Re: Question to all candidates: Ongoing/future legal projects

[Paul Wise]

On Tue, 2022-03-29 at 20:47 -0700, Felix Lechner wrote:

> It furthermore seems that I did not follow the proper process when
> filing my request, as Paul Wise pointed out.

My reference to the Hardware/Wanted wiki page was referring to the
procedure for after you have bought hardware, no longer need it and
want to pass it on to someone else.

Admittedly the page also includes info about posting hardware
wishlists, but that section of the page doesn't really apply to your
case since it is more for a Debian service, so just buying hardware and
getting a reimbursement is a better process, this is documented in the
section on hosting, which I've just rewritten to be a bit clearer.

PS: I also just added mention of hardware purchases to the page and
uses of Debian money to the MemberBenefits page.

-- 
bye,
pabs

https://wiki.debian.org/PaulWise


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

Re: Question to all candidates: Ongoing/future legal projects

[Felix Lechner]

Hi Tiago,

On Sun, Mar 27, 2022 at 11:09 AM Tiago Bortoletto Vaz  wrote:
>
> Given that Jonathan, after lots of research as he describes in this
> thread, has stated that such request never reached him, can you clarify
> how can you have even complied to a request for more information from him?
>
> Also, if that's the case, why didn't you try to reach him in private at
> the time rather than bringing it to -vote months later, during an DPL
> campaign period in which you are both candidates?
>
> In my view, your request was totally eligible and could/should be
> quickly approved.

Look, the discussion with Richard was about the process for
disbursements. I like the idea of committees that are on schedule and
open to the public.

Despite the collateral damage Jonathan suffered for the missing
reimbursement (which people in power have to endure) I did not intend
to embarrass him, yet that's where we are taking this thread.

Jonathan is a busy guy. In fact, he is so busy he too would benefit if
other folks handled the disbursements. It would be a win-win for
everyone.

My messages may also have gotten stuck in Jonathan's spam filter.

It furthermore seems that I did not follow the proper process when
filing my request, as Paul Wise pointed out.

Either way, you challenged me for the true record. Below, you will
find the exchange you are interested in. I redacted both of Jonathan's
responses in case he wrote them. As you can see, I wrote a lot in
private, but was ineffective.

Similar to my other responses, I am committed to transparency when possible.

For context you will need to know that my internal hindrances with
operating lintian.d.o—which we resurrected after years of spotty
service—did not start (or end) with the misplaced reimbursement. In
RT#8464 from November 2020, you will find a partial record of the
dispute. (The ticket contains one of the few irate emails I have
written in Debian; it may have contributed to my DAM warning.) To this
day, there has been nothing but obstruction.

In my mind, the missing reimbursement simply made that point one more time.

Jonathan witnessed some of my issues, but he did not cause them. In
Debian, too much power is held away from the public eye. We have Setec
Astronomy. Hence my open and public committees.

Thanks for supporting my reimbursement request!

Kind regards,
Felix Lechner

-- Forwarded message -
From: Felix Lechner 
Date: Wed, Nov 3, 2021 at 7:45 AM
Subject: Re: Spending Debian money, [redacted]
To: Jonathan Carter 

Hi Jonathan,

On Sun, Aug 9, 2020 at 9:01 AM Jonathan Carter  wrote:
>
> [excerpt of posting to debian-private]

For over a year, I contemplated writing this request. As part of my
contributions to Debian I operate the website lintian.d.o. Untarring
and scanning the archive uses lots of resources. The service is
heavily disk-bound.

It would help to upgrade one of my machines to an NVMe SSD. The
machine also needs more memory (currently 24 GB). I would not normally
make the upgrades. Would the project please help out with the proposed
purchase?

The details in the amount of approximately US$217 are below. Thanks!

Kind regards
Felix Lechner

* * *

SAMSUNG (MZ-V8V1T0B/AM) 980 SSD 1TB, $120
Dual M.2 PCIE Adapter for SATA or PCIE NVMe, $16
Patriot 16GB(2x8GB) Viper III DDR3 1600MHz CL9, $60
Subtotal $196
Sales Tax $21
Total $217

All amounts are in US dollars. The seller is Amazon.com.

-- Forwarded message -
From: Jonathan Carter 
Date: Thu, Nov 4, 2021 at 11:57 AM
Subject: Re: Spending Debian money, [redacted]
To: Felix Lechner 

[redacted for privacy]

-Jonathan

-- Forwarded message -
From: Jonathan Carter 
Date: Fri, Nov 5, 2021 at 11:53 AM
Subject: Re: Spending Debian money, [redacted]
To: Felix Lechner 

Hi Felix

[redacted for privacy]

-Jonathan

-- Forwarded message -
From: Felix Lechner 
Date: Fri, Nov 5, 2021 at 12:59 PM
Subject: Re: Spending Debian money, [redacted]
To: Jonathan Carter 

Hi Jonathan,

On Fri, Nov 5, 2021 at 11:53 AM Jonathan Carter  wrote:
>
> [redacted for privacy]

The website consists of three parts. There is a web server, which
[redacted] hosts presently; a database on my personal VPS; and the
machinery that generates the data.

Some time ago, [redacted] offered to host the database (part
[redacted]). I believe I met all their requirements (most notably,
packaging semver [1] for bullseye), but no one has responded to my
most recent message from October 4.

At this point, I believe I earned the right to suspect that the delay
was intended to provoke me into writing another angry letter, but none
is coming. It would therefore be fine to proceed with the database.
(The website will also work better if DAM kicks me out, as [redacted]
suggested.) The database is currently hosted on a VPS with 1 GB of
RAM, which is under dimensioned. I already offer reduced services.

Having received no response from [redacted], I approached [redacted]
about the database

Re: Results for Voting secrecy

[Kurt Roeckx]

On Tue, Mar 29, 2022 at 03:54:28PM -0700, Russ Allbery wrote:
> 
> My understanding of the implications of this process (and Kurt is
> authoritative here, of course) is that if you rank NOTA equally with an
> option, that vote is not part of V(A,D) or V(D,A) since neither option is
> preferred over the other, and therefore has no effect either way on
> whether an option is discarded because it doesn't meet majority.

That is also my understanding. Ranking an option equal to NOTA means
that it will have no effect on the majority requirements, only other
people's votes will have an effect on the majority requirements for that
option.

In the vote results you'll see that the majority calculation uses
185/61 for option 2. That means 185 people voted option 2 above NOTA,
61 below NOTA. Since there were 258 people who voted it means
258-185-61=12 voted option 2 equal to NOTA.


Kurt

Re: Results for Voting secrecy

[Philip Hands]

Russ Allbery  writes:

> Philip Hands  writes:
>
>> The blurb that's sent out with the votes says:
>
>>   To vote "no, no matter what", rank "None of the above" as more
>>   desirable than the unacceptable choices, or you may rank the "None of
>>   the above" choice and leave choices you consider unacceptable blank.
>
>> which to me suggests that if one ranks something as equal to NotA then
>> one is not marking it as unacceptable, so presumably it is counted as
>> acceptable -- is that how such votes are calculated?
>
> The relevant provision of the constitution is:
>
> A.5.3. Any (non-default) option which does not defeat the default
> option by its required majority ratio is dropped from consideration. 
>
>   1. Given two options A and B, V(A,B) is the number of voters who
>  prefer option A over option B.
>   2. An option A defeats the default option D by a majority ratio N,
>  if V(A,D) is greater or equal to N * V(D,A) and V(A,D) is
>  strictly greater than V(D,A).
>   3. If a supermajority of S:1 is required for A, its majority ratio
>  is S; otherwise, its majority ratio is 1.
>
> My understanding of the implications of this process (and Kurt is
> authoritative here, of course) is that if you rank NOTA equally with an
> option, that vote is not part of V(A,D) or V(D,A) since neither option is
> preferred over the other, and therefore has no effect either way on
> whether an option is discarded because it doesn't meet majority.

Ah, right -- it seems that I had a typo in the grep pattern I was
pointing at the tally file, which was confusing me. Having fixed that,
I can now see that it is exactly as you describe.  Thanks :-)

In that case, a vote of '--1-' does make sense, if one doesn't actually
want to block a sufficiently large vote for options 1 or 2. In fact it's
probably what I should have voted rather than '--12' in order to reflect
my actual view.

Cheers, Phil.
-- 
|)|  Philip Hands  [+44 (0)20 8530 9560]  HANDS.COM Ltd.
|-|  http://www.hands.com/http://ftp.uk.debian.org/
|(|  Hugo-Klemm-Strasse 34,   21075 Hamburg,GERMANY


signature.asc
Description: PGP signature

Re: To all candidates: Debian and people with disabilities

[Samuel Thibault]

Hello,

Wouter Verhelst, le mar. 29 mars 2022 20:00:19 +0200, a ecrit:
> On Tue, Mar 22, 2022 at 09:23:38AM +0100, Samuel Thibault wrote:
> > Hello,
> > 
> > Devin Prater, le lun. 21 mars 2022 22:10:15 -0500, a ecrit:
> > > As far as backports, my problem is enabling it. Normal desktop users 
> > > probably
> > > won't even know what that is, and the syntax is rather ugly, to me at 
> > > least.
> > 
> > Ok, that's one point that could be worked on: creating an easy way to
> > enable backports.
> 
> This is what my project, "extrepo", wants to accomplish: to make it
> easily possible to enable repositories that are not enabled by default.

Nice!

I added a not on the accessibility wiki page.

Samuel

Re: Results for Voting secrecy

[Russ Allbery]

Philip Hands  writes:

> The blurb that's sent out with the votes says:

>   To vote "no, no matter what", rank "None of the above" as more
>   desirable than the unacceptable choices, or you may rank the "None of
>   the above" choice and leave choices you consider unacceptable blank.

> which to me suggests that if one ranks something as equal to NotA then
> one is not marking it as unacceptable, so presumably it is counted as
> acceptable -- is that how such votes are calculated?

The relevant provision of the constitution is:

A.5.3. Any (non-default) option which does not defeat the default
option by its required majority ratio is dropped from consideration. 

  1. Given two options A and B, V(A,B) is the number of voters who
 prefer option A over option B.
  2. An option A defeats the default option D by a majority ratio N,
 if V(A,D) is greater or equal to N * V(D,A) and V(A,D) is
 strictly greater than V(D,A).
  3. If a supermajority of S:1 is required for A, its majority ratio
 is S; otherwise, its majority ratio is 1.

My understanding of the implications of this process (and Kurt is
authoritative here, of course) is that if you rank NOTA equally with an
option, that vote is not part of V(A,D) or V(D,A) since neither option is
preferred over the other, and therefore has no effect either way on
whether an option is discarded because it doesn't meet majority.

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

Re: Question to all candidates: how is Debian doing?

[Felix Lechner]

Hi Ted,

On Tue, Mar 29, 2022 at 1:39 PM Theodore Ts'o  wrote:
>
> I'll note that you spent a lot of time about how the council would
> appoint a chair and vice-chair, create bylaws, meet monthly, etc.
>
> However you didn't really answer the question regarding what the
> authorities that the Strategy Council would have.  You've said that
> the strategy council would expand on your answers that were given in
> the top of this thread --- but then what?
>
> If the Strategy Council were to decide that a strategy might be, say,
> "a mouse should put a bell on the cat", how would that strategy be
> carried out?  Debian is a volunteer organization  some have said,
> "do-ocracy".  So I'm not sure what you, as the DPL, would do with the
> conclusions that might be made by such a Strategy Council?

In my reading of Lucas's original questions, he merely probed the
thinking of the candidates. Lucas did not express a concern that any
of us would fail for lack of authority. My responses were meant to
answer his questions.

At the same time, I will answer your questions as well.

The idea behind the committees is to establish, by experiment, that
two or more heads think better together than one alone. Large parts of
the project still operate according to the "strong maintainer" model,
so that's a significant paradigm shift. The committees provide a soft
counterweight that cares about the project as a whole. A better group
spirit alone will be a noticeable advance for Debian.

As to your specific concern about how strategic conclusions might be
used, a project leader can benefit from the advice in many ways. For
example, I would make sure that my own actions work for us in the long
term, and not against us.

As a passionate optimist, I may also need an occasional reality check
from our best and brightest. Since everything is public, all members
can do the same.

Perhaps most significantly, I hope to present the membership with a
wide range of novel and creative proposals. The findings of a Strategy
Council could create a sense of urgency—a rare sentiment in Debian—to
embrace new ideas.

In short, the Strategy Council will help us, as a group, to get
unstuck in many ways. That's how we will get stuff done.

Thank you for your challenging question!

Kind regards,
Felix Lechner

Re: Results for Voting secrecy

[Philip Hands]

Kurt Roeckx  writes:

> On Mon, Mar 28, 2022 at 12:26:51PM -0600, Sam Hartman wrote:
>> > "Kurt" == Kurt Roeckx  writes:
>> 
>> 
>> >> It inadvertently weakened the constitutional protection against
>> >> changes to the constitution.
>> 
>> Kurt> I currently fail to see how it does.
>> 
>> I think Felix's point is that if we had choice 1, 2 and Nota,
>> 
>> People who preferred option 3 would vote N>2=1 or some such.
>> 
>> Because choice 3 was on the ballot, people had options that reflected
>> their preferences and so some of those people voted 3>2>N.
>
> So the only thing I see is that they now had the option to express there
> preferences, while they were limited in how to express their preference
> without option 3.
>
> One way of interpreting the NOTA option is to look say what you think is
> acceptable or not. Without option 3 on the ballot, you can not say you
> think option 1 and 2 are acceptable but prefer option 3. You need to say
> option 1 and 2 are not acceptable, while you actually think they are
> acceptable. With option 3 on the ballot you can really talk about it
> being acceptable or not.
>
> Without option 3, it's probably beter to talk about preference rather
> than being acceptable. If you prefer no change, you just mark it below
> the NOTA option, even when you think option 1 or 2 is acceptable.
>
> Option 3 being on the ballot can make it more likely for option 1 and
> 2 to pass, but that's because people can actually express their opinion.
>
> Our voting system works best when all option are on the ballot. Adding
> more options is not a problem, it has clone independence.
>
>> Felix's point is that the voters who preferred option 3 actually had the
>> power to make it win, provided they were willing to say that they found
>> option 2 unacceptable.
>> Felix's assumption is that if they realized they had that power, they
>> would have exercised it.
>
> But option 2 won, so even if there were people who voted strategically,
> it's not a problem in this vote.

I don't actually mind the outcome (despite my '--12' vote, which was
tactical in the way described above I'm afraid -- probably as a result
of growing up under the UK's first-past-the-post system, which
pretty-much forces people to vote tactically, so I tend to do it out of
habit).

However, I'm failing to understand how the votes are calculated and/or
what certain votes were expected to achieve by the people casting them.

The blurb that's sent out with the votes says:

  To vote "no, no matter what", rank "None of the above" as more
  desirable than the unacceptable choices, or you may rank the "None of
  the above" choice and leave choices you consider unacceptable blank.

which to me suggests that if one ranks something as equal to NotA then
one is not marking it as unacceptable, so presumably it is counted as
acceptable -- is that how such votes are calculated?

It seems 8 people voted '--1-' and 3 voted '1---'.

Did those all contribute to option 2 getting its 3:1 majority?

If so, do we think that someone casting either of those votes was
expecting their vote to be interpreted thus?

Cheers, Phil.
-- 
|)|  Philip Hands  [+44 (0)20 8530 9560]  HANDS.COM Ltd.
|-|  http://www.hands.com/http://ftp.uk.debian.org/
|(|  Hugo-Klemm-Strasse 34,   21075 Hamburg,GERMANY


signature.asc
Description: PGP signature

Re: Results for Voting secrecy

[Pierre-Elliott Bécue]

Felix Lechner  wrote on 27/03/2022 at 22:30:53+0200:

> Hi Kurt,
>
> On Sun, Mar 27, 2022 at 11:03 AM Kurt Roeckx  wrote:
>>
>> Clearly people don't think it's identical, otherwise it would not have
>> been an option, or people would have voted it equally.
>
> People were confused.
>
> Given the stated intent of Option 3 that "early 2022 is not the time
> for rushed changes like this", the Secretary should not have admitted
> that option to the ballot. It inadvertently weakened the
> constitutional protection against changes to the constitution.
>
> The constitution is the project's foundational document.
>
> Neither the option's proponents nor the voters understood the
> deleterious effect. (Nor did I.) At a minimum, the public was entitled
> to a warning from the Project Secretary.
>
> The vote was procedurally defective.
>
>> Option 3 has no effect on the majority results. The options are compared
>> to the NOTA option.
>
> Folks opposing "secret votes" should never have placed Option 2 ahead
> of NOTA, and would not have done so if Option 3 had been absent.
>
> I do not believe it is possible to reconstruct the electorate's intent
> solely from the beat matrix. A better approximation, however, would be
> to also consider the 107 votes who placed Option 3 ahead of Option 2
> in the latter's majority test. That would yield 185 / (61 + 107) = 1.1
> which is less than the factor of 3 mandated by section 4.1.2 of the
> constitution.
>
> As far as I can see, the result is unconstitutional and thus invalid.
>
>> I currently don't see anything wrong with this vote, so I see no reason
>> to redo it.
>
> Please reconsider. Otherwise the project's sole alternative may be to
> replace the Project Secretary.

So… you state the vote is unconstitutional based on an interpretation
you are making of the potential misleading state some ballot option
would have had. And your solution to "fix" this is to *demand* an
_unconstitutional_ decision from the Project Secretary who is
responsible for making sure things stay constitutional, otherwise you
threaten him to get replaced for doing exactly what the project expects
him to do?

I know these times are rough. I know some people in some countries are
trying to show one can look like a democrat while not being one at all.
But seeing this, there and now is kind of painful. I see it as quite
violent and inappropriate.

As far as I'm concerned, I'd be quite worried to have you as a DPL.

FWIW, Kurt, I'd like to express to you my full sympathy and support.

-- 
PEB


signature.asc
Description: PGP signature

Re: Question to all candidates: how is Debian doing?

[Theodore Ts'o]

On Thu, Mar 17, 2022 at 01:57:59PM -0700, Felix Lechner wrote:
> Hi Lucas,
> 
> On Thu, Mar 17, 2022 at 12:52 PM Lucas Nussbaum  wrote:
> >
> > Interesting. What would be the composition, roles, duties of that
> > Strategy Council ?
> 
> Like all committees and boards I envision, the Strategy Council should
> have at least five but no more than twelve members. I personally like
> nine, if you can get that many volunteers

I'll note that you spent a lot of time about how the council would
appoint a chair and vice-chair, create bylaws, meet monthly, etc.

However you didn't really answer the question regarding what the
authorities that the Strategy Council would have.  You've said that
the strategy council would expand on your answers that were given in
the top of this thread --- but then what?

If the Strategy Council were to decide that a strategy might be, say,
"a mouse should put a bell on the cat", how would that strategy be
carried out?  Debian is a volunteer organization  some have said,
"do-ocracy".  So I'm not sure what you, as the DPL, would do with the
conclusions that might be made by such a Strategy Council?

Cheers,

- Ted

General Resolution: Voting secrecy result

[Debian Project Secretary - Kurt Roeckx]

Hi,

The result of the General Resolution is:
Option 2 "Hide identities of Developers casting a particular vote and
allow verification"

The details of the results are available at:
https://www.debian.org/vote/2022/vote_001


Kurt Roeckx
Debian Project Secretary



signature.asc
Description: PGP signature

Re: Results for Voting secrecy

[Thomas Goirand]

On 3/28/22 01:30, Felix Lechner wrote:

Meanwhile, the uncertainty you and I both suffer would be resolved by
a simple redo of the vote with a ballot that carries the appropriate
warning. That is all I asked for.


IMO we shouldn't have voted for this in the first place (for many 
reasons, like not knowing the implementation details, and the fact we 
voted only on one property of the vote).


Please do not make it even more painful asking for another vote (because 
you do not like the outcome).


Cheers,

Thomas Goirand (zigo)

Re: Questions about Debian derivatives

[Wouter Verhelst]

On Mon, Mar 28, 2022 at 10:27:48AM +0800, Paul Wise wrote:
[...]
> > Not sure if you're familiar with extrepo?
> 
> As I understand it, extrepo is more for things like the Mozilla Firefox
> or PostgreSQL repositories than things like Ubuntu? Probably a
> discussion for the extrepo maintainer, or potentially there is room for
> an extrepo extension containing apt configs/keys for derivatives,
> perhaps pulled from the census. As an example of where this would be
> useful, I'm pulling the Ubuntu census apt config into a chdist so I can
> list Ubuntu package versions etc on the command-line.

The URL format of the extrepo metadata currently contains "/debian/" in
it, precisely because I foresaw the possibility to also include
repositories for (or of) our derivatives, and then you might not
necessarily want to mix that with Debian *all* the time.

The current extrepo client does not have the ability to select a
different distribution, mostly because it hasn't come up yet and there
are other things that are important currently; but if there is a need
and/or interest from derivatives, I'm definitely open to adding the
metadata to extrepo and extending the client with functionality that
would be useful.

-- 
 w@uter.{be,co.za}
wouter@{grep.be,fosdem.org,debian.org}

Re: To all candidates: Debian and people with disabilities

[Wouter Verhelst]

Hi Samuel,

On Tue, Mar 22, 2022 at 09:23:38AM +0100, Samuel Thibault wrote:
> Hello,
> 
> Devin Prater, le lun. 21 mars 2022 22:10:15 -0500, a ecrit:
> > As far as backports, my problem is enabling it. Normal desktop users 
> > probably
> > won't even know what that is, and the syntax is rather ugly, to me at least.
> 
> Ok, that's one point that could be worked on: creating an easy way to
> enable backports.

This is what my project, "extrepo", wants to accomplish: to make it
easily possible to enable repositories that are not enabled by default.
You can enable backports on bullseye by way of the following two
commands:

sudo apt install extrepo
sudo extrepo enable debian_backports

(there are a lot of other repositories you can enable that way, btw; to
get a full list, run "extrepo search", although you might want to pipe
it into a pager ;-) )

> At least as a question in the installer,

Adding an installer module for extrepo is on my TODO list, but I do have
a lot on my plate and thus am not sure I'll be able to finish that work
in time for the release.

[...]
-- 
 w@uter.{be,co.za}
wouter@{grep.be,fosdem.org,debian.org}

Re: Results for Voting secrecy

[Kurt Roeckx]

On Mon, Mar 28, 2022 at 12:26:51PM -0600, Sam Hartman wrote:
> > "Kurt" == Kurt Roeckx  writes:
> 
> 
> >> It inadvertently weakened the constitutional protection against
> >> changes to the constitution.
> 
> Kurt> I currently fail to see how it does.
> 
> I think Felix's point is that if we had choice 1, 2 and Nota,
> 
> People who preferred option 3 would vote N>2=1 or some such.
> 
> Because choice 3 was on the ballot, people had options that reflected
> their preferences and so some of those people voted 3>2>N.

So the only thing I see is that they now had the option to express there
preferences, while they were limited in how to express their preference
without option 3.

One way of interpreting the NOTA option is to look say what you think is
acceptable or not. Without option 3 on the ballot, you can not say you
think option 1 and 2 are acceptable but prefer option 3. You need to say
option 1 and 2 are not acceptable, while you actually think they are
acceptable. With option 3 on the ballot you can really talk about it
being acceptable or not.

Without option 3, it's probably beter to talk about preference rather
than being acceptable. If you prefer no change, you just mark it below
the NOTA option, even when you think option 1 or 2 is acceptable.

Option 3 being on the ballot can make it more likely for option 1 and
2 to pass, but that's because people can actually express their opinion.

Our voting system works best when all option are on the ballot. Adding
more options is not a problem, it has clone independence.

> Felix's point is that the voters who preferred option 3 actually had the
> power to make it win, provided they were willing to say that they found
> option 2 unacceptable.
> Felix's assumption is that if they realized they had that power, they
> would have exercised it.

But option 2 won, so even if there were people who voted strategically,
it's not a problem in this vote.

On a ballot with mixed majority requirements the option with the lowest
majority requirement clearly has an advantage. It might be possible to
fix that by requiring all option to have the highest majority
requirement, but I'm not really sure it's better or not.


Kurt