . . o . . . o . o
Maître de conférences . . . . . http://upsilon.cc/zack . . . o . . . o o
Former Debian Project Leader  . . @zack on identi.ca . . o o o . . . o .
« the first rule of tautology club is the first rule of tautology club »


signature.asc
Description: Digital signature


Re: PaySwarm-based Debian donations (was: Re: KickStarter for Debian packages)

2013-06-16 Thread Lars Wirzenius
On Sat, Jun 15, 2013 at 11:20:47PM -0400, Manu Sporny wrote:
> Thanks to everyone that has participated in the discussion thus far. :)
> I think there have been a number of solid concerns and issues raised,
> which I'm going to try and wrap into a proposal below.
> 
> I think it might help simplify the donations goal by framing it in the
> following way:
> 
> Ultimately, where to send a donation is the decision of the person or
> organization doing the donation (the benefactor).
> 
> Package maintainers, software developers, and project organizations can
> lobby for where they'd like to see the money go, but it's the benefactor
> that decides where they'd like to send the money in the end. Given that
> premise, all a package maintainer, software developer, or organization
> can do is make suggestions to the benefactor.

I am afraid I find this whole approach not just questionable, but
likely to distort, and damage, the free software development processes
in general, and Debian development in particular. I suggest we, the
Debian project, approach this very carefully.

While the reality is slightly more complex, we are currently in a
state where we make technical decisions mostly based on what is the
right thing to do. We sometimes disagree on what the right thing is,
but the disagreements are based on our interpretations of shared goals
and values, and different evaluations of the various solutions, and
different emphasis on various technical virtues.

The more we introduce money into the development process, the higher
the risk is that we get away from making decisions based on what the
technically right thing is for us and for our users, and the more we
will decide things based on how we can maximise our income.

Be careful what you reward, because you will get more of it. Even if
you don't actually want more of it.

You suggest that package maintainers get to suggest where donations go.
There's two glaring problems there. First, it disregards all the great
things people do to make Debian better that are _not_ about packaging
at all.  We have translators, documentation writers, wiki gardeners, bug
triagers, people who answer questions on the debian-user mailing lists
in various languages, people who help staff Debian booths at various
conferences, people who run user groups. And so on, and so forth. None
of this work is highly visible, and it's really hard to target them with
donations, yet it's often more work than maintaining packages.

Second, even considering package maintainers only, targeted donations
would unfairly favour those maintaining the most visible packages.
If maintaining, say, Iceweasel or GNOME or Emacs results in getting
money, that will certainly lessen the interest in maintaining, say, 
Make, coreutils, or Grub. If having your name on four hundred packages
gives you ten times more money than maintaining one package well,
what happens to average package quality?

These are not unsolveable problems, I'm sure. However, I don't think
your attitude to solving them will result in a good solution, and you
may wreck things on the way. You push for a particular payment solution,
and dismiss the experiences and concerns of your critics. I fear this
is not a recipe for success.

(Disclaimer: I used to have a consulting contract to "improve the
technical quality of Debian", which gave me a livelihood for about
1.5 years. The lasting result of that was piuparts.)

-- 
http://www.cafepress.com/trunktees -- geeky funny T-shirts
http://gtdfh.branchable.com/ -- GTD for hackers


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20130616090548.GD31164@havelock



Re: KickStarter for Debian packages - crowdfunding/donations for development

2013-06-16 Thread Philip Hands
Manu Sporny  writes:
...
> With respect to Debian-packaged software, if we address both issues,
> the benefit is that more resources can be directed toward Free
> Software development.

That is an assumption that I happen to think is completely unfounded.

IBM tested various ways of incentivising coders decades ago -- almost
all of them were disastrously counter-productive.

We tried DuncTank -- I'd contend that the net amount of productive work
done was reduced by that initiative, and some very active contributors
were demotivated to the point that they went away and didn't come back.

It is bound to direct money to highly visible projects, regardless of
the effort required to package them, while people working on vital but
largely invisible infrastructure will get nothing much -- how good is
that going to be for the project? (when the Morlocks see the Eloi having
all the fun, I fear that they may start to get hungry ;-) ).

How do we determine a fair split between a couple of developers, one
living in a penthouse in New York, and another living in a shanty town
on a dollar a day.

I presume we'd be open about what people were being paid?  How about if
we end up publishing that we've given someone what amounts to a fortune
in their locale?

I'm not against people being paid for Free Software work -- that's what
pays my mortgage after all, and much of my income for the last 20 years
has been at least peripherally related to Debian.  I just don't like the
idea of Debian being the conduit for the money.  I think it's even
problematic for Debian to act as the advertiser.

If a developer and their customer negotiate a deal, nobody but the
developer need worry if they think it's a fair deal, and nobody but the
developer's reputation is at risk.  Otherwise we'll start to see
complaints like: "I gave Debian $1000 and they don't even acknowledge my
bug reports"

In conclusion, I think this is a very dangerous idea, and that it would
cause nothing but trouble.  The main underlying assumption is wrong.
People work on Debian as amateurs, in the best sense of the word
(i.e. motivated by the love of it, not for financial gain).  An influx
of mercenaries would not be a net gain.

If it were needed or useful, Debian would not exist.

If it was a really good idea then we'd all be using something like
Mandrake instead.

Cheers, Phil.

P.S. in answer to:
> What do you think about the counter-argument to that statement posed by 
> Martin Owens?
>
> http://lists.debian.org/debian-project/2013/06/msg00031.html

The idea that it's currently impossible to fund Free Software is
nonsense. See IBM, HP, Canonical, my customers, anyone that's ever said
to a DD (or anyone else for that matter): "I'll buy you a beer if you
help me package this..."

Where payments to work on Debian make sense, removing friction is a good
thing for all involved, but that should all be done (far) outside Debian.
-- 
|)|  Philip Hands [+44 (0)20 8530 9560]http://www.hands.com/
|-|  HANDS.COM Ltd.http://www.uk.debian.org/
|(|  10 Onslow Gardens, South Woodford, London  E18 1NE  ENGLAND


pgpccuxZ0gsvc.pgp
Description: PGP signature


Re: PaySwarm-based Debian donations (was: Re: KickStarter for Debian packages)

2013-06-16 Thread Holger Levsen
On Sonntag, 16. Juni 2013, Manu Sporny wrote:
> Thanks to everyone that has participated in the discussion thus far. :)
> I think there have been a number of solid concerns and issues raised,
> which I'm going to try and wrap into a proposal below.

and then you continue to ignore these concerns and continue with your agenda 
:-(


please go away and "improve" something else. 


signature.asc
Description: This is a digitally signed message part.


Re: KickStarter for Debian packages - crowdfunding/donations for development

2013-06-16 Thread Andrei POPESCU
On Sb, 15 iun 13, 21:38:24, Manu Sporny wrote:
> 
> Yes, we probably don't want to create a Mos Eisley in Debian. However,
> knowing that you can go somewhere to hire Debian developers to fix
> issues that you're having with the system would be very helpful for
> companies (like ours and the ones we sub-contract from) that build
> products on top of Debian.

http://www.debian.org/consultants/

Kind regards,
Andrei
-- 
http://wiki.debian.org/FAQsFromDebianUser
Offtopic discussions among Debian users and developers:
http://lists.alioth.debian.org/mailman/listinfo/d-community-offtopic


signature.asc
Description: Digital signature


PaySwarm-based Debian donations (was: Re: KickStarter for Debian packages)

2013-06-15 Thread Manu Sporny
Thanks to everyone that has participated in the discussion thus far. :)
I think there have been a number of solid concerns and issues raised,
which I'm going to try and wrap into a proposal below.

I think it might help simplify the donations goal by framing it in the
following way:

Ultimately, where to send a donation is the decision of the person or
organization doing the donation (the benefactor).

Package maintainers, software developers, and project organizations can
lobby for where they'd like to see the money go, but it's the benefactor
that decides where they'd like to send the money in the end. Given that
premise, all a package maintainer, software developer, or organization
can do is make suggestions to the benefactor.

Let's assume that there is a file named DONATE that is included with
source distributions of Free Software. Its use is analogous to a README.
A DONATE file lists a PaySwarm-compatible list of financial accounts
(and percentages associated with each account) that should be the target
of donations to the particular software project.

This file can also be included in the debian/ subdirectory. When placed
in that directory, it lists donations that the package maintainers feel
should be performed on top of the upstream authors' DONATE file (if it
exists).

Finally, this file can also be included in the /etc/ directory (or maybe
/etc/donate.conf?) to specify a list of donations that the distribution
feels should be performed on top of all other donations.

The files are composed together to suggest where donations should go to
the sender. They are composed in this order:

1. Upstream project's DONATE file.
2. Package maintainers DONATE file.
3. System's DONATE file.

So, when a benefactor types 'apt-donate apache2 $5', assuming 90% goes
to ASF and 10% goes to Debian, they are provided with the following
suggestion:

"""
Of your $5 donation:
  1. $4.50 will go to the Apache Software Foundation
  2. $0.50 will go to the Debian Project

If you would like to adjust the amounts, enter the number beside the
amount that you'd like to adjust.

Do these amounts look good to you? (y/n)
"""

At this point, the benefactor can modify amounts or choose not to
perform the donation at all.

Things that are currently not provided in the proposal:

1. The format of the DONATE file.
2. How these DONATE files are installed onto the system, or
   discovered by apt-donate.
3. Whether or not these files should specify "for-hire" or
   "bounty board" URLs.

This proposal attempts to address a number of concerns raised related to
package maintainers having a variety of differing opinions on how we
should handle donations. After reading through the proposal above, are
there any issues or concerns that remain as a high to moderate risk?

-- manu

-- 
Manu Sporny (skype: msporny, twitter: manusporny, G+: +Manu Sporny)
Founder/CEO - Digital Bazaar, Inc.
blog: Meritora - Web payments commercial launch
http://blog.meritora.com/launch/


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/51bd2f0f.1070...@digitalbazaar.com



Re: KickStarter for Debian packages - crowdfunding/donations for development

2013-06-15 Thread Manu Sporny
On 06/15/2013 02:18 AM, Charles Plessy wrote:
> Le Sat, Jun 15, 2013 at 12:25:26AM -0400, Joey Hess a écrit :
>> Charles Plessy wrote:
>>> In the case of Debian, I share with others the concern of having 
>>> the packages as a source of revenue
>> 
>> How about making fixed bugs a source of revenue?
> 
> I do not see how to fit this in the PaySwarm model proposed here, 
> unless there are URLs for billing?

There is a PaySwarm spec called "Payment Intents", which is basically a
way to express "If X happens, I want to pay $Y". This can be used to do
all sorts of limit-based financial transactions:

"If bug 23489 is fixed, I'll pay $5."
"If $500 is raised to address bug 32892, I'll put in $10."

> In that case, the bounties could accumulate until somebody takes them
> by fixing the bug.

Have you heard of? https://www.bountysource.com/

Similar model.

> We would probably need some review system unless there are easy way
> for refunds.

The way PaySwarm Payment Intents work is by collecting "promises to
pay". When a limit is hit, all the payments go through. The model is
similar to KickStarter's all-or-nothing model. However, that model is a
bit constrained by the credit card and banking system (holds are put on
the funds in certain instances), which PaySwarm is not constrained by.

-- manu

-- 
Manu Sporny (skype: msporny, twitter: manusporny, G+: +Manu Sporny)
Founder/CEO - Digital Bazaar, Inc.
blog: Meritora - Web payments commercial launch
http://blog.meritora.com/launch/


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/51bd24d6.10...@digitalbazaar.com



Re: KickStarter for Debian packages - crowdfunding/donations for development

2013-06-15 Thread Manu Sporny
On 06/14/2013 09:45 PM, Charles Plessy wrote:
> thanks for your work on web payments.  I hope that they will be a 
> lead contribution to the development of Free works.

And thanks to you for taking part in the conversation and helping us
build something that will hopefully lead to more resources funneled to
Free Software development. :)

> I think that for Debian, we need methods that pay for the work to be 
> done, rather than for the work that has been done.  Debian has a 
> large number of contributors, and we should encourage mobility 
> between projects, rather than make ourselves dependant to the
> project that would be currently paying for our bills.

I see, good point. So your opinion is that something like a
bounty/crowdfunding mechanism would be more helpful than a donation system?

> Moreover, some of our upstreams who are already collecting donations
> may be worried of losing opportunities to reveive when their work is
> distributed through Debian packages. Here, PaySwarm can solve our
> problem.

I think I understand.

> As Paul showed you, we are collecting URLs for donations, and we 
> probably miss a lot of them.  If the PaySwarm project would provide 
> with mechanisms to easily communicate or discover this information, 
> that would be very useful.  Then, together with the information that 
> the donation they received comes from a Debian user, we could pass 
> the URL to our upstream guide
> (http://wiki.debian.org/UpstreamGuide), so that the packaging work
> (and therefore the need for support) is reduced thanks to Upstream
> standardising his work.

Got it, I'm working on a proposal now that might work. Your input was
very helpful as it helps flesh out some of the bounty/crowdfunding stuff.

-- manu

-- 
Manu Sporny (skype: msporny, twitter: manusporny, G+: +Manu Sporny)
Founder/CEO - Digital Bazaar, Inc.
blog: Meritora - Web payments commercial launch
http://blog.meritora.com/launch/


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/51bd2330.3060...@digitalbazaar.com



Re: KickStarter for Debian packages - crowdfunding/donations for development

2013-06-15 Thread Manu Sporny
On 06/14/2013 07:29 PM, David Kalnischkies wrote:
>> I disagree. Some packages make _a lot_ of work and some people 
>> spend thousands of hours to make Debian an excellent distribution 
>> and the package in particular useful and maintainable. This is in 
>> many cases not less work than being upstream, it's just another 
>> type of work which is often invisible.
> 
> My train of thought is that I will never be able to accept money for
>  the package 'apt'. Sure, I might do some stuff, but this package is
>  17 years old, has seen contribution from all round the world and all
>  that. What would give me the right to claim it rather than anyone 
> else who has ever worked on it. And even if you haven't worked 
> directly on it, we still have dependencies, be it code wise or 
> infrastructure (hardware, people, …). It can't be my job as the 
> maintainer¹ to rate the contribution of everyone even only remotely 
> involved to decide who is getting, who isn't and how much exactly. (¹
> which I am not, but lets keep it simple here)

Why wouldn't you, as maintainer, just say that all donations go to the
Debian Project? Isn't that the most logical thing to do in this instance?

> I have on the other hand accepted money for working on APT (e.g. as 
> GSoC student) so the concept of APT being a magical cash-cow I can 
> milk isn't completely alien for me, but for me there is a difference
>  between claiming that I am the one who should get the money if you 
> like the package and being up for hire if you need something to like
>  it (even more)

So is it fair to say that you'd like to be able to do the following:

1. List the Debian Project as the receiver for all donations.
2. List yourself as someone that is willing to be hired to do work on APT?

-- manu

-- 
Manu Sporny (skype: msporny, twitter: manusporny, G+: +Manu Sporny)
Founder/CEO - Digital Bazaar, Inc.
blog: Meritora - Web payments commercial launch
http://blog.meritora.com/launch/


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/51bd20c4.9000...@digitalbazaar.com



Re: KickStarter for Debian packages - crowdfunding/donations for development

2013-06-15 Thread Manu Sporny
On 06/14/2013 06:50 PM, Jonas Smedegaard wrote:
> Anyone can locate a particular Debian contributor and wire them 15 
> Bitcoin.  No need for Debian to support that.

I don't think you mean 'anyone'. I think you mean 'a highly skilled
programmer that understands how open source software is built and how
the Debian project operates...' :)

In the same vein, you could claim that many people on the planet can
find someone else on the planet and get money to them. There are two
issues with attempting to do that, however: the first is finding the
right person, and the second is figuring out how to get something of
value to them. :)

With respect to Debian-packaged software, if we address both issues, the
benefit is that more resources can be directed toward Free Software
development.

Even finding the correct Debian Developer to send money to is a very
difficult proposition for someone that is just using a piece of
software. It's very difficult to find this information. Even if you find
the information, you have to figure out if the target Debian developer
accepts Bitcoin. Many don't, and would prefer to be paid in their
national currency.

Merely stating that something is possible does not mean that it's easy
to do. :)

> Debian do not promote a capitalist-driven boost, nor any other 
> specific gaming of the system.  Debian is indifferent about how its 
> developers find time and devotion for our shared project.
> 
> I really thought Gunnar nailed it:
> 
>> We have never needed it

What do you think about the counter-argument to that statement posed by
Martin Owens?

http://lists.debian.org/debian-project/2013/06/msg00031.html

-- manu

-- 
Manu Sporny (skype: msporny, twitter: manusporny, G+: +Manu Sporny)
Founder/CEO - Digital Bazaar, Inc.
blog: Meritora - Web payments commercial launch
http://blog.meritora.com/launch/


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/51bd1f32.7000...@digitalbazaar.com



Re: KickStarter for Debian packages - crowdfunding/donations for development

2013-06-15 Thread Manu Sporny
On 06/14/2013 05:24 PM, Paul Tagliamonte wrote:
> On Fri, Jun 14, 2013 at 05:14:27PM -0400, Manu Sporny wrote:
>> I agree, which is why the payment details live completely outside
>> the Debian systems. The only thing you'd need to initiate payment
>> is an e-mail address, or a PaySwarm financial account address,
>> which looks like this:
>> 
>> https://dev.payswarm.com/i/manu/accounts/public-account
> 
> Interesting. So one would have to set up a payswarm instance to
> collect, or piggy back on a public instance?

You would piggy back on a public instance. Basically, you'd pick a
payment provider (kinda like you pick your grocery store, school, or
bank today). You would pick a payment processor you trust, and since
they're using PaySwarm, they're guaranteed to inter-operate with every
other PaySwarm-compatible payment processor out there.

We're also considering supporting financial account re-directs from a
personal website (as long as it's served over HTTPS). You could setup a
re-direct from your personal website to a URL on your payment provider.
There's isn't much advantage in doing this other than you being in full
control of your financial accounts in the event that your payment
provider goes rogue and/or freezes your account.

>> requirements when you process payments; know your customer (KYC) is
>> one of them. There are also anti-money laundering regulations that
>> you must ensure you follow to comply with the law. All of these
>> things are things
> 
> Which law? US law? We're an international org, are the laws standard 
> worldwide? We have Developers in every jurisdiction you can imagine
> :)

If you want to get very technical about it, the "law" is the combination
of all local, state, and federal laws applicable to the financial
transaction. What you're exchanging, how you're exchanging it, who is
exchanging it, trade embargoes... all of these things apply when you're
doing a financial transaction. So yes, it's a complicated problem due to
all the laws that apply, which is why payment processors exist.

>> Right, which is why one solution is making it up to the package 
>> maintainers and software authors to figure out how payments should
>> be split up. I think we all agree that tips should be distributed
>> based on merit and that the maintainers have a pretty good idea of
>> how that should go. In the event that the maintainers and upstream
>> can't come to an agreement, they could always just opt to send the
>> donation upstream to the Debian Project.
> 
> Right, but this leads to one of two things:

Wouter Verhelst had an excellent response to your concerns. There was a
counter to his line of argumentation, which I'll respond to shortly.

>>> Payment systems in general tend to lead to "paybullying", which
>>> is something I'd really (really) like to avoid. I've always loved
>>> how un-corporate the Debian community is.
>> 
>> I agree that we really don't want that. Any suggestions about how
>> we might be able to avoid it?
> 
> I'm not sure, but I've seen at least one high-profile F/OSS project 
> maintainer (with project email, writing from it) saying "I've written
> a patch for this bug, it's done, but you need to give me money before
> I release it". Putting an official system into place might make this
> more common / easier to make look "official".

I don't quite understand. I thought your concern was preventing that
sort of behavior, not condoning it?

That is, this is my read from the community so far:

People holding patches for ransom seems to be bad form and
anti-community. People stating that they'd like to work on X, but would
like to be paid for the work is better. People doing the work for free
and retroactively getting a donation for it is better. All donations to
large teams going to the Debian project seems to be the least
controversial solution (ignoring the fact that upstream developers don't
see any of that money).

I'm formulating a proposal that attempts to address all of the concerns
raised so far in this thread. More later...

-- manu

-- 
Manu Sporny (skype: msporny, twitter: manusporny, G+: +Manu Sporny)
Founder/CEO - Digital Bazaar, Inc.
blog: Meritora - Web payments commercial launch
http://blog.meritora.com/launch/


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/51bd1ba0.1080...@digitalbazaar.com



Re: KickStarter for Debian packages - crowdfunding/donations for development

2013-06-15 Thread Manu Sporny
On 06/14/2013 02:54 PM, Scott Howard wrote:
> As a member of the bitcoin team in debian, I also see how easy it 
> could be in the future to semi-anonymously receive payment from and 
> to anywhere in the world for such work, and that this may already be 
> going on already in Debian or other distros.

As an aside, PaySwarm is currency agnostic. The commercial
implementation of it (Meritora) deals with USD today, has plans for Euro
(and a few other national currencies) within a year, and Bitcoin shortly
after that.

> I don't know how I feel about this, fundamentally, yet. But my gut 
> didn't like the idea of having a wiki page where DDs can post their 
> hourly rates for packaging software for Debian...

Yes, we probably don't want to create a Mos Eisley in Debian. However,
knowing that you can go somewhere to hire Debian developers to fix
issues that you're having with the system would be very helpful for
companies (like ours and the ones we sub-contract from) that build
products on top of Debian.

There is already a fantastic amount of money that is involved in the
development of Debian, it's just hidden.

-- manu

-- 
Manu Sporny (skype: msporny, twitter: manusporny, G+: +Manu Sporny)
Founder/CEO - Digital Bazaar, Inc.
blog: Meritora - Web payments commercial launch
http://blog.meritora.com/launch/


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/51bd1710.5020...@digitalbazaar.com



Re: KickStarter for Debian packages - crowdfunding/donations for development

2013-06-15 Thread Paul Tagliamonte
On Sat, Jun 15, 2013 at 11:48:27AM +0200, Wouter Verhelst wrote:
> On 14-06-13 23:24, Paul Tagliamonte wrote:
> > Right, but this leads to one of two things:
> > 
> >   - No money is shared with dependencies (leading to people flocking to
> > awesomewm, gnome, kde, chrome, wine, apache2, etc)
> > 
> >   - Money is shared with dependencies (leading to people flocking to
> > gcc, linux, libc6)
> 
> Which is a problem, why?

Because, given a naive implementation, people could latch on and feed
off all the money trickling in without doing work. Unless you want to
"rate" people in Debian, which I think is a very bad idea.

Oh you? You're worth 1%. You? Oh, you get nothing. Kbai.

> - The maintainer(s) decide to put all the money in a fund that is used
>   for things like meetings among the package's maintainers. In other
>   words, there is no direct financial benefit to be had, and thus I
>   don't expect people to be interested in joining purely for financial
>   benefit.

This would be a nice result of this system. Very very nice. I would love
if donations for a package went to a single discretionary fund to help
them with development. Things like:

  - going to upstream development sprints
  - meeting in a central location for a sprint
  - [other brilliant ideas here]

> - The maintainer(s) set up some complex scheme by which financial
>   benefit is equally distributed among contributors based on size of

I think this is a can of worms we should very much avoid.

[..]

> Unless you think money is dirty, I don't see how any of this would
> involve "flocking" in a problematic manner.

I don't think it's dirty, but it distorts views. When the person next to
you doing less work than you is making $MONEY, and you're not, you don't
want to work on $THING anymore.

> Am I missing something?

No, just not looking in the same places as me.

> If/when this were to happen in Debian, I think it would be fair to kick
> said developer out of the project, on the basis of them violating the
> "do not stand in the way" rule of constitution §2.1.1.

He claimed he'd not stand in the way of anyone writing their own patch,
and I don't think you could claim not writing the patch is standing in
the way, so I think this doesn't apply.

It's shady, but I don't think we have a mechinism for enforcing people
don't do this. Having a big-professional system where you say "Oh, just
apt-donate me $10 usd and I'll release this patch", people would believe
this is how things work.


Cheers,
   Paul


-- 
 .''`.  Paul Tagliamonte 
: :'  : Proud Debian Developer
`. `'`  4096R / 8F04 9AD8 2C92 066C 7352  D28A 7B58 5B30 807C 2A87
 `- http://people.debian.org/~paultag


signature.asc
Description: Digital signature


Re: KickStarter for Debian packages - crowdfunding/donations for development

2013-06-15 Thread Jakub Wilk

* Wouter Verhelst , 2013-06-15, 11:48:
What happens with the money should be decided by the maintainers of the 
package. Before you'll see "flocking", there will have been such a 
decision already (otherwise there's no money and thus no "flocking"). 
Given that, I can see only a few possible outcomes:

[...]
- The (main/sole) maintainer decides to egoistically take all the money 
for themselves. They alienate all other maintainers, and nobody ever 
wants to work with them anymore.


Yeah, that could happen with or without money involved.

The package suffers, until it is hijacked or taken away by decree of 
the TC or similar.


Or rather s/, until.*/ forever/. I'm talking from experience here.

--
Jakub Wilk


--
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20130615115151.gb4...@jwilk.net



Re: KickStarter for Debian packages - crowdfunding/donations for development

2013-06-15 Thread Wouter Verhelst
On 14-06-13 23:24, Paul Tagliamonte wrote:
> Right, but this leads to one of two things:
> 
>   - No money is shared with dependencies (leading to people flocking to
> awesomewm, gnome, kde, chrome, wine, apache2, etc)
> 
>   - Money is shared with dependencies (leading to people flocking to
> gcc, linux, libc6)

Which is a problem, why?

What happens with the money should be decided by the maintainers of the
package. Before you'll see "flocking", there will have been such a
decision already (otherwise there's no money and thus no "flocking").
Given that, I can see only a few possible outcomes:

- The maintainer(s) decide to put all the money in a fund that is used
  for things like meetings among the package's maintainers. In other
  words, there is no direct financial benefit to be had, and thus I
  don't expect people to be interested in joining purely for financial
  benefit.
- The maintainer(s) set up some complex scheme by which financial
  benefit is equally distributed among contributors based on size of
  contribution. People may flock to that package, but the net result
  will be a better package. This is a good thing. If many people try to
  join the packaging team, eventually all the low-hanging fruit will be
  gone and people will start looking for other packaging teams to join,
  because the possible benefit no longer outweighs the investment to be
  made.
- The (main/sole) maintainer decides to egoistically take all the money
  for themselves. They alienate all other maintainers, and nobody ever
  wants to work with them anymore. The package suffers, until it is
  hijacked or taken away by decree of the TC or similar. This is a
  possible problem in the proposed scheme, but I don't think it could
  involve any "flocking", since there's no benefit to be had.
- The (main/sole) maintainer decides to quit their job and live off
  donated money instead. Debian gets a more focused developer (and,
  thus, better packages) as a result. This is essentially similar to the
  second scheme, except that other people will have a harder time
  keeping up with the full-time developer.

Unless you think money is dirty, I don't see how any of this would
involve "flocking" in a problematic manner.

Am I missing something?

[...]
> I'm not sure, but I've seen at least one high-profile F/OSS project
> maintainer (with project email, writing from it) saying "I've written a
> patch for this bug, it's done, but you need to give me money before I
> release it". Putting an official system into place might make this more
> common / easier to make look "official".

If/when this were to happen in Debian, I think it would be fair to kick
said developer out of the project, on the basis of them violating the
"do not stand in the way" rule of constitution §2.1.1.

-- 
This end should point toward the ground if you want to go to space.

If it starts pointing toward space you are having a bad problem and you
will not go to space today.

  -- http://xkcd.com/1133/



signature.asc
Description: OpenPGP digital signature


Re: KickStarter for Debian packages - crowdfunding/donations for development

2013-06-15 Thread Wouter Verhelst
On 15-06-13 06:07, Russ Allbery wrote:
> I think the tension gets much worse if
> the project is explicitly deciding to pay some people and not others.

While that is true, I do not believe that the proposed scheme results in
"the project (...) deciding to pay". If done right, this could simply be
a way for people to communicate that a) they welcome donations, and b)
where these donations should go to.

Every maintainer in Debian can, for themselves, decide whether that
metadata should signal that they accept donations, and if so where these
donations should go. The project as a whole would not seem to factor in
to that decision.

One of the major arguments against the scheme of trying to arbitrarily
pay project members with project funds was that it was fairly difficult
for project members who just happen to be active in other areas to
solicit payment. This is explicitly not the case with the proposed scheme.

-- 
This end should point toward the ground if you want to go to space.

If it starts pointing toward space you are having a bad problem and you
will not go to space today.

  -- http://xkcd.com/1133/


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/51bc3434.3030...@debian.org



Re: KickStarter for Debian packages - crowdfunding/donations for development

2013-06-15 Thread Holger Levsen
On Samstag, 15. Juni 2013, Nikolaus Rath wrote:
> I agree. It would add a whole new dimension to NMU'ing, orphaning,
> adopting, and salvaging packages with a large user base. For example,
> currently most people would probably happily accept co-maintainers even
> if they're confident that they could do the work alone -- but what if
> adding a co-maintainer means that a potential stream of personal revenue
> will be cut in half?

thanks for nailing this one aspect. (there are more but this one is a good 
example...)




signature.asc
Description: This is a digitally signed message part.


Re: KickStarter for Debian packages - crowdfunding/donations for development

2013-06-15 Thread Holger Levsen
Hi,

On Samstag, 15. Juni 2013, Russ Allbery wrote:
> There were some past experiments with this in Debian, and they caused a
> lot of social controversy.
> 
> One of the problems with paying for work in the Debian context is that
> we're a world-wide project that welcomes contributions from everyone as
> equally as we can manage.  I think this is one of the major strengths of
> the project.  For the most part, we can ignore such things as differences
> in compensation rates in different parts of the world.  But if we get into
> paying for work, that immediately highlights that amounts that are
> inadequate to pay for skilled time in some areas where there are project
> contributors are far more than a typical wage in other areas.  This
> creates, or at least highlights, an awkward inequality in the project.
> 
> Another problem is that when some people are paid for doing the same work
> that other people are doing on a volunteer basis, it creates a lot of
> tension that's difficult to manage.  While this is true anyway (for
> example, my employer pays me to do some packaging work), currently it's
> quite indirect, and the payment isn't officially blessed by the project or
> part of the project structure.  Every contributor to Debian just manages
> their finances in their own ways.  I think the tension gets much worse if
> the project is explicitly deciding to pay some people and not others.

FWIW I share these concerns and also don't like the idea to as a project 
endorse ways to give money to developers. I don't want to see money persons 
join Debian and I dont want see people leave Debian because of it. I also dont 
want to see people fight over money. ("Hey, I must maintain this software 
because I deserve the money users pay for it.")

Donations to Debian OTOH, which benefit all developers (by paying for 
hardware, developer sprints, etc) are very much welcome.


cheers,
Holger


signature.asc
Description: This is a digitally signed message part.


Re: KickStarter for Debian packages - crowdfunding/donations for development

2013-06-14 Thread Charles Plessy
Le Sat, Jun 15, 2013 at 12:25:26AM -0400, Joey Hess a écrit :
> Charles Plessy wrote:
> > In the case of Debian, I share with others the concern of having the 
> > packages
> > as a source of revenue
> 
> How about making fixed bugs a source of revenue?

I do not see how to fit this in the PaySwarm model proposed here, unless there
are URLs for billing ?  In that case, the bounties could accumulate until
somebody takes them by fixing the bug.  We would probably need some review
system unless there are easy way for refunds.  For instance, one person would
claim the bounties by sending a patch, and a Debian Developer or Maintainer
would acknowledge that the patch actually solves the problem by uploading the
package.  But there are probably better ways to do it.

Cheers,

-- 
Charles Plessy
Tsurumi, Kanagawa, Japan


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20130615061824.ge11...@falafel.plessy.net



Re: KickStarter for Debian packages - crowdfunding/donations for development

2013-06-14 Thread Ben Hutchings
On Sat, 2013-06-15 at 00:25 -0400, Joey Hess wrote:
> Charles Plessy wrote:
> > In the case of Debian, I share with others the concern of having the 
> > packages
> > as a source of revenue
> 
> How about making fixed bugs a source of revenue?

http://dilbert.com/strips/comic/1995-11-13/

Ben.

-- 
Ben Hutchings
If the facts do not conform to your theory, they must be disposed of.


signature.asc
Description: This is a digitally signed message part


Re: KickStarter for Debian packages - crowdfunding/donations for development

2013-06-14 Thread Joey Hess
Charles Plessy wrote:
> In the case of Debian, I share with others the concern of having the packages
> as a source of revenue

How about making fixed bugs a source of revenue?

-- 
see shy jo


signature.asc
Description: Digital signature


Re: KickStarter for Debian packages - crowdfunding/donations for development

2013-06-14 Thread Russ Allbery
Martin Owens  writes:

> I feel much better about paying for work to be done than tipping, it
> feels more right. Although it requires larger pools of money and a more
> forthright approach deciding where this resource should be plowed in
> order to curate the larger design. Debian hasn't been much for curating
> anything, since it's more anarchistic than planned. But maybe it'd be
> possible to have project goals at the debian level which such money
> could be focused.

There were some past experiments with this in Debian, and they caused a
lot of social controversy.

One of the problems with paying for work in the Debian context is that
we're a world-wide project that welcomes contributions from everyone as
equally as we can manage.  I think this is one of the major strengths of
the project.  For the most part, we can ignore such things as differences
in compensation rates in different parts of the world.  But if we get into
paying for work, that immediately highlights that amounts that are
inadequate to pay for skilled time in some areas where there are project
contributors are far more than a typical wage in other areas.  This
creates, or at least highlights, an awkward inequality in the project.

Another problem is that when some people are paid for doing the same work
that other people are doing on a volunteer basis, it creates a lot of
tension that's difficult to manage.  While this is true anyway (for
example, my employer pays me to do some packaging work), currently it's
quite indirect, and the payment isn't officially blessed by the project or
part of the project structure.  Every contributor to Debian just manages
their finances in their own ways.  I think the tension gets much worse if
the project is explicitly deciding to pay some people and not others.

-- 
Russ Allbery (r...@debian.org)   


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/87bo78koez@windlord.stanford.edu



Re: KickStarter for Debian packages - crowdfunding/donations for development

2013-06-14 Thread Martin Owens
On Fri, 2013-06-14 at 07:20 -0400, Paul Tagliamonte wrote:
> Payment systems in general tend to lead to "paybullying", which is
> something I'd really (really) like to avoid. I've always loved how
> un-corporate the Debian community is.

There are two reasons money gets given to people.

a. Money is freely given retroactively as a reward for having done
something. This requires good record keeping about how much work someone
actually did and how much of the reward they should be likely to claim.
The down side is that it reduces enthusiasm. This is 'tipping' in my
vernacular.
b. Money is given pro-actively, for work you are about to, are doing or
have done and not published. This requires good quality checking and
being able to decide if the work was completed as agreed. This requires
a bunch of work to have been done to 'design' the work to be done
strictly. This is 'paying' in my lingo.

I feel much better about paying for work to be done than tipping, it
feels more right. Although it requires larger pools of money and a more
forthright approach deciding where this resource should be plowed in
order to curate the larger design. Debian hasn't been much for curating
anything, since it's more anarchistic than planned. But maybe it'd be
possible to have project goals at the debian level which such money
could be focused.

I think pro-active funding works better at the project level. If
Inkscape wants to spend it's money on sending contributors to
conferences, then that's a valid way to spend money on project goals. If
Blender wants to spend money on funding developers and movie makers to
stretch their software, also valid.

What we can't be is scared to develop the social and administrative
systems that make handling money effective. If we don't try, we won't
learn.

On Sat, 2013-06-15 at 00:50 +0200, Jonas Smedegaard wrote:
>  We should just be indifferent about it.  And that's what we practice
now as I see it.

We could be indifferent about lots of things in Debian. Licensing for
instance. \"Who cares! as long as we get the software delivered, only
lawyers care about some silly license and besides it could happen
outside of debian!\"

No, we care because it's important. Having meta data that describes the
connection between the user and the developer is an important
responsibility of the distributing platform. It is in fact the power of
the platform to deliver projects in a way that sustains and supplies
(users|attention|money|developers) to the upstream.

I don't think we're doing our job if we are not communicating an
important facet of the life-blood of our upstreams to our users. We are
instead acting to cut them off from each other unintentionally.

Martin Owens


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/1371260896.18562.26.camel@delen



Re: KickStarter for Debian packages - crowdfunding/donations for development

2013-06-14 Thread Charles Plessy
Le Fri, Jun 14, 2013 at 04:42:54PM -0400, Manu Sporny a écrit :
> 
> What if we did the following:
> 
> 1. If information exists in the package to distribute the donation
>automatically (these 35 people should get exactly these
>percentages), then we do that.
> 2. If the package lists the upstream project donation account, then we
>do that.
> 3. If there is no such information, then we send the money to the
>Debian project.

Dear Manu,

thanks for your work on web payments.  I hope that they will be a lead
contribution to the development of Free works.

In the case of Debian, I share with others the concern of having the packages
as a source of revenue, that would strongly reduce incentive for collaborative
maintainance of packages.  I think that for Debian, we need methods that pay
for the work to be done, rather than for the work that has been done.  Debian
has a large number of contributors, and we should encourage mobility between
projects, rather than make ourselves dependant to the project that would be
currently paying for our bills.

Most of the works we distribute are of a much smaller scale, where the approach
that you propose would not have such a negative side effect.  Moreover, some of
our upstreams who are already collecting donations may be worried of losing
opportunities to reveive when their work is distributed through Debian packages.
Here, PaySwarm can solve our problem.

As Paul showed you, we are collecting URLs for donations, and we probably miss
a lot of them.  If the PaySwarm project would provide with mechanisms to easily
communicate or discover this information, that would be very useful.  Then,
together with the information that the donation they received comes from a
Debian user, we could pass the URL to our upstream guide
(http://wiki.debian.org/UpstreamGuide), so that the packaging work (and
therefore the need for support) is reduced thanks to Upstream standardising his
work.

Have a nice week-end,

-- 
Charles Plessy
Debian Med packaging team,
http://www.debian.org/devel/debian-med
Tsurumi, Kanagawa, Japan


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20130615014503.ga11...@falafel.plessy.net



Re: KickStarter for Debian packages - crowdfunding/donations for development

2013-06-14 Thread Nikolaus Rath
Paul Wise  writes:
> I am very concerned about motivations of Debian project volunteers
> being distorted by money so I would suggest only allowing donations to
> Debian as a whole or directly to individual upstream projects.

I agree. It would add a whole new dimension to NMU'ing, orphaning,
adopting, and salvaging packages with a large user base. For example,
currently most people would probably happily accept co-maintainers even
if they're confident that they could do the work alone -- but what if
adding a co-maintainer means that a potential stream of personal revenue
will be cut in half?


Best,

   -Nikolaus

-- 
 »Time flies like an arrow, fruit flies like a Banana.«

  PGP fingerprint: 5B93 61F8 4EA2 E279 ABF6  02CF A9AD B7F8 AE4E 425C


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/871u84mg1n@vostro.rath.org



Re: KickStarter for Debian packages - crowdfunding/donations for development

2013-06-14 Thread David Kalnischkies
On Fri, Jun 14, 2013 at 4:44 PM, Arno Töll  wrote:
> On 14.06.2013 15:52, David Kalnischkies wrote:
>> She/he is doing a lot of work for sure, but I appreciate the software, not
>> that it is packaged with this. That I can get this software easily while
>> using Debian is something the whole project is responsible for, not just
>> the person who happens to maintain this specific package, so appreciation
>> for the package itself should go to the project as a whole.

(in the last sentence, I meant "Debian project", I see now that this
 might be mistaken as refering to upstream.)

> I disagree. Some packages make _a lot_ of work and some people spend
> thousands of hours to make Debian an excellent distribution and the
> package in particular useful and maintainable. This is in many cases
> not less work than being upstream, it's just another type of work which
> is often invisible.

My train of thought is that I will never be able to accept money for the
package 'apt'. Sure, I might do some stuff, but this package is 17 years old,
has seen contribution from all round the world and all that. What would give
me the right to claim it rather than anyone else who has ever worked on it.
And even if you haven't worked directly on it, we still have dependencies,
be it code wise or infrastructure (hardware, people, …). It can't be my job
as the maintainer¹ to rate the contribution of everyone even only remotely
involved to decide who is getting, who isn't and how much exactly.
(¹ which I am not, but lets keep it simple here)

Thankfully, 'apt' is a native package, so it has by default an entity which
accepts donations and is able to divide the money raised in a good way
across all involved people by sponsoring meetings, hardware and stuff.
And I believe the same entity should collect money channeled to "us" for
packaging non-native software rather than to individual maintainers.


I have on the other hand accepted money for working on APT (e.g. as GSoC
student) so the concept of APT being a magical cash-cow I can milk isn't
completely alien for me, but for me there is a difference between claiming
that I am the one who should get the money if you like the package and
being up for hire if you need something to like it (even more) – even
though its on the etch of being double-think.


Best regards

David Kalnischkies


--
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CAAZ6_fBeTULS1Q7gv8GthCHty0z=a+pte6pown03165omc1...@mail.gmail.com



Re: KickStarter for Debian packages - crowdfunding/donations for development

2013-06-14 Thread Jonas Smedegaard
Quoting Manu Sporny (2013-06-14 22:57:46)
> On 06/14/2013 05:36 AM, Martin Owens wrote:
> > On Thu, 2013-06-13 at 23:03 -0500, Gunnar Wolf wrote:
> >> We have never needed it
> > 
> > Such a mechanism would never be needed if the only people Debian 
> > should serve are technical users who are able to involve themselves 
> > in the project when they need it.
> > 
> > Excluding people with money is just another way of excluding people 
> > from Free Software development. I'm not so sure it's been as healthy 
> > for us or our users as assumed here.
> 
> I think this is an excellent point. The people that use Debian, more 
> specifically Ubuntu, are not always capable of jumping in and doing 
> the technical work themselves. It would be nice to give them a way to 
> contribute (donations), or promote work they want to see happen 
> (bounties).

Anyone can locate a particular Debian contributor and wire them 15 
Bitcoin.  No need for Debian to support that.

IMO Debian should not be against bounties or other forms of developer- 
or package-specific monetary boosting (read: distortion).  We should 
just be indifferent about it.  And that's what we practice now as I see 
it.

Some developers may be paid full time for the things they do in Debian, 
while others don't get a dime.

Some packages may get lots of love from maintainer with a full stomach 
and a house with a pool, while other package may survive with less 
attentive maintainers poking a bit now and then when not busy hunting or 
farming to support their families.

Debian do not promote a capitalist-driven boost, nor any other specific 
gaming of the system.  Debian is indifferent about how its developers 
find time and devotion for our shared project.

I really thought Gunnar nailed it:

> We have never needed it


 - Jonas

-- 
 * Jonas Smedegaard - idealist & Internet-arkitekt
 * Tlf.: +45 40843136  Website: http://dr.jones.dk/

 [x] quote me freely  [ ] ask before reusing  [ ] keep private


signature.asc
Description: signature


Re: KickStarter for Debian packages - crowdfunding/donations for development

2013-06-14 Thread Manu Sporny
On 06/14/2013 09:52 AM, David Kalnischkies wrote:
> On Thu, Jun 13, 2013 at 11:50 PM, Manu Sporny 
>  wrote:
>> A deeper discussion of the social concerns and solutions with
>> 
>> http://lists.debian.org/deity/2013/06/msg00054.html
> 
> I thought this was pretty shallow, but anyway:

Well, I meant deeper than the introductory e-mail about the topic. :)

> I believe its in some sense our responsibility to also collect the 
> information of how to show appreciation for the hard work of
> everyone involved in this project.

+1 - If we can get this right, we can make it easier to donate in a way
that is equitable. There are a significant number of organizations that
spend most of their donations on administrative staff to do funding drives.

PaySwarm allows you to see exactly where your money is going. We did
this by design in order to ensure transparency for people that want it.
The one down-side to sending money to an organization or a foundation is
that you have no idea where your money goes after you send it. More below...

> This especially means though, that it must be very very very clear 
> who is the receiver of the appreciation. I would personally be 
> surprised if the maintainer of package XYZ is getting a share via 
> "apt-donate xyz '5 coins'"

Yes, exactly right. PaySwarm lists every single entity that will receive
funds. If there are 50 maintainers and upstream authors listed with
varying amounts, the donor will be able to see exactly where every
single bit of their donation goes (if they want to see the details).

There is a video demonstrating what a purchase looks like from the browser:

https://www.youtube.com/watch?v=02vX36Ntxxk

and what one looks like from the command line (note the list of who is
getting paid and how much they're getting paid):

https://www.youtube.com/watch?v=0SEUnX-H2Qc

How you specify the entities being paid, which will be visible to the
person doing the donation, is outlined in this blog post:

https://hacks.mozilla.org/2013/04/payswarm-part-2/

-- manu

-- 
Manu Sporny (skype: msporny, twitter: manusporny, G+: +Manu Sporny)
Founder/CEO - Digital Bazaar, Inc.
blog: Meritora - Web payments commercial launch
http://blog.meritora.com/launch/


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/51bb8cda.5070...@digitalbazaar.com



Re: KickStarter for Debian packages - crowdfunding/donations for development

2013-06-14 Thread Paul Tagliamonte
On Fri, Jun 14, 2013 at 05:14:27PM -0400, Manu Sporny wrote:
> I agree, which is why the payment details live completely outside the
> Debian systems. The only thing you'd need to initiate payment is an
> e-mail address, or a PaySwarm financial account address, which looks
> like this:
> 
> https://dev.payswarm.com/i/manu/accounts/public-account

Interesting. So one would have to set up a payswarm instance to collect,
or piggy back on a public instance?

> requirements when you process payments; know your customer (KYC) is one
> of them. There are also anti-money laundering regulations that you must
> ensure you follow to comply with the law. All of these things are things

Which law? US law? We're an international org, are the laws standard
worldwide? We have Developers in every jurisdiction you can imagine :)

> that the Debian project doesn't want to (or have to) deal with wrt. the
> current proposal.
> 
> > I think allowing users to 'tip' would be nice, but it creates this 
> > system where now everyone wants to co-maint gcc, bash, libc, linux 
> > and not really "do" anything.
> 
> Right, which is why one solution is making it up to the package
> maintainers and software authors to figure out how payments should be
> split up. I think we all agree that tips should be distributed based on
> merit and that the maintainers have a pretty good idea of how that
> should go. In the event that the maintainers and upstream can't come to
> an agreement, they could always just opt to send the donation upstream
> to the Debian Project.

Right, but this leads to one of two things:

  - No money is shared with dependencies (leading to people flocking to
awesomewm, gnome, kde, chrome, wine, apache2, etc)

  - Money is shared with dependencies (leading to people flocking to
gcc, linux, libc6)

In both cases, Debian Project Members (non-uploading / non-maintainers)
don't get any love either. We could use pseudo-packages to help (e.g.
donate www.debian.org $1,000,000), but I can't imagine that's complete
(we'd never donate gsoc-team-alpha $500, for instance).

In the case where we do always split it, it kills a bit of the motivation
for tipping packages in general.


> > Payment systems in general tend to lead to "paybullying", which is 
> > something I'd really (really) like to avoid. I've always loved how 
> > un-corporate the Debian community is.
> 
> I agree that we really don't want that. Any suggestions about how we
> might be able to avoid it?

I'm not sure, but I've seen at least one high-profile F/OSS project
maintainer (with project email, writing from it) saying "I've written a
patch for this bug, it's done, but you need to give me money before I
release it". Putting an official system into place might make this more
common / easier to make look "official".

> Let me know if the above makes sense. I'd be happy to answer more
> questions if it would help. :)

Thank you! :)

> 
> > P.S., I do like your work on JSON-LD, Manu!
> 
> Thanks! Fun fact: We built JSON-LD because of PaySwarm. We needed to
> make sure that the core financial protocol could be extensible in a
> distributed way, and JSON-LD ended up being the solution for that.

Neat!

> 
> You can see how JSON-LD is used for the PaySwarm financial protocol and
> digital receipts here:
> 
> https://hacks.mozilla.org/2013/04/payswarm-part-2/
> 
> https://hacks.mozilla.org/2013/04/web-payments-with-payswarm-purchasing-part-3-of-3/
> 
> -- manu
> 
> -- 
> Manu Sporny (skype: msporny, twitter: manusporny, G+: +Manu Sporny)
> Founder/CEO - Digital Bazaar, Inc.
> blog: Meritora - Web payments commercial launch
> http://blog.meritora.com/launch/


Cheers,
  Paul

-- 
 .''`.  Paul Tagliamonte 
: :'  : Proud Debian Developer
`. `'`  4096R / 8F04 9AD8 2C92 066C 7352  D28A 7B58 5B30 807C 2A87
 `- http://people.debian.org/~paultag


signature.asc
Description: Digital signature


Re: KickStarter for Debian packages - crowdfunding/donations for development

2013-06-14 Thread Manu Sporny
On 06/14/2013 07:20 AM, Paul Tagliamonte wrote:
> In particular, I have concerns on how people *collect* this money.
> In the case of a package maintainer, we don't require we know who
> they are, nor their legal name, or even where the live. Just an
> active, working email and good changes that we can upload (hell, they
> could even just get sponsorship via debdiffs without a GPG key for
> all we know). In that case, verifiying identity wouldn't be something
> we could do, nor do I think such people would give their bank numbers
>  up.

I agree, which is why the payment details live completely outside the
Debian systems. The only thing you'd need to initiate payment is an
e-mail address, or a PaySwarm financial account address, which looks
like this:

https://dev.payswarm.com/i/manu/accounts/public-account

If a developer wants a payment, they can list a valid PaySwarm account,
or just provide their e-mail address. If they provide their e-mail
address, they'll get an e-mail from the PaySwarm payment processor
stating that someone has sent them some money.

It's up to the PaySwarm payment processors to verify names, addresses,
bank accounts, and credit card numbers. There are a number of legal
requirements when you process payments; know your customer (KYC) is one
of them. There are also anti-money laundering regulations that you must
ensure you follow to comply with the law. All of these things are things
that the Debian project doesn't want to (or have to) deal with wrt. the
current proposal.

> I think allowing users to 'tip' would be nice, but it creates this 
> system where now everyone wants to co-maint gcc, bash, libc, linux 
> and not really "do" anything.

Right, which is why one solution is making it up to the package
maintainers and software authors to figure out how payments should be
split up. I think we all agree that tips should be distributed based on
merit and that the maintainers have a pretty good idea of how that
should go. In the event that the maintainers and upstream can't come to
an agreement, they could always just opt to send the donation upstream
to the Debian Project.

> Payment systems in general tend to lead to "paybullying", which is 
> something I'd really (really) like to avoid. I've always loved how 
> un-corporate the Debian community is.

I agree that we really don't want that. Any suggestions about how we
might be able to avoid it?

> I'm very lukewarm about this right now, but I think with some sound 
> arguments, I'd warm up to it.

Let me know if the above makes sense. I'd be happy to answer more
questions if it would help. :)

> P.S., I do like your work on JSON-LD, Manu!

Thanks! Fun fact: We built JSON-LD because of PaySwarm. We needed to
make sure that the core financial protocol could be extensible in a
distributed way, and JSON-LD ended up being the solution for that.

You can see how JSON-LD is used for the PaySwarm financial protocol and
digital receipts here:

https://hacks.mozilla.org/2013/04/payswarm-part-2/

https://hacks.mozilla.org/2013/04/web-payments-with-payswarm-purchasing-part-3-of-3/

-- manu

-- 
Manu Sporny (skype: msporny, twitter: manusporny, G+: +Manu Sporny)
Founder/CEO - Digital Bazaar, Inc.
blog: Meritora - Web payments commercial launch
http://blog.meritora.com/launch/


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/51bb87b3.8000...@digitalbazaar.com



Re: KickStarter for Debian packages - crowdfunding/donations for development

2013-06-14 Thread Manu Sporny
On 06/14/2013 05:36 AM, Martin Owens wrote:
> On Thu, 2013-06-13 at 23:03 -0500, Gunnar Wolf wrote:
>> We have never needed it
> 
> Such a mechanism would never be needed if the only people Debian
> should serve are technical users who are able to involve themselves
> in the project when they need it.
> 
> Excluding people with money is just another way of excluding people
> from Free Software development. I'm not so sure it's been as healthy
> for us or our users as assumed here.

I think this is an excellent point. The people that use Debian, more
specifically Ubuntu, are not always capable of jumping in and doing the
technical work themselves. It would be nice to give them a way to
contribute (donations), or promote work they want to see happen (bounties).

Even if there are some developers that oppose monetary contribution in
any form, there are others that would love to develop for Debian more
(but can't feed their families on volunteering). That is, it's important
to not encode a personal opinion in software that silences those that
have a different opinion.

We can build this system such that those that don't want any donations
can pass it on to the Debian Project, or reject all donations outright.
We can also build this system such that those that would like donations
can state exactly how they want the donation to be split up among the
project participants. This approach ends up being fair to both sides and
doesn't silence either opinion.

-- manu

-- 
Manu Sporny (skype: msporny, twitter: manusporny, G+: +Manu Sporny)
Founder/CEO - Digital Bazaar, Inc.
blog: Meritora - Web payments commercial launch
http://blog.meritora.com/launch/


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/51bb83ca.1010...@digitalbazaar.com



Re: KickStarter for Debian packages - crowdfunding/donations for development

2013-06-14 Thread Manu Sporny
On 06/13/2013 10:33 PM, Paul Wise wrote:
> We already have means of donating to Debian. Making that more 
> accessible is on the todo and probably your work could help out 
> here.
> 
> http://www.debian.org/donations 
> http://lists.debian.org/debian-www/2013/05/msg00025.html

Great, would love to help out in some way. Who should I chat with about
that?

> We have a mechanism for maintainers to point folks at upstream 
> donation pages, it is only used by two packages so far though:
> 
> http://wiki.debian.org/UpstreamMetadata 
> http://upstream-metadata.debian.net/table/Donation

Yes, that's one of the problems we had identified. If the donation
mechanism is more complicated than 'apt-donate apache2 $5', then the
likelihood that it will be used goes down dramatically.

> Tying donations to one payment processor doesn't sound like a good 
> idea to me.

I agree. PaySwarm is designed to be an open payment standard. Once the
standard is finalized you'll be able to port your financial account from
one PaySwarm payment processor to another just as easily as people can
port their cellphones today. The specs are designed to prevent vendor
lock-in. There is only one payment processor today (Meritora), but once
other PaySwarm payment processors pop up, if you don't like Meritora for
whatever reason, you can switch (while preserving all of your financial
account data).

If you'd like to learn more about PaySwarm, there are a set of blog
posts that we did for Mozilla Hacks here:

https://hacks.mozilla.org/2013/04/web-payments-with-payswarm-identity-part-1-of-3/

https://hacks.mozilla.org/2013/04/payswarm-part-2/

https://hacks.mozilla.org/2013/04/web-payments-with-payswarm-purchasing-part-3-of-3/

> I am very concerned about motivations of Debian project volunteers 
> being distorted by money so I would suggest only allowing donations 
> to Debian as a whole or directly to individual upstream projects.

We could certainly do the former, although the upstream software authors
may not like that because it could be construed as the overall Debian
project overreaching into what should be a conversation between the
package maintainer and the upstream authors.

We could also do the latter, but I'd like there to be a possibility of a
package maintainer getting a reward of some kind for past work
performed. The Debian Security team does an excellent job, and have
saved us on numerous occasions. I'd love to have some way of sending
them something in appreciation.

What if we did the following:

1. If information exists in the package to distribute the donation
   automatically (these 35 people should get exactly these
   percentages), then we do that.
2. If the package lists the upstream project donation account, then we
   do that.
3. If there is no such information, then we send the money to the
   Debian project.

Does that sound reasonable?

> I am also concerned about the distortions that monetisation has had 
> on the web and worry about the consequences of embedding this into 
> browsers. Both the modern web and modern web browsers are very 
> concerning in general though.

I agree, although we may differ on what the root of the problem is. I
don't think it's money or monetization, as that is a neutral thing, just
like technology (and Debian). It's not inherently evil or good. We tend
to think of it more from a systems perspective. Certain systems result
in certain behaviors.

The Web's monetization has happened mainly based on advertising dollars
(Google, Facebook, Yahoo, Twitter, etc.). The system of advertising
results in an attempt to please the advertisers, since they're the
customers. Citizens are consumers, and tend to be viewed as 2nd class
citizens to the advertisers. We think that there are better models for
the Web, but that the core infrastructure for those models doesn't exist
yet.

The fundamental thing that we're trying to do with PaySwarm is to
democratize finance. That means that we're trying to put the tools that
are only available to large corporations and Wall street into the hands
of most people. It's analogous to how the Web democratized publishing...
pretty easy to put something out there these days that can reach
hundreds of millions of people. We want to do that to our global
financial infrastructure.

I think software developers are going to play a major role in building
this new Web-based financial infrastructure. I also agree with you that
we have to be careful and methodical about how we build this system. The
hope is that by doing it right, we can bring more money into the
ecosystem (in a good way) and tackle some long-standing problems
(under-staffing, interface design, testing, infrastructure, etc.).

I've been using Debian since the late 1990s. I love this project. I'm
primarily interested in trying to build the core donations
infrastructure in a way that is going to truly help the Debian community.

-- manu

-- 
Manu Sporny (skype: msporny, twitter: manusporny, G+: +Manu Sporny)
Foun

Re: KickStarter for Debian packages - crowdfunding/donations for development

2013-06-14 Thread Scott Howard
On Thu, Jun 13, 2013 at 5:50 PM, Manu Sporny  wrote:
> I'd like to add a feature to apt that enables people to donate money or
> crowdfund a particular project or developer. The inspiration for the
> project comes from [1]Gittip.

Another issue/can of worms: Someone started a thread on an ubuntu list
asking how to get a company's proprietary software into multiverse.
The generic response, as usual, is to either us a PPA or try to twist
a DDs arm to get it into non-free (if it is redistributable). DDs
don't want to use their volunteer time on non-free projects, but while
thinking of the response I was aware about how financial incentive may
change a DD's opinion. As a member of the bitcoin team in debian, I
also see how easy it could be in the future to semi-anonymously
receive payment from and to anywhere in the world for such work, and
that this may already be going on already in Debian or other distros.

Allowing quick in-distribution monetization of effort potentially
opens up a market for DDs to work on packaging specific commercial
(and free) projects for pay. On one hand, it feels like it ruins the
"free-ness" (both in beer and speech) of Debian (although there is no
rule saying you have to pay to get in), but on the other hand
explicitly rejecting monetization feels like we are discriminating
against the commercial field of use, which is also a non-free act
(kind of what Martin Owen is saying).

I don't know how I feel about this, fundamentally, yet. But my gut
didn't like the idea of having a wiki page where DDs can post their
hourly rates for packaging software for Debian...

This is just a warning of one way I can see such an official donation
system evolving in a few years. I still prefer donating to the project
as a whole just to avoid this mess for now.


~Scott


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CANg8-dDcpXTXEhdz+LbnQsz1kWKYS3zXZpSJ1krRzQN3==v...@mail.gmail.com



Re: KickStarter for Debian packages - crowdfunding/donations for development

2013-06-14 Thread Arno Töll
Hi,

On 14.06.2013 15:52, David Kalnischkies wrote:
> She/he is doing a lot of work for sure, but I appreciate the software, not
> that it is packaged with this. That I can get this software easily while
> using Debian is something the whole project is responsible for, not just
> the person who happens to maintain this specific package, so appreciation
> for the package itself should go to the project as a whole.

I disagree. Some packages make _a lot_ of work and some people spend
thousands of hours to make Debian an excellent distribution and the
package in particular useful and maintainable. This is in many cases
not less work than being upstream, it's just another type of work which
is often invisible.

Not saying we should implement a user interface for (micro-) donations
to Debian package maintainers, but I'm sure people being grown up enough
to become contractually capable, are also able to differentiate whether
they would like to donate to upstream or to the distributor and both
deserve their part of the cake.

Now, if comes up to discuss whether we should proxy donations from
people to maintainers I'm unsure. I'd rather be in favor of collecting
these donations to a common pool and divide it equal amongst maintainers
who opt-in for very simple reasons:

* Opt-In for the reason Mr. Tagliamonte mentioned: Some of us are
pseudonymous identities, some may not want to share bank account data etc.

* A maintainer is not responsible alone for the work. Personally I
cannot value enough the tremendous work the Security Team spends with
packages I and $boss uses daily for profit.

* A maintainer builds his work on the shoulders of others, like you,
being the APT maintainer, like maintainers of dependencies, like the
Release Team hating your debdiffs. And, of course, our non-uploading DDs
who spend lots of time to maintain the website, make translations, etc.

* Obviously some very important packages virtually get zero attention.
You know, people donate what they see, be it apache2 (taken because I
maintain this), but who would donate for libz we depend upon? This is an
imbalance to have in mind.

However, is all of that a reason not to accept money from people at all?
I don't think so. We should just ensure it's fair and balanced _within_
Debian and clear to people donating that the money goes (to what
fraction) to whom.

-- 
with kind regards,
Arno Töll
IRC: daemonkeeper on Freenode/OFTC
GnuPG Key-ID: 0x9D80F36D



signature.asc
Description: OpenPGP digital signature


Re: KickStarter for Debian packages - crowdfunding/donations for development

2013-06-14 Thread David Kalnischkies
On Thu, Jun 13, 2013 at 11:50 PM, Manu Sporny  wrote:
> A deeper discussion of the social concerns and solutions with
> integrating donations and fundraising into Debian packages can be found
> here:
>
> http://lists.debian.org/deity/2013/06/msg00054.html


I thought this was pretty shallow, but anyway:

My personal opinion is that while many projects have means of accepting
donations most of the time its not as easy to discover or to use.

"Donating" to projects which are not free-to-use is a lot easier as the same
interface over which you discover this incredible masterpiece of a project
is also the place where you find how to channel money to it – simply because
you have to pay for it before you use it.

And I am not talking about money exclusively here: Some projects have
licenses encouraging buying the developer a beer if you see her/him,
to send a postcard, to donate to favorite entity, … all of which isn't
that accessible if you aren't highly determined to find it.

As a distribution we work hard on making the best software available via
an easy interface, we collect screenshots, reviews and votes for them (in
various way with various states of completion in various downstreams) so
I believe its in some sense our responsibility to also collect the
information of how to show appreciation for the hard work of everyone
involved in this project.

That Debian is able to collect and deliver appreciations to developers via
a simple interface was proven in 2010 with thanks.debian.net (the original
is gone) which was like the best idea since sliced bread. :)
[I know that others didn't like the "spam" – so much for uncontroversial]

So I don't believe Debian should be the entity actually collecting (at least)
the more physical/dangerous forms of appreciation, this is really something
the projects itself have to work out, but we are good at collection metadata
and I think many would appreciate being able to appreciate more easily.


This especially means though, that it must be very very very clear who is
the receiver of the appreciation. I would personally be surprised if the
maintainer of package XYZ is getting a share via "apt-donate xyz '5 coins'"
(I use this interface now just for brevity).
She/he is doing a lot of work for sure, but I appreciate the software, not
that it is packaged with this. That I can get this software easily while
using Debian is something the whole project is responsible for, not just
the person who happens to maintain this specific package, so appreciation
for the package itself should go to the project as a whole.


Best regards

David Kalnischkies


--
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CAAZ6_fBEh6grc=u=+xoghj_fmqlo3nf0enx2y9jkocumk2g...@mail.gmail.com



Re: KickStarter for Debian packages - crowdfunding/donations for development

2013-06-14 Thread Paul Tagliamonte
On Thu, Jun 13, 2013 at 11:03:36PM -0500, Gunnar Wolf wrote:
[..]
> work. That's all fine. We have a mechanism in place to help people
> donate money to Debian as a whole. That's also fine. But I'd very much
> rather keep both things separate — Not instate mechanisms in Debian to
> get funds to individual developer. We have never needed it, and from
> the discussions I have taken part in or witnessed, I really doubt we
> would need it now.


In particular, I have concerns on how people *collect* this money. In
the case of a package maintainer, we don't require we know who they are,
nor their legal name, or even where the live. Just an active, working
email and good changes that we can upload (hell, they could even just
get sponsorship via debdiffs without a GPG key for all we know). In that
case, verifiying identity wouldn't be something we could do, nor do I
think such people would give their bank numbers up.


I think allowing users to 'tip' would be nice, but it creates this
system where now everyone wants to co-maint gcc, bash, libc, linux and
not really "do" anything.


Payment systems in general tend to lead to "paybullying", which is
something I'd really (really) like to avoid. I've always loved how
un-corporate the Debian community is.


I'm very lukewarm about this right now, but I think with some sound
arguments, I'd warm up to it.


P.S., I do like your work on JSON-LD, Manu!

Cheers,
  Paul

-- 
 .''`.  Paul Tagliamonte 
: :'  : Proud Debian Developer
`. `'`  4096R / 8F04 9AD8 2C92 066C 7352  D28A 7B58 5B30 807C 2A87
 `- http://people.debian.org/~paultag


signature.asc
Description: Digital signature


Re: KickStarter for Debian packages - crowdfunding/donations for development

2013-06-14 Thread Martin Owens
On Thu, 2013-06-13 at 23:03 -0500, Gunnar Wolf wrote:
> We have never needed it

Such a mechanism would never be needed if the only people Debian should
serve are technical users who are able to involve themselves in the
project when they need it.

Excluding people with money is just another way of excluding people from
Free Software development. I'm not so sure it's been as healthy for us
or our users as assumed here.

Martin,


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/1371202594.18562.3.camel@delen



Re: KickStarter for Debian packages - crowdfunding/donations for development

2013-06-13 Thread Gunnar Wolf
Paul Wise dijo [Fri, Jun 14, 2013 at 10:33:58AM +0800]:
> (...)
> Tying donations to one payment processor doesn't sound like a good idea to me.
> 
> I am very concerned about motivations of Debian project volunteers
> being distorted by money so I would suggest only allowing donations to
> Debian as a whole or directly to individual upstream projects.
> 
> I am also concerned about the distortions that monetisation has had on
> the web and worry about the consequences of embedding this into
> browsers. Both the modern web and modern web browsers are very
> concerning in general though.

FWIW, I agree with Paul here. Some Debian people have requested
(individually) for public sponsorship to their free software-related
work. That's all fine. We have a mechanism in place to help people
donate money to Debian as a whole. That's also fine. But I'd very much
rather keep both things separate — Not instate mechanisms in Debian to
get funds to individual developer. We have never needed it, and from
the discussions I have taken part in or witnessed, I really doubt we
would need it now.

Of course, I cannot decline the offer in the name of the whole
project. I just state my opinion.


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20130614040336.gb41...@gwolf.org



Re: KickStarter for Debian packages - crowdfunding/donations for development

2013-06-13 Thread Paul Wise
We already have means of donating to Debian. Making that more
accessible is on the todo and probably your work could help out here.

http://www.debian.org/donations
http://lists.debian.org/debian-www/2013/05/msg00025.html

We have a mechanism for maintainers to point folks at upstream
donation pages, it is only used by two packages so far though:

http://wiki.debian.org/UpstreamMetadata
http://upstream-metadata.debian.net/table/Donation

Tying donations to one payment processor doesn't sound like a good idea to me.

I am very concerned about motivations of Debian project volunteers
being distorted by money so I would suggest only allowing donations to
Debian as a whole or directly to individual upstream projects.

I am also concerned about the distortions that monetisation has had on
the web and worry about the consequences of embedding this into
browsers. Both the modern web and modern web browsers are very
concerning in general though.

-- 
bye,
pabs

http://wiki.debian.org/PaulWise


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/CAKTje6HrFjkLbckHYUD6361p89CPisqNo1Z0OW1Fx1aXX=z...@mail.gmail.com



KickStarter for Debian packages - crowdfunding/donations for development

2013-06-13 Thread Manu Sporny
I'd like to add a feature to apt that enables people to donate money or
crowdfund a particular project or developer. The inspiration for the
project comes from [1]Gittip.

I've already contacted the APT project and they directed me here, as
they felt it needed to be discussed by the larger Debian community.

My name is [2]Manu Sporny and I'm the current chair of the Web Payments
group at the World Wide Web Consortium (W3C). I'm also the project
leader for the [3]PaySwarm project, which is an initiative to create a
universal payment standard for the Web. We're also currently [4]working
with Mozilla to build payments into the browser. Each of these
initiatives is a patent-free and royalty-free open standard that we'd
like to try and apply to the Debian project.

The general idea with apt-based donations is to enable every package in
Debian to be able to accept funding and/or donations for the project.
Here are some of the ideas we were kicking around:

 1. Simple donations. Donate a fixed dollar amount $1, $2.50,
 $5, or $10 to a project. This could be opt-in or automatic.

 2. Split donations. Donate a fixed dollar amount to a project,
 where 50% of the donation goes to the project and 50% of the
 donation is split equally among the project dependencies (by
 reading the Debian control file).

 3. Recurring donations. Same as the first item above, but
 happens on a recurring basis.

 4. Variable split donations. Same as the second item above, but
 the percentages of where the money goes is specified by the
 maintainer in the Debian control file or other included file.

 5. The ability to donate arbitrary amounts to registered
 Debian developers, or registered project goals.

We could add these features to a new command in apt that can be executed
via the command-line:

apt-donate [package] [amount]

A deeper discussion of the social concerns and solutions with
integrating donations and fundraising into Debian packages can be found
here:

http://lists.debian.org/deity/2013/06/msg00054.html

There is a series on PaySwarm for developers on the Mozilla Hacks blog,
if you want more background information on how PaySwarm works:

[5]PaySwarm: Identity (part 1 of 3)
[6]PaySwarm: Assets and Listings (part 2 of 3)
[7]PaySwarm: Purchasing (part 3 of 3)

We'd tie donations to the [8]Meritora PaySwarm payment processor. If
there is interest in adding this sort of feature in apt (or apt-utils),
then we'd be happy to do the work on the software and make it happen.

I'd like to understand what the expectations might be for a feature like
this. I'd also like to answer any questions about the Web Payments /
PaySwarm standardization work as well. If it would help, I can jump on a
quick voice chat with someone closely associated with the Debian Project
to discuss in more depth.

My contact info (in preferred order):
Skype: msporny
VoIP: sip:mspo...@digitalbazaar.com
G+ Hangout: +Manu Sporny
Mobile: 540-641-1942

Is there interest in this feature?

-- manu

1. https://www.gittip.com/
2. http://www.linkedin.com/in/manusporny
3. https://payswarm.com/
4. http://manu.sporny.org/2013/browser-payments/
5.
https://hacks.mozilla.org/2013/04/web-payments-with-payswarm-identity-part-1-of-3/
6. https://hacks.mozilla.org/2013/04/payswarm-part-2/
7.
https://hacks.mozilla.org/2013/04/web-payments-with-payswarm-purchasing-part-3-of-3/
8. http://blog.meritora.com/launch/

-- 
Manu Sporny (skype: msporny, twitter: manusporny, G+: +Manu Sporny)
Founder/CEO - Digital Bazaar, Inc.
blog: Meritora - Web payments commercial launch
http://blog.meritora.com/launch/


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/51ba3e97.7010...@digitalbazaar.com



Re: Debian packages

2009-03-11 Thread Goswin von Brederlow
Gabriel McCall  writes:

> Dear Debian Package Maintainers,
>
> Is there a definitive way to download packages for a machine that,
> due to it's remote location, does not have an internet connection?
> What I'm looking for is a complete file that would contain all the
> packages that a specific application is dependant upon whether or
> not a given operating system already included said packages in it's
> makeup.

This sounds like you want a partial mirror. There is apt-zip for
package management with sneakernet (you carry an USB stick/disk
around) but that will only list needed packages and skip already
installed ones.

If I had this problem I would use reprepro. Reprepro is an easy tool
to create or mirror a debian archive for use with apt. It also alows
filtering the package list to create partial mirrors. It is easy to
mirror just a list of packages. The hard part and the part that is
missing is to create the package list so that all
depends/recommends[/suggests] are satisfied. Maybe you can borrow code
from debian-cd for that.

> If something like that is not available, would there be a way to
> create files full of packages that would cover a large range of
> basic packages in an index such as a "pool"?  This would allow users
> to download and burn a CD full of the latest packages for specific
> uses such as Multimedia, games, development, or perhaps entire
> desktops that could be added to existing operating systems.
>
> For example, I cannot install a second desktop environment such as
> Xfce or Kde on a machine that has Gnome installed without using a
> package manager and an internet connection.  However, if I were able
> to download a set of package files which came with thier
> dependencies neatly bundled in an index that could be read by
> something like synaptic; I could then install the alternate desktop
> environment without the need for an internet connection.
>
> Simpler yet would be a multimedia bundle, or perhaps a networking
> bundle.  If I simply wanted a media player installed on my system
> and could download a package bundle which contained everything
> needed to install that specific media player it would be much easier
> than trying to download each individual package and install them all
> as individual binary packages.

There are way too many kinds of bundles people might want for debian
to provide them all and only providing a select few will always leave
most people wanting. So I think that would be a bad idea.

> Please help me understand why this is not something that is easliy found.
>
> Thank you for your time
>
> Gabriel  (hellion_darkl...@inbox.com)

MfG
Goswin


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org



Re: Debian packages

2009-03-11 Thread Mehdi Dogguy
Gabriel McCall wrote:
> Dear Debian Package Maintainers,
> 
> Is there a definitive way to download packages for a machine that, due to 
> it's remote location, does not have an internet connection?  What I'm looking 
> for is a complete file that would contain all the packages that a specific 
> application is dependant upon whether or not a given operating system already 
> included said packages in it's makeup.  
> 

I think that AptOnCd is what your are looking for:

http://aptoncd.sourceforge.net/

Cheers,

-- Mehdi Dogguy


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org



Re: Debian packages

2009-03-10 Thread Daniel Dickinson
On Tue, 10 Mar 2009 09:54:31 -0800
Gabriel McCall  wrote:

> Dear Debian Package Maintainers,
> 
> Is there a definitive way to download packages for a machine that,
> 
> Simpler yet would be a multimedia bundle, or perhaps a networking
> bundle.  If I simply wanted a media player installed on my system and
> could download a package bundle which contained everything needed to
> install that specific media player it would be much easier than
> trying to download each individual package and install them all as
> individual binary packages.

You may want to look at http://wiki.debian.org/DebianCustomCD if
apt-zip doesn't fulfill your needs.

> 
> Please help me understand why this is not something that is easliy
> found.

Because it's not easy to do.  Development of easier alternatives for
making custom pools is ongoing (though slowly as most developers who
need a pool also have hard drive space for a mirror for a particular
architecture).

Regards,

Daniel

-- 
And that's my crabbing done for the day.  Got it out of the way early, 
now I have the rest of the afternoon to sniff fragrant tea-roses or 
strangle cute bunnies or something.   -- Michael Devore
GnuPG Key Fingerprint 86 F5 81 A5 D4 2E 1F 1C  http://gnupg.org


signature.asc
Description: PGP signature


Re: Debian packages

2009-03-10 Thread Julien BLACHE
Gabriel McCall  wrote:

Hi,

> Is there a definitive way to download packages for a machine that, due
> to it's remote location, does not have an internet connection?  What
> I'm looking for is a complete file that would contain all the packages
> that a specific application is dependant upon whether or not a given
> operating system already included said packages in it's makeup.

apt-zip might be what you're looking for.

JB.

-- 
 Julien BLACHE - Debian & GNU/Linux Developer -  
 
 Public key available on  - KeyID: F5D6 5169 
 GPG Fingerprint : 935A 79F1 C8B3 3521 FD62 7CC7 CD61 4FD7 F5D6 5169 


-- 
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org



Debian packages

2009-03-10 Thread Gabriel McCall
Dear Debian Package Maintainers,

Is there a definitive way to download packages for a machine that, due to it's 
remote location, does not have an internet connection?  What I'm looking for is 
a complete file that would contain all the packages that a specific application 
is dependant upon whether or not a given operating system already included said 
packages in it's makeup.  

If something like that is not available, would there be a way to create files 
full of packages that would cover a large range of basic packages in an index 
such as a "pool"?  This would allow users to download and burn a CD full of the 
latest packages for specific uses such as Multimedia, games, development, or 
perhaps entire desktops that could be added to existing operating systems.  

For example,  I cannot install a second desktop environment such as Xfce or Kde 
on a machine that has Gnome installed without using a package manager and an 
internet connection.  However, if I were able to download a set of package 
files which came with thier dependencies neatly bundled in an index that could 
be read by something like synaptic; I could then install the alternate desktop 
environment without the need for an internet connection.  

Simpler yet would be a multimedia bundle, or perhaps a networking bundle.  If I 
simply wanted a media player installed on my system and could download a 
package bundle which contained everything needed to install that specific media 
player it would be much easier than trying to download each individual package 
and install them all as individual binary packages.

Please help me understand why this is not something that is easliy found.

Thank you for your time

Gabriel  (hellion_darkl...@inbox.com)

 


--
To UNSUBSCRIBE, email to debian-project-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org



Re: buildd redundancy (was Re: Recompilation of ALL Debian packages ...

2006-10-04 Thread Wouter Verhelst
On Sun, Sep 24, 2006 at 02:32:27PM -0400, Nathanael Nerode wrote:
> It's not reasonable to rely on one single machine like that: apart from the 
> mess that would happen if it went down or the person uploading its packages 
> took a week's vacation,

As opposed to the "mess" that does happen right now when that same
person who does powerpc, mips, mipsel, and a slew of other buildd
machines goes on vacation?

Right.

-- 
 Home is where you have to wash the dishes.
  -- #debian-devel, Freenode, 2004-09-22


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Re: Recompilation of ALL Debian packages ...

2006-09-24 Thread Nathanael Nerode
martin f krafft wrote:

> also sprach Russ Allbery <[EMAIL PROTECTED]> [2006.09.01.0241 +0200]:
>> Rebuilding every package really doesn't buy you that much in the
>> way of security.
> 
> This is arguable and I don't want to go there. The reason I am
> pushing for this is because of two of my clients, who have been
> wanting to use Debian for three years now but consciously decided
> against it, because it is not guaranteed that the sources and the
> binaries in our archives correspond for all architectures. They are
> well aware that trojans can still exist, but it's an entirely
> different thing whether they exist in source and hence in all
> architectures (which would result in some serious negative feedback
> or even revocation of upload rights), or just in one of the binaries
> and hence would be much harder to detect/analyse.

How big are your clients?  If they're good-sized companies with a spare 
computer, they can compile all the packages they use locally from Debian 
source with not *too* much work.

-- 
Nathanael Nerode  <[EMAIL PROTECTED]>

Bush admitted to violating FISA and said he was proud of it.
So why isn't he in prison yet?...


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Re: Recompilation of ALL Debian packages ...

2006-09-24 Thread Nathanael Nerode
Roberto C. Sanchez wrote:

> Is it not part of the process of becoming a DD (or sponsorship of
> packages for non-DDs) learning the "responsible" way to build packages.
> That is, developers are taught to use tools like pbuilder or sbuild in
> order to ensure that packages build cleanly.  I'm not saying that
> mistakes will never occur.  However, I would think that the vast
> majority of the people will be responsible and do it correctly the vast
> majority of the time.

Yeah, but a 5% error rate, or even a 1% error rate, is still a hell of a lot
of packages for QA to find and fix.

-- 
Nathanael Nerode  <[EMAIL PROTECTED]>

Bush admitted to violating FISA and said he was proud of it.
So why isn't he in prison yet?...


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



buildd redundancy (was Re: Recompilation of ALL Debian packages ...

2006-09-24 Thread Nathanael Nerode
martin f krafft wrote:

> also sprach Henning Makholm <[EMAIL PROTECTED]> [2006.08.31.1641 +0200]:
>> Please read up on the regular (every few months) discussions about
>> "source-only uploads" in the list archives. (Capsule summary: yes,
>> it would be easy to do, but there is no consensus that it would be
>> *desirable* to do so).
> 
> For instance: http://lists.debian.org/debian-devel/2006/07/msg00544.html
> 
> You say people have argued against source-only uploads, but I think
> source-only is actually not what we are talking about. The subject
> line says: "recompile...".
> 
> Whether or not requiring binary packages in the upload as a QA sort
> of measure is a good thing is a tangential debate.
> 
> I would like to know why we can't just discard those binaries and
> rebuild them on trusted machines. Then we get the best of all
> worlds.

This is really a nice idea.  It is *not* reasonable to do this until
we get the long awaited redundancy in the i386 buildd system.
Currently there is one buildd with one maintainer, unless something
has changed since http://release.debian.org/etch_arch_qualify.html was
last updated.  (Has something changed?  That looks out of date, but I don't
know who to ask.)

It's not reasonable to rely on one single machine like that: apart from the 
mess that would happen if it went down or the person uploading its packages 
took a week's vacation,  a compromise of that one machine will root every 
i386 Debian system.  (I hope it's at least as secure as ftpmaster, but two 
single points of failure are worse than one.)

Redundancy is needed for alpha, amd64, hppa, arm, and ia64 as well,
according to that chart.

-- 
Nathanael Nerode  <[EMAIL PROTECTED]>

Bush admitted to violating FISA and said he was proud of it.
So why isn't he in prison yet?...


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



buildd redundancy (was Re: Recompilation of ALL Debian packages ...

2006-09-24 Thread Nathanael Nerode
Sven Luther wrote:

> Accordying to James Troup, whom i asked exactly that at some past debconf,
> this is because if there is some lag in the x86 buildd, then loads of user
> will complain to them about non-installable packages, and the
> ftp-masters/buildd administrators being volunteers and not paid and
> everything, don't care about being given such pressure.

Right: so if we had *two* x86 buildds at *different* locations and *two* 
people who could check the logs and upload from *each* buildd, then as long 
as both people didn't go on vacation simultaneously, and the machines didn't
crash simultaneously, then the buildds could handle it.

Buildd redundancy is supposed to be an architecture release requirement;
it was waived because so many architectures were flunking it.  Is any 
progress being made on fixing this for etch and the future?

-- 
Nathanael Nerode  <[EMAIL PROTECTED]>

Bush admitted to violating FISA and said he was proud of it.
So why isn't he in prison yet?...


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Re: Recompilation of ALL Debian packages ...

2006-09-05 Thread Henning Makholm
Scripsit "James R. Van Zandt" <[EMAIL PROTECTED]>

>  - Allow an automated comparison of the two .debs.  This would take
>some work to set up, but I would hope to detect a binary that
>doesn't correspond to the claimed sources.  Also incorrect version
>of a compiler and different library versions than claimed in the
>dependencies.

There are many build tools that embed timestamps into the files being
built, each in their own way and with their own format.  Building the
same package twice in the same, clean, environment will in general
lead to .debs where the content of binary files differ in many places.

An automated comparison would need package-specific overrides for a
nontrivial percentage of our source packages. If the maintainer
declares the overrides, we don't gain security against deliberate
trojanings. If not, then whom _do_ we trust enough to maintain the
override database? And what useful work would they have to skip in
order to main comparison overrides?

-- 
Henning Makholm  "40.9931 lightfortnight-barns per nanopint"


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Re: Recompilation of ALL Debian packages ...

2006-09-04 Thread Henrique de Moraes Holschuh
On Mon, 04 Sep 2006, James R. Van Zandt wrote:
> Henrique de Moraes Holschuh <[EMAIL PROTECTED]> wrote:
> >   >  - Eliminate the wait for the buildd for the first architecture.
> >
> >   Not acceptable.
> 
> Rather, you would not find that acceptable.

No, it's just that such "install and override later" would not be an
acceptable way to implement the solution, given the reasons why the solution
was proposed.

> >   If you are to replace the uploaded binary debs with ones rebuilt from
> >   source, do it right: do not install the "untrusted" binary debs to the
> >   archive anywhere, and don't let them get to incoming.d.o, either.
> 
> I would rather that Debian offered users the choice of a more timely
> binary compiled by the DD or a later binary compiled by a buildd.

As long as the debs end up in another suite, or somewhere else away from the
main archive.

Otherwise, we are better off just leaving things as they are right now.

> >>  - Allow an automated comparison of the two .debs.  
> >
> >This is worth doing, but difficult to get right.
> 
> Having both versions available would help in diagnosing any
> differences, especially while the comparison utility was still being
> tuned.

Nothing against this, but the original debs should not be anywhere they
could be confused with the autobuilt ones.  Otherwise, you defeat the
purpose of such debs.

There are alternate ways, but they are much more complicated to deploy
(except maybe for signed debs, which are basically already supported in most
tools but dak).

-- 
  "One disk to rule them all, One disk to find them. One disk to bring
  them all and in the darkness grind them. In the Land of Redmond
  where the shadows lie." -- The Silicon Valley Tarot
  Henrique Holschuh


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Re: Recompilation of ALL Debian packages ...

2006-09-04 Thread James R. Van Zandt

Henrique de Moraes Holschuh <[EMAIL PROTECTED]> wrote:
>   >  - Eliminate the wait for the buildd for the first architecture.
>
>   Not acceptable.

Rather, you would not find that acceptable.

>   It will cause a time window where a trojaned binary package
>   might be active, 

True.

>   and since it would later have a new clean one replacing it,
>   it would be even worse to detect the problem.
>
>   If you are to replace the uploaded binary debs with ones rebuilt from
>   source, do it right: do not install the "untrusted" binary debs to the
>   archive anywhere, and don't let them get to incoming.d.o, either.

I would rather that Debian offered users the choice of a more timely
binary compiled by the DD or a later binary compiled by a buildd.

>>  - Allow an automated comparison of the two .debs.  
>
>This is worth doing, but difficult to get right.

Having both versions available would help in diagnosing any
differences, especially while the comparison utility was still being
tuned.

- Jim Van Zandt


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Re: Recompilation of ALL Debian packages ...

2006-09-04 Thread Henrique de Moraes Holschuh
On Mon, 04 Sep 2006, James R. Van Zandt wrote:
> >   You are right, I wrote source-only upload, but obviously 
> >   upload-binary-and-remove-it is better policy.
> 
> I suggest that the uploaded binary be kept temporarily, for two
> purposes:  
> 
>  - Eliminate the wait for the buildd for the first architecture.

Not acceptable.  It will cause a time window where a trojaned binary package
might be active, and since it would later have a new clean one replacing it,
it would be even worse to detect the problem.

If you are to replace the uploaded binary debs with ones rebuilt from
source, do it right: do not install the "untrusted" binary debs to the
archive anywhere, and don't let them get to incoming.d.o, either.

>  - Allow an automated comparison of the two .debs.  This would take

This is worth doing, but difficult to get right.

-- 
  "One disk to rule them all, One disk to find them. One disk to bring
  them all and in the darkness grind them. In the Land of Redmond
  where the shadows lie." -- The Silicon Valley Tarot
  Henrique Holschuh


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Re: Recompilation of ALL Debian packages ...

2006-09-04 Thread Gunnar Wolf
martin f krafft dijo [Sat, Sep 02, 2006 at 08:42:34AM +0200]:
> also sprach Russ Allbery <[EMAIL PROTECTED]> [2006.09.02.0141 +0200]:
> > I honestly think the security argument for doing this is silly.
> 
> Clients do not want to hear something like that.

Please... Do you mean they trust me (as an unknown person with upload
privileges to Debian) to produce proper sources, but to trojan the
binary packages? Do they think that all of the other DDs (or a
significant number of them anyway) will check my .orig.tar.gz is the
same as upstream's, and that my .diff.gz is sane?

I don't buy that as an argument. I do support rebuilding everything,
to ensure buildability of arch: all packages and to ensure
buildability under the architecture on which arch-dependent packages
were originally built, ensuring dependencies are complete and so.

Greetings,

-- 
Gunnar Wolf - [EMAIL PROTECTED] - (+52-55)5623-0154 / 1451-2244
PGP key 1024D/8BB527AF 2001-10-23
Fingerprint: 0C79 D2D1 2C4E 9CE4 5973  F800 D80E F35A 8BB5 27AF


signature.asc
Description: Digital signature


Re: Recompilation of ALL Debian packages ...

2006-09-04 Thread James R. Van Zandt

Matej Cepl <[EMAIL PROTECTED]> wrote:
>   On Thu 31. August 2006 12:47, you wrote:
>   > Without a binary version someone upload (and therefor should
>   > have tested), he could always claim his upload would have
>   > worked if the buildds would not have mangled it. So there is
>   > at least one package everyone can check who is to blame for
>   > it.
>
>   You are right, I wrote source-only upload, but obviously 
>   upload-binary-and-remove-it is better policy.

I suggest that the uploaded binary be kept temporarily, for two
purposes:  

 - Eliminate the wait for the buildd for the first architecture.

 - Allow an automated comparison of the two .debs.  This would take
   some work to set up, but I would hope to detect a binary that
   doesn't correspond to the claimed sources.  Also incorrect version
   of a compiler and different library versions than claimed in the
   dependencies.

- Jim Van Zandt


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Re: Recompilation of ALL Debian packages ...

2006-09-04 Thread Mark Brown
On Fri, Sep 01, 2006 at 05:03:29PM +0200, Michelle Konzack wrote:

> I have tried to RECOMPILE some packages in Sarge but failed.
> The Binaries are working.  It seems, thet the Maintainer had
> used a machine where the Build was successfull, but no other
> one can do it because it FTBFS

Source uploads don't help that much with that: the buildds will notice
for things that produce binary packages and all too often the break
occurs when some other package changes some time after the package was
uploaded.

This is one of the reasons why people do test builds of the archive from
time to time, particularly just before release.

-- 
"You grabbed my hand and we fell into it, like a daydream - or a fever."


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Re: Recompilation of ALL Debian packages ...

2006-09-04 Thread Michelle Konzack
Hello Martin and *,

Am 2006-08-31 17:11:03, schrieb martin f krafft:

> I would like to know why we can't just discard those binaries and
> rebuild them on trusted machines. Then we get the best of all
> worlds.

I have tried to RECOMPILE some packages in Sarge but failed.
The Binaries are working.  It seems, thet the Maintainer had
used a machine where the Build was successfull, but no other
one can do it because it FTBFS

I am pro source-only uploads, which mean, it will build on a
clean system... 

Thanks, Greetings and nice Day
Michelle Konzack
Systemadministrator
Tamay Dogan Network
Debian GNU/Linux Consultant


-- 
Linux-User #280138 with the Linux Counter, http://counter.li.org/
# Debian GNU/Linux Consultant #
Michelle Konzack   Apt. 917  ICQ #328449886
   50, rue de Soultz MSM LinuxMichi
0033/6/6192519367100 Strasbourg/France   IRC #Debian (irc.icq.com)


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Re: Recompilation of ALL Debian packages ...

2006-09-03 Thread martin f krafft
also sprach Bastian Blank <[EMAIL PROTECTED]> [2006.09.02.1841 +0200]:
> > Don't porters work on DSA-controlled machines?
> 
> Nope. They are controlled by the porters themself.

Then I guess this thread taught me something new. Not sure I wanted
to hear this.

-- 
Please do not send copies of list mail to me; I read the list!
 
 .''`.   martin f. krafft <[EMAIL PROTECTED]>
: :'  :  proud Debian developer, author, administrator, and user
`. `'`   http://people.debian.org/~madduck - http://debiansystem.info
  `-  Debian - when you have better things to do than fixing systems
 
NP: Eels / Blinking Lights and other Revelations (Disk 1)


signature.asc
Description: Digital signature (GPG/PGP)


Re: Recompilation of ALL Debian packages ...

2006-09-02 Thread Bastian Blank
On Sat, Sep 02, 2006 at 04:24:24PM +0200, martin f krafft wrote:
> Don't porters work on DSA-controlled machines?

Nope. They are controlled by the porters themself.

Bastian

-- 
Fascinating is a word I use for the unexpected.
-- Spock, "The Squire of Gothos", stardate 2124.5


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Re: Recompilation of ALL Debian packages ...

2006-09-02 Thread Ben Pfaff
martin f krafft <[EMAIL PROTECTED]> writes:

> The important thing to consider is that there are always two
> types of clients: executives and clued people. The clued people
> understand your reasoning (and I claim I do too, which makes me
> clued; woohoo!). The executives don't.

You seem to be saying that we should make technical decisions in
Debian based on what a bunch of unclued executives think.  I
disagree.
-- 
Ben Pfaff 
email: [EMAIL PROTECTED]
web: http://benpfaff.org


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Re: Recompilation of ALL Debian packages ...

2006-09-02 Thread martin f krafft
also sprach Henning Makholm <[EMAIL PROTECTED]> [2006.09.02.1552 +0200]:
> > And yes, I still think there's a difference between the two
> > scnearios: a clean source, 11 clean binaries, but one trojaned one
> > against an unclean source and 12 unclean binaries. As someone else
> > said, post-mortem it'll be *much* easier to deal with the latter.
> 
> You seem to be assuming that porters are more trustworthy than
> other DDs. Why?

Don't porters work on DSA-controlled machines? It's not so much
about trusting the one doing the work as it is about not trusting
the environment in which a package was built.

-- 
Please do not send copies of list mail to me; I read the list!
 
 .''`.   martin f. krafft <[EMAIL PROTECTED]>
: :'  :  proud Debian developer, author, administrator, and user
`. `'`   http://people.debian.org/~madduck - http://debiansystem.info
  `-  Debian - when you have better things to do than fixing systems
 
NP: Pond / Pond


signature.asc
Description: Digital signature (GPG/PGP)


Re: Recompilation of ALL Debian packages ...

2006-09-02 Thread Henning Makholm
Scripsit martin f krafft <[EMAIL PROTECTED]>

> And yes, I still think there's a difference between the two
> scnearios: a clean source, 11 clean binaries, but one trojaned one
> against an unclean source and 12 unclean binaries. As someone else
> said, post-mortem it'll be *much* easier to deal with the latter.

You seem to be assuming that porters are more trustworthy than
other DDs. Why?

-- 
Henning Makholm"Vi skal nok ikke begynde at undervise hinanden i
den store regnekunst her, men jeg vil foreslå, at vi fra
 Kulturministeriets side sørger for at fremsende tallene og også
  give en beskrivelse af, hvordan man læser tallene. Tak for i dag!"



Re: Recompilation of ALL Debian packages ...

2006-09-02 Thread Michael Poole
Russ Allbery writes:

> Source-code trojans are more dangerous because people fear binaries but
> think that if they've compiled it, it's fine, when the only real
> distinction is between code that's been audited and code that hasn't.
> Binaries built and uploaded by a maintainer who audits the upstream code
> are significantly safer than uncompiled source code uploaded by a
> maintainer who doesn't.

This compares apples (a maintainer who audits the upstream code) to
oranges (one who doesn't).  Even given human error, the approach to
auditing a source code package is reasonably well-understood.  For
binary packages, it is not, but it is clear that it is much more
labor-intensive.

Michael Poole


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Re: Recompilation of ALL Debian packages ...

2006-09-02 Thread Russ Allbery
martin f krafft <[EMAIL PROTECTED]> writes:

> And yes, I still think there's a difference between the two scnearios: a
> clean source, 11 clean binaries, but one trojaned one against an unclean
> source and 12 unclean binaries. As someone else said, post-mortem it'll
> be *much* easier to deal with the latter.

I've thought about this for a while now (you mentioned it earlier as well)
and I can't see why the latter would be easier to deal with.  I'm curious
what difference you're seeing.  Either way, you still have to verify that
the source is now clean; there's no reason to assume, once a trojan is
discovered, that there was only *one* trojan, and source trojans can be
written to only manifest on one particular platform.

Certainly, binaries are essentially impossible to audit, so as soon as you
want the security of an audit, you have to start with source.  But
blocking upload of binaries doesn't help with that process at all.  It
only would if the ftpmasters or buildd admins were then going to audit the
source, which of course they're not and couldn't given the millions of
lines of source in Debian.

I can construct several artificial scenarios where source-only uploads
would lead to better security (such as postulating the existence of roving
source code auditors who look at all the Debian source packages), but none
of them describe Debian today or seem particularly likely to describe a
future Debian.

The reason why people get uncomfortable around this area of Debian's
security is because they *should* be; what they may miss is that the same
issues apply to *all* software, and they should be equally nervous about
any software they download off the net, in any form.  The largest
mitigation of the risk is that most software comes with chains of trust,
breaks in those chains of trust usually have other symptoms and are
discovered through other methods, and as soon as someone finds a problem
the word spreads fairly fast.  It would surprise me a great deal if SuSE
were any better at auditing the code they incorporate into their
distribution than Debian is.

-- 
Russ Allbery ([EMAIL PROTECTED])   


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Re: Recompilation of ALL Debian packages ...

2006-09-02 Thread Russ Allbery
Sven Luther <[EMAIL PROTECTED]> writes:
> On Fri, Sep 01, 2006 at 11:52:17PM -0700, Russ Allbery wrote:

>> Source-code trojans are more dangerous because people fear binaries but
>> think that if they've compiled it, it's fine, when the only real
>> distinction is between code that's been audited and code that hasn't.
>> Binaries built and uploaded by a maintainer who audits the upstream
>> code are significantly safer than uncompiled source code uploaded by a
>> maintainer who doesn't.

> The thing is, if you have no guarantee that the binaries effectively
> correspond to the sources you are auditing, then auditing is not going
> to do you any kind of good, don't you think ?

Which just says that you shouldn't audit the source code and then use
someone else's binaries supposedly built from it.  Any organization
willing and capable of going to the work of auditing the source code can
certainly then also build their own binaries from exactly the source code
they audited, and should.  Debian upload policies don't help them, one way
or the other, since the work of building the package is negligible next to
the work required for the audit and in that sort of situation not building
the package themselves would just reintroduce more risks than are worth
it.

> I still believe that rebuilding everything is the best way to go, since
> it avoids any number of silly errors on the developer's part, and this
> would be a good thing for stability if nothing else.

Agreed.

-- 
Russ Allbery ([EMAIL PROTECTED])   


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Re: Recompilation of ALL Debian packages ...

2006-09-02 Thread Sven Luther
On Fri, Sep 01, 2006 at 11:52:17PM -0700, Russ Allbery wrote:
> martin f krafft <[EMAIL PROTECTED]> writes:
> > also sprach Russ Allbery <[EMAIL PROTECTED]> [2006.09.02.0141 +0200]:
> 
> >> I honestly think the security argument for doing this is silly.
> 
> > Clients do not want to hear something like that.
> 
> People frequently don't want to hear that ideas they've latched on to
> don't really have much basis in fact.  If I were expressing that directly
> with a client, I would probably use a softer expression of the idea than
> "silly," of course.  I would, however, not want to let someone keep the
> notion that binaries are dangerous but source code is somehow safer.  It's
> not true (at least in any significant sense), nor is it true that
> source-only uploads provide any more accountability than the system we
> have now.
> 
> Source-code trojans are more dangerous because people fear binaries but
> think that if they've compiled it, it's fine, when the only real
> distinction is between code that's been audited and code that hasn't.
> Binaries built and uploaded by a maintainer who audits the upstream code
> are significantly safer than uncompiled source code uploaded by a
> maintainer who doesn't.

The thing is, if you have no guarantee that the binaries effectively
correspond to the sources you are auditing, then auditing is not going to do
you any kind of good, don't you think ?

I still believe that rebuilding everything is the best way to go, since it
avoids any number of silly errors on the developer's part, and this would be a
good thing for stability if nothing else.

Also, imagine a statically linked binary, which happens to have been built
with an non-official library on the developper machine, or with devel X
libs,or whatever else. I guess all those already happened in the past.

Not to count the few packages which are not buildable out of main, but need
some extra non-official packages or manipulation to bootstrap (like the
gcc/glibc pair for example).

Friendly,

Sven Luther


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Re: Recompilation of ALL Debian packages ...

2006-09-02 Thread martin f krafft
also sprach Russ Allbery <[EMAIL PROTECTED]> [2006.09.02.0912 +0200]:
> Feh, I think that's a cop-out.  It's not that hard to explain, or
> that hard to understand, and I've worked with plenty of executives
> who can understand that concept just fine when explained in terms
> that they're familiar with.

So have I. You're welcome to try these two clients of mine! :)
(and they are huge clients as well as potential sponsors)

> Besides, one of the major reasons why I work on free software
> projects like Debian is precisely because I have no intention of
> considering the opinions of those ignorant of the issues involved
> when determining questions of technical policy.

The technicians at the place think the same. It's them and I against
the executives. I like free software because people help each other
readily. So that's my job: to help them, because they are sick and
tired of using Suse.

And yes, I still think there's a difference between the two
scnearios: a clean source, 11 clean binaries, but one trojaned one
against an unclean source and 12 unclean binaries. As someone else
said, post-mortem it'll be *much* easier to deal with the latter.

-- 
Please do not send copies of list mail to me; I read the list!
 
 .''`.   martin f. krafft <[EMAIL PROTECTED]>
: :'  :  proud Debian developer, author, administrator, and user
`. `'`   http://people.debian.org/~madduck - http://debiansystem.info
  `-  Debian - when you have better things to do than fixing systems
 
there are two groups of people in the world: those who believe that
the world can be divided into two groups of people, and those who
don't.


signature.asc
Description: Digital signature (GPG/PGP)


Re: Recompilation of ALL Debian packages ...

2006-09-02 Thread Russ Allbery
martin f krafft <[EMAIL PROTECTED]> writes:
> also sprach Russ Allbery <[EMAIL PROTECTED]> [2006.09.02.0852 +0200]:

>> You're probably not going to convince me on this, so it may not be
>> worth wasting time on arguing about it when we both agree on the
>> fundamental goal.

> Neither have you convinced me. The important thing to consider is that
> there are always two types of clients: executives and clued people. The
> clued people understand your reasoning (and I claim I do too, which
> makes me clued; woohoo!). The executives don't. And it's easier to teach
> Go or Bridge to a group of monkeys than it is to hammer sense into
> executives when it comes to technical stuff.

Feh, I think that's a cop-out.  It's not that hard to explain, or that
hard to understand, and I've worked with plenty of executives who can
understand that concept just fine when explained in terms that they're
familiar with.  Besides, one of the major reasons why I work on free
software projects like Debian is precisely because I have no intention of
considering the opinions of those ignorant of the issues involved when
determining questions of technical policy.

The idea that source code can be trojaned just as easily as binaries is a
very old one, even apart from impressive demonstrations involving C
compilers.  Consider this from twelve years ago, for instance:



I can explain that to an executive just fine.

-- 
Russ Allbery ([EMAIL PROTECTED])   


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Re: Recompilation of ALL Debian packages ...

2006-09-02 Thread martin f krafft
also sprach Russ Allbery <[EMAIL PROTECTED]> [2006.09.02.0852 +0200]:
> You're probably not going to convince me on this, so it may not be
> worth wasting time on arguing about it when we both agree on the
> fundamental goal.

Neither have you convinced me. The important thing to consider is
that there are always two types of clients: executives and clued
people. The clued people understand your reasoning (and I claim I do
too, which makes me clued; woohoo!). The executives don't. And it's
easier to teach Go or Bridge to a group of monkeys than it is to
hammer sense into executives when it comes to technical stuff.

-- 
Please do not send copies of list mail to me; I read the list!
 
 .''`.   martin f. krafft <[EMAIL PROTECTED]>
: :'  :  proud Debian developer, author, administrator, and user
`. `'`   http://people.debian.org/~madduck - http://debiansystem.info
  `-  Debian - when you have better things to do than fixing systems
 
"geld ist das brecheisen der macht."
 - friedrich nietzsche


signature.asc
Description: Digital signature (GPG/PGP)


Re: Recompilation of ALL Debian packages ...

2006-09-01 Thread Russ Allbery
George Danchev <[EMAIL PROTECTED]> writes:

> True, and Martin's reasoning is about consistency across the
> architectures, not that much after security, as I read it.

That argument I agree with.

> On Saturday 02 September 2006 02:41, Russ Allbery wrote:
>> However, that does not mean I think it's a bad idea.  I actually think
>> it's a good idea, but for a somewhat different reason.  Every single
>> time we get ready to release stable, someone builds every package in
>> the distribution and then encounters a bunch of FTBFS errors,
>> particularly for arch: all packages.  Many of those errors were always
>> there and were never detected because we don't build arch: all packages
>> anywhere outside the maintainer's system.

> Fortunately there are lots of people running personal autobuilders and 
> reporting FTBFS's lately, even in the arch:all packages.

Sure, because we're approaching a release.  My impression is that this
activity ramps up considerably before a release, which adds to the problem
of growing RC bug counts right when we're trying to shrink them.  The
solution is to find the problems earlier and in a way that breaks
something the maintainer immediately cares about.  If packages couldn't
get into the archive unless they built on at least one autobuilder, I
think that would cut down on a lot of RC bugs that are only found later
and get maintainers to be more aggressive about fixing them immediately.

-- 
Russ Allbery ([EMAIL PROTECTED])   


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Re: Recompilation of ALL Debian packages ...

2006-09-01 Thread Russ Allbery
martin f krafft <[EMAIL PROTECTED]> writes:
> also sprach Russ Allbery <[EMAIL PROTECTED]> [2006.09.02.0141 +0200]:

>> I honestly think the security argument for doing this is silly.

> Clients do not want to hear something like that.

People frequently don't want to hear that ideas they've latched on to
don't really have much basis in fact.  If I were expressing that directly
with a client, I would probably use a softer expression of the idea than
"silly," of course.  I would, however, not want to let someone keep the
notion that binaries are dangerous but source code is somehow safer.  It's
not true (at least in any significant sense), nor is it true that
source-only uploads provide any more accountability than the system we
have now.

Source-code trojans are more dangerous because people fear binaries but
think that if they've compiled it, it's fine, when the only real
distinction is between code that's been audited and code that hasn't.
Binaries built and uploaded by a maintainer who audits the upstream code
are significantly safer than uncompiled source code uploaded by a
maintainer who doesn't.

You're probably not going to convince me on this, so it may not be worth
wasting time on arguing about it when we both agree on the fundamental
goal.

-- 
Russ Allbery ([EMAIL PROTECTED])   


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Re: Recompilation of ALL Debian packages ...

2006-09-01 Thread George Danchev
On Saturday 02 September 2006 02:41, Russ Allbery wrote:
> martin f krafft <[EMAIL PROTECTED]> writes:
> > The reason I am pushing for this is because of two of my clients, who
> > have been wanting to use Debian for three years now but consciously
> > decided against it, because it is not guaranteed that the sources and
> > the binaries in our archives correspond for all architectures. They are
> > well aware that trojans can still exist, but it's an entirely different
> > thing whether they exist in source and hence in all architectures (which
> > would result in some serious negative feedback or even revocation of
> > upload rights), or just in one of the binaries and hence would be much
> > harder to detect/analyse.
>
> I honestly think the security argument for doing this is silly.

True, and Martin's reasoning is about consistency across the architectures, 
not that much after security, as I read it.

> However, that does not mean I think it's a bad idea.  I actually think
> it's a good idea, but for a somewhat different reason.  Every single time
> we get ready to release stable, someone builds every package in the
> distribution and then encounters a bunch of FTBFS errors, particularly for
> arch: all packages.  Many of those errors were always there and were never
> detected because we don't build arch: all packages anywhere outside the
> maintainer's system.  

Fortunately there are lots of people running personal autobuilders and 
reporting FTBFS's lately, even in the arch:all packages.

-- 
pub 4096R/0E4BD0AB 2003-03-18 
fingerprint 1AE7 7C66 0A26 5BFF DF22 5D55 1C57 0C89 0E4B D0AB 


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Re: Recompilation of ALL Debian packages ...

2006-09-01 Thread martin f krafft
also sprach Russ Allbery <[EMAIL PROTECTED]> [2006.09.02.0141 +0200]:
> I honestly think the security argument for doing this is silly.

Clients do not want to hear something like that.

-- 
Please do not send copies of list mail to me; I read the list!
 
 .''`.   martin f. krafft <[EMAIL PROTECTED]>
: :'  :  proud Debian developer, author, administrator, and user
`. `'`   http://people.debian.org/~madduck - http://debiansystem.info
  `-  Debian - when you have better things to do than fixing systems
 
what do you mean, it's not packaged in debian?


signature.asc
Description: Digital signature (GPG/PGP)


Re: Recompilation of ALL Debian packages ...

2006-09-01 Thread Russ Allbery
martin f krafft <[EMAIL PROTECTED]> writes:

> The reason I am pushing for this is because of two of my clients, who
> have been wanting to use Debian for three years now but consciously
> decided against it, because it is not guaranteed that the sources and
> the binaries in our archives correspond for all architectures. They are
> well aware that trojans can still exist, but it's an entirely different
> thing whether they exist in source and hence in all architectures (which
> would result in some serious negative feedback or even revocation of
> upload rights), or just in one of the binaries and hence would be much
> harder to detect/analyse.

I honestly think the security argument for doing this is silly.

However, that does not mean I think it's a bad idea.  I actually think
it's a good idea, but for a somewhat different reason.  Every single time
we get ready to release stable, someone builds every package in the
distribution and then encounters a bunch of FTBFS errors, particularly for
arch: all packages.  Many of those errors were always there and were never
detected because we don't build arch: all packages anywhere outside the
maintainer's system.  Similarly, there have been packages in the archive
with significantly different configured features and library dependencies
on x86 than on any other platform because of where the maintainer built
the package.

So I'm not disagreeing with the goal.  I just don't like the security
argument for it and don't find it persuasive.  But I would vote in favor
of building all *.debs on central build servers.  (Whether we still
require a *.deb during upload is actually a separate question -- I think
there's an argument, perhaps not persuasive, in favor of requiring that
the upload contain built packages for at least one platform as a basic
sanity check but just throwing away that build after verifying it exists.)

-- 
Russ Allbery ([EMAIL PROTECTED])   


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Re: Recompilation of ALL Debian packages ...

2006-09-01 Thread martin f krafft
also sprach Russ Allbery <[EMAIL PROTECTED]> [2006.09.01.0241 +0200]:
> Rebuilding every package really doesn't buy you that much in the
> way of security.

This is arguable and I don't want to go there. The reason I am
pushing for this is because of two of my clients, who have been
wanting to use Debian for three years now but consciously decided
against it, because it is not guaranteed that the sources and the
binaries in our archives correspond for all architectures. They are
well aware that trojans can still exist, but it's an entirely
different thing whether they exist in source and hence in all
architectures (which would result in some serious negative feedback
or even revocation of upload rights), or just in one of the binaries
and hence would be much harder to detect/analyse.

> It makes it harder to hide what you did, but only harder; a rogue
> uploader could obfuscate a trojan in source code rather well.  In
> the end, we still trust people in the keyring.

We do. Does that mean our clients do? Does it mean our clients have
to trust their machines?

I realise that on an academic level you are absolutely right, and
our users effectively trust every machine in use by developers.

However, security is not about secure vs. insecure, it's about
building blocks, and the harder we make it, the better. Every single
step counts, as long as its doable with reasonable effort.

> About the only thing you gain is the potential ability to do more
> detailed post-mortem analysis after something already exploded.

Accountability is very important to businesses.

We could argue endlessly about this, but I'd much rather move
forward. You say that it won't buy much, but you don't voice
a concern or vote against it. Therefore let's see how much effort it
would be and then assess whether it's a viable means forward to
recompile on all architectures.

-- 
Please do not send copies of list mail to me; I read the list!
 
 .''`. martin f. krafft <[EMAIL PROTECTED]>
: :'  :proud Debian developer, author, administrator, and user
`. `'` http://people.debian.org/~madduck http://debiansystem.info
  `-  Debian - when you have better things to do than fixing systems
 
whatever you do will be insignificant,
but it is very important that you do it.
 -- mahatma gandhi


signature.asc
Description: Digital signature (GPG/PGP)


Re: Recompilation of ALL Debian packages ...

2006-09-01 Thread Roberto C. Sanchez
On Fri, Sep 01, 2006 at 02:57:27AM +0200, Sven Luther wrote:
> On Thu, Aug 31, 2006 at 05:41:11PM -0700, Russ Allbery wrote:
> > 
> > Rebuilding every package really doesn't buy you that much in the way of
> > security.  It makes it harder to hide what you did, but only harder; a
> > rogue uploader could obfuscate a trojan in source code rather well.  In
> > the end, we still trust people in the keyring.  About the only thing you
> > gain is the potential ability to do more detailed post-mortem analysis
> > after something already exploded.
> 
> And the amount of breakage caused by actual mistakes of the uploader, like
> having random sets of non-official libraries installed and such.
> 

Is it not part of the process of becoming a DD (or sponsorship of
packages for non-DDs) learning the "responsible" way to build packages.
That is, developers are taught to use tools like pbuilder or sbuild in
order to ensure that packages build cleanly.  I'm not saying that
mistakes will never occur.  However, I would think that the vast
majority of the people will be responsible and do it correctly the vast
majority of the time.

Regards,

-Roberto

-- 
Roberto C. Sanchez
http://familiasanchez.net/~roberto


signature.asc
Description: Digital signature


Re: Recompilation of ALL Debian packages ...

2006-08-31 Thread Sven Luther
On Thu, Aug 31, 2006 at 05:41:11PM -0700, Russ Allbery wrote:
> Matej Cepl <[EMAIL PROTECTED]> writes:
> 
> > No, it is matter of accountability and being able to tell to the bank
> > (mentioned in Martin's presentation) that we know who compiled the
> > package and we have made reasonable precautions to be sure there are no
> > trojans inside.
> 
> Rebuilding every package really doesn't buy you that much in the way of
> security.  It makes it harder to hide what you did, but only harder; a
> rogue uploader could obfuscate a trojan in source code rather well.  In
> the end, we still trust people in the keyring.  About the only thing you
> gain is the potential ability to do more detailed post-mortem analysis
> after something already exploded.

And the amount of breakage caused by actual mistakes of the uploader, like
having random sets of non-official libraries installed and such.

Friendly,

Sven Luther


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Re: Recompilation of ALL Debian packages ...

2006-08-31 Thread Russ Allbery
Matej Cepl <[EMAIL PROTECTED]> writes:

> No, it is matter of accountability and being able to tell to the bank
> (mentioned in Martin's presentation) that we know who compiled the
> package and we have made reasonable precautions to be sure there are no
> trojans inside.

Rebuilding every package really doesn't buy you that much in the way of
security.  It makes it harder to hide what you did, but only harder; a
rogue uploader could obfuscate a trojan in source code rather well.  In
the end, we still trust people in the keyring.  About the only thing you
gain is the potential ability to do more detailed post-mortem analysis
after something already exploded.

-- 
Russ Allbery ([EMAIL PROTECTED])   


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Re: Recompilation of ALL Debian packages ...

2006-08-31 Thread Matej Cepl
On Thu 31. August 2006 12:47, you wrote:
> Debian is not other distributions. Other distributions have
> dependency hell with source-only uploads. This is a matter of
> policy and being able to blame people if something fails.

No, it is matter of accountability and being able to tell to the 
bank (mentioned in Martin's presentation) that we know who 
compiled the package and we have made reasonable precautions to 
be sure there are no trojans inside. Checking for dependencies 
was meant to be just cherry on the top -- probably it isn't, but 
it wasn't something which I would be that much worried about.

> Without a binary version someone upload (and therefor should
> have tested), he could always claim his upload would have
> worked if the buildds would not have mangled it. So there is
> at least one package everyone can check who is to blame for
> it.

You are right, I wrote source-only upload, but obviously 
upload-binary-and-remove-it is better policy.

> well, besides showing your ignorance for how debian packages

I have never claimed that I am not ignorant -- just curious and 
obviously by madduck's answers I am not that much off the point.

> it by the programs processing the .changes files), I think
> this would be one of the most stupid things to do. Take a look
> at ubuntu, who do this, and how many utterly broken packages
> they get out of this.

Chmmm, either you are smart and wise and Ubuntu people (among 
others) are stupid, or you are trying to bully me out of the 
discussion on petty technical details and you are over-reacting 
in order to avoid the technical discussion. Guess, which option 
I think is more probable? :-)

Best,

Matěj
P.S.: I think I would rather keep the discussion on the list, if 
you won't mind.
-- 
GPG Finger: 89EF 4BC6 288A BF43 1BAB  25C3 E09F EF25 D964 84AC
http://www.ceplovi.cz/matej/blog/, Jabber: [EMAIL PROTECTED]
23 Marion St. #3, (617) 876-1259, ICQ 132822213
 
Just remember, brothers and sisters--their skins may be white,
but their souls are just as black as ours!
   -- a black preacher



Re: Recompilation of ALL Debian packages ...

2006-08-31 Thread Sven Luther
On Thu, Aug 31, 2006 at 05:03:10PM +0200, martin f krafft wrote:
> [bcc'd to ftpmaster to make it easier for them to reply if they
> don't read the list]
> 
> also sprach Matej Cepl <[EMAIL PROTECTED]> [2006.08.31.1621 +0200]:
> > Wouldn't it be sensible to add that line to crontab (e.g., rm -f 
> > $INCOMING_QUEUE/*.deb; we have even advantage over Red Hat, that 
> 
> I don't think it's as simple as that; we definitely need dak
> support, because for all I know, dak requires at least one .deb to
> accept a source package.

Some 4-5 years ago, source-only uploads where definitively working, and it was
really nice to do those, when i was behind a slow modem line, and the
binary/source ratio was very big.

It was disabled because some unresponsible guys used to upload
not-even-checked source packages i believe.

Friendly,

Sven Luther


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Re: Recompilation of ALL Debian packages ...

2006-08-31 Thread Sven Luther
On Thu, Aug 31, 2006 at 10:21:06AM -0400, Matej Cepl wrote:
> Hi,
> 
> I was listening to madduck's presentation for Irish LUG 
> (http://blog.signal2noise.co.uk/cgi-bin/blosxom.pl\
> /technical/martinfkrafft_talk.html) and I was quite shocked to 
> learn, that not all binary packages are compiled through buildd 
> network, but that most binary packages (mostly those created on 
> i386 platform) are uploaded directly by DD.
> 
> In bad old days when I was using RedHat I dimly remember that 
> contributions to contrib/ directory at RedHat.com were allowed 
> to be only as .src.rpm packages and 386.rpm were immediately 
> deleted without RH system even thinking about them (through 
> crontab). Why in the world is this not done Debian is besides me 
> (actually, I thought, it IS done) -- I don't think there would 
> be not enough buildd machines with i386, if Debian people would 
> try to find them.

Accordying to James Troup, whom i asked exactly that at some past debconf,
this is because if there is some lag in the x86 buildd, then loads of user
will complain to them about non-installable packages, and the
ftp-masters/buildd administrators being volunteers and not paid and everything, 
don't care about being given such pressure.

Not to mention that this is the case for almost all non-x86 arches, and the
fact that elmo does exactly the same thing when working for ubuntu.

Notice also that source-only uploads where allowed in some distant past (some
4-5 years ago), but explicitly disabled.

Friendly,

Sven Luther


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Re: Recompilation of ALL Debian packages ...

2006-08-31 Thread martin f krafft
also sprach Henning Makholm <[EMAIL PROTECTED]> [2006.08.31.1641 +0200]:
> Please read up on the regular (every few months) discussions about
> "source-only uploads" in the list archives. (Capsule summary: yes,
> it would be easy to do, but there is no consensus that it would be
> *desirable* to do so).

For instance: http://lists.debian.org/debian-devel/2006/07/msg00544.html

You say people have argued against source-only uploads, but I think
source-only is actually not what we are talking about. The subject
line says: "recompile...".

Whether or not requiring binary packages in the upload as a QA sort
of measure is a good thing is a tangential debate.

I would like to know why we can't just discard those binaries and
rebuild them on trusted machines. Then we get the best of all
worlds.

(Arguably I failed to express myself properly in the last post).

-- 
Please do not send copies of list mail to me; I read the list!
 
 .''`. martin f. krafft <[EMAIL PROTECTED]>
: :'  :proud Debian developer, author, administrator, and user
`. `'` http://people.debian.org/~madduck http://debiansystem.info
  `-  Debian - when you have better things to do than fixing systems
 
"i feel sorry for people who don't drink. when they wake up in the
 morning, that's as good as they're going to feel all day."
-- frank sinatra


signature.asc
Description: Digital signature (GPG/PGP)


Re: Recompilation of ALL Debian packages ...

2006-08-31 Thread martin f krafft
[bcc'd to ftpmaster to make it easier for them to reply if they
don't read the list]

also sprach Matej Cepl <[EMAIL PROTECTED]> [2006.08.31.1621 +0200]:
> Wouldn't it be sensible to add that line to crontab (e.g., rm -f 
> $INCOMING_QUEUE/*.deb; we have even advantage over Red Hat, that 

I don't think it's as simple as that; we definitely need dak
support, because for all I know, dak requires at least one .deb to
accept a source package.

But maybe one of the people with more insight into the matter could
give a brief overview of what would be needed to support this.

Or give the reasons why it hasn't been done yet. I surely don't want
to raise hell (again) why our ftpmasters aren't doing anything, when
they have good reason not to. :)

-- 
Please do not send copies of list mail to me; I read the list!
 
 .''`. martin f. krafft <[EMAIL PROTECTED]>
: :'  :proud Debian developer and author: http://debiansystem.info
`. `'`
  `-  Debian - when you have better things to do than fixing a system
 
"a woman does not want the truth; what is truth to women? from the
 beginning, nothing has been more alien, repugnant, and hostile to
 woman than the truth - her great art is the lie, her highest
 concern is mere appearance and beauty."
  -- friedrich nietzsche


signature.asc
Description: Digital signature (GPG/PGP)


Re: Recompilation of ALL Debian packages ...

2006-08-31 Thread Henning Makholm
Scripsit Matej Cepl <[EMAIL PROTECTED]>

> Wouldn't it be sensible to add that line to crontab (e.g., rm -f 
> $INCOMING_QUEUE/*.deb; we have even advantage over Red Hat, that 
> we don't have to fiddle with find to delete just binary *.rpm 
> and preserve *.src.rpm :-)) and to recompile everything?

Please read up on the regular (every few months) discussions about
"source-only uploads" in the list archives. (Capsule summary:
yes, it would be easy to do, but there is no consensus that it
would be *desirable* to do so).

-- 
Henning Makholm"They have a word for people our age.
  They call us children and treat us like mice."


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



Recompilation of ALL Debian packages ...

2006-08-31 Thread Matej Cepl
Hi,

I was listening to madduck's presentation for Irish LUG 
(http://blog.signal2noise.co.uk/cgi-bin/blosxom.pl\
/technical/martinfkrafft_talk.html) and I was quite shocked to 
learn, that not all binary packages are compiled through buildd 
network, but that most binary packages (mostly those created on 
i386 platform) are uploaded directly by DD.

In bad old days when I was using RedHat I dimly remember that 
contributions to contrib/ directory at RedHat.com were allowed 
to be only as .src.rpm packages and 386.rpm were immediately 
deleted without RH system even thinking about them (through 
crontab). Why in the world is this not done Debian is besides me 
(actually, I thought, it IS done) -- I don't think there would 
be not enough buildd machines with i386, if Debian people would 
try to find them.

Madduck pointed me towards older discussion on this issue 
(http://www.gatago.com/linux/debian/security/14510447.html) from 
May 2004 [!], which is really strange -- everybody agrees that 
it is real security issue, that it wouldn't be that difficult to 
resolve, but if I am not mistaken, nothing happened so far.

Another comment from that discussion made even more thinking 
about this -- somebody mentioned, that of course many DDs are 
not compiling their packages in clean chroot. I have always 
thought that beauty of the fact (which I believed to be true 
then) that everything is recompiled was that all Dependencies 
are really tried in the hard way -- in clean chroot. Apparently 
this is not the case, and Debian is not that much different than 
some more adventurous distributions where dependency hell may 
happen.

Wouldn't it be sensible to add that line to crontab (e.g., rm -f 
$INCOMING_QUEUE/*.deb; we have even advantage over Red Hat, that 
we don't have to fiddle with find to delete just binary *.rpm 
and preserve *.src.rpm :-)) and to recompile everything?

Best,

Matěj

-- 
GPG Finger: 89EF 4BC6 288A BF43 1BAB  25C3 E09F EF25 D964 84AC
http://www.ceplovi.cz/matej/blog/, Jabber: [EMAIL PROTECTED]
23 Marion St. #3, (617) 876-1259, ICQ 132822213
 
He is a self-made man and worships his creator.
  -- John Bright



Re: Pacotes Debian / Packages MD5 sum

2005-11-28 Thread A F Machado
English: 

Hello, Rubem
At every directory containing files for download (in Debian mirrors)
there are MD5SUMS files containing lists of md5 for each image.
Try to download using jigdo sw or bit torrent (to save bandwidth of
servers). 
This list language is English.
It is very likely that you will get faster answers if you ask technical
or usage questions at one of the suitable user-lists at your native
language http://lists.debian.org/users.html  .
This debian-project list is for discussing non-technical issues of the
project organization.
Good luck and welcome to Debian.
Andre Felipe

Português:

Olá, Rubem
Em cada diretório contendo arquivos para download (nos espelhos Debian)
http://www.debian.org.br/CD/ há arquivos MD5SUMS contendo listas de md5
para cada imagem.
Tente baixar usando o software jigdo ou o bit torrent (para poupar banda
de transmissão nos servidores).
O idioma desta lista aqui é o inglês.
É bem mais provável que você consiga respostas mais rápido e profundas
se você fizer perguntas técnicas ou sobre uso em alguma das listas de
usuários adequadas, em seu idioma nativo em
http://lists.debian.org/users.html .
Esta lista debian-project é para debater aspectos não técnicos da
organização do projeto.
Boa sorte e bem vindo ao Debian.
André Felipe




Em Seg, 2005-11-28 às 14:54 +, Rubem Figueredo escreveu:
> Prezados,
> os pacotes iso do Debian contidos no site www.debian.org possuem md5 ?
> precisava ver se os pacotes "baixados" vieram corrompidos ou não.
> Grato,
> Rubem
> 
> 
> __
> Lar doce lar. Faça do Yahoo! sua homepage.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]



  1   2   >