date:20040220

Stefano Mazzocchi dijo:
> Antonio Gallardo wrote:
>
>> Torsten Curdt dijo:
>>
>>>Antonio Gallardo wrote:
>>>
>>>
Carsten Ziegeler dijo:


>What about JCS?
>
>http://jakarta.apache.org/turbine/jcs/index.html


jisp is part of the dependencies. :-(
See:
http://jakarta.apache.org/turbine/jcs/dependencies.html
>>>
>>>hm... what do they do?
>>>maybe we should talk to them and coordinate efforts?
>>
>> Hi Torsten!
>>
>> Long time don't saw a post from you. Welcome back!
>>
>> It is a good idea. I think this is the best we can do. Also I think we
>> will need a database to replace jisp, I thought another good candidate
>> can
>> be hsqldb (already on out CVS) or - Axion
>> http://axion.tigris.org/quickstart.html
>>
>> WDYT?
>>
>> Best Regards,
>>
>> Antonio Gallardo
>
> Antonio, Jisp is not a database! is a binary object storage system. The
> closest thing I can think of is BerkeleyDB or, in short, a persistent
> hashtable.
>
> That's all we need.

Yep. This is a not good option. Thanks for the advise! :-D

Best Regards,

Antonio Gallardo

RE: Jisp 3.0 moved to GPL licence

Reinhard Poetz dijo:
>
>
>> -Original Message-
>> From: Stefano Mazzocchi [mailto:[EMAIL PROTECTED]
>> Sent: Friday, February 20, 2004 3:55 PM
>> To: [EMAIL PROTECTED]
>> Subject: Re: Jisp 3.0 moved to GPL licence
>>
>>
>> Steven Noels wrote:
>>
>> > On 20 Feb 2004, at 12:52, Torsten Curdt wrote:
>> >
>> >> hm... what do they do?
>> >> maybe we should talk to them and coordinate efforts?
>> >
>> >
>> > +1 - seems like the best plan to go forward without over-reacting.
>> >
>> > FWIW: I think we should actively weed out one-man-effort
>> dependencies.
>> > People have the freedom to change their mind, but we shouldn't be a
>> > victim of that.
>>
>> +1000
>>
>> But keep in mind that we are never locked in as we always
>> have the right
>> to fork. So, let's keep reasonable and not panic.
>
> +1
> As Antonio already said it's no emergency becaue we *do use* a former
> version under a 'friendly' licency and we have time to evaluate all
> possible options without getting panic. For now this is
>
>  - forking Jisp 2.6
>  - JCS
>  - OSCache from Opensymphonie
>
> Some mentioned RDBMS systems or XMLDBs - of course an option but I don't
> think they are designed for our caching needs (but maybe I'm wrong
> here).

I will prefer to eat our own "dog food". I think Apache JCS is a good
candidate. If there are bugs, let fix them! Please keep in mind that
excalibur need to be changed too.

Best Regards,

Antonio Gallardo

Re: Jisp 3.0 moved to GPL licence

Scott Robert Ladd dijo:
>> On 20.02.2004 12:25, Antonio Gallardo wrote:
>>> AFAIK, we must move away from jisp. That was an ugly move! :-(
>
> I'm willing to work with Apache on this issue, so how am I being ugly?

Hi:

Sorry, Scott It was not addressed to you. I am glad to hear that. :-D

Best Regards,

Antonio Gallardo

Re: Jisp 3.0 moved to GPL licence

Joerg Heinicke dijo:
> On 20.02.2004 12:25, Antonio Gallardo wrote:
>
>> AFAIK, we must move away from jisp. That was an ugly move! :-(
>
> IMO you can not convict somebody just because he chose a license that
> does not match your requirements. He probably did not do that to make us
> angry. Maybe you chose only a wrong word (as the English language is not
> the natural language of most of us) but 'ugly' is a really hard word in
> this context I think.

Sorry, for the word. I know everybody is free to choose the License that
meet the best for him. I just tried to point this was a not good move from
us. I don't really know people behind jisp and not tried to flame them. I
saw this just from my own (Cocoon) perspective. Please forgive me if it
can be interpreted in other way. English is not my natural language.
Please don't take the words in a wrong sense.

Best Regards,

Antonio Gallardo

Re: Of One-man Efforts and the like

Stefano Mazzocchi wrote:
there are many ways to describe the spirit of the apache community, but 
there is one that I like more than all the others: "we care about people 
more than we care about code".
Well then, we are on the same philosophical ground! :)

Now. You, as the author and copyright holder, have the freedom to modify 
licensing at any time and for whatever reason. Here, nobody will tell 
you what's good or bad for you, you know that better than we all do.
As I said in another message, I'm going to ponder the issues (again) 
this weekend. I appreciate the detailed responses, though.

Now for apache pretending and not giving back.

telnet www.coyotegulch.com 80
Trying 64.70.152.229...
Connected to coyotegulch.com.
Escape character is '^]'.
HEAD / HTTP/1.0
HTTP/1.1 200 OK
Date: Fri, 20 Feb 2004 22:11:31 GMT
Server: Apache
Connection: close
Content-Type: text/html
Point taken, though given the complexity of my personal website, I could 
probably use just about any web server. I don't do much server-side work 
anymore; my specialty is high-performance and numeric applications. I 
suspect Apache has more than its share of experts on parallel coding and 
optimization...

That given, I'll consider what everyone's said. I'll do a bit more 
research and thinking.

..Scott

--
Scott Robert Ladd
Coyote Gulch Productions (http://www.coyotegulch.com)
Software Invention for High-Performance Computing

Re: The (new) Jisp 3.0 license

I'm going to consider some of the philosophical ramification this 
weekend; I have a 3.1 version of Jisp in the Wings, and I can always 
relicense it.

One issue that I did want to address:

Steven Noels wrote:
I'm intrigued though by the questions you receive about Jisp from the 
Apache side: do these originate from Cocoon users? While Jisp is 
considered a core (though switchable) part of Cocoon (it's the default 
of two cache implementations), its existence is shielded from the 
passing eye of a normal user. Also, we would expect that Jisp issues 
within Cocoon would be reported on the Cocoon lists, which happens, but 
not regularly enough to suspect people to come and nag you directly. Or 
is it because of other ASF projects using Jisp?
I don't know the precise origin (within the Apache framework) of the 
questions; I do know that, since last fall, three separate companies 
contacted me about Jisp, based on their discovery of it in the Apache 
code base.

In one case, the company paid me a small amount to help them try to win 
a contract with a Big Bank; their bid failed, though not due to Jisp. 
They used Jisp for a purpose similar to that of Cocoon.

The other two companies approached me with similar propisitions; for 
some reason both were incapable of some rather simple extensions of Jisp 
required for their own commercial projects. Since both were in a rush, 
and I'm always looking to make new customers, I jumped in and wrote what 
they asked for; neither paid me, and one was quite rude about the affair.

After the above experiences, I relicensed Jisp under the GPL.

I am open to being convinced that another license would be more beneficial.

..Scott

--
Scott Robert Ladd
Coyote Gulch Productions (http://www.coyotegulch.com)
Software Invention for High-Performance Computing

Re: Of One-man Efforts and the like

On 20 Feb 2004, at 18:52, Scott Robert Ladd wrote:

> FWIW: I think we should actively weed out one-man-effort 
dependencies.
> People have the freedom to change their mind, but we shouldn't be a
> victim of that.
See Stefano's last remark: we are actively pursuing project stability 
by making sure Cocoon has a serious legal foundation, and by choosing 
"sound" dependencies. This choice can be based on both technical or 
community criteria, and hopefully on both at the same time. So yes, we 
have been debating dependencies quite a lot lately, and you should see 
my quote in this context: we got bitten by this before, and it hurts. 
You didn't bite us yet, but we see your teeth. ;-)

As I stated before, I'm willing to provide Apache with a specific 
license if it makes everyone's life easier.
As you obviously see now, it ain't that simple. We need to care about 
"the users of our users" as well.

Now about your pertinent question on "community", you should go and 
inquire the incubating ASF projects. If a company like BEA still sees 
additional marketing value by incubating a project at Apache and not 
going open source on their own, I'm utterly convinced that single-man 
projects can reap lots of exposure by trying to join the ASF 
(http://incubator.apache.org/)

I've helped the BEA peeps with XMLBeans, and I'm happy to help any 
other viable project. But the burden of the work wil be on your side - 
and potential benefit will be related to that.

--
Steven Noelshttp://outerthought.org/
Outerthought - Open Source Java & XMLAn Orixo Member
Read my weblog athttp://blogs.cocoondev.org/stevenn/
stevenn at outerthought.orgstevenn at apache.org

Re: Of One-man Efforts and the like

Scott Robert Ladd wrote:

After looking at the archives of recent messages, I decided to join this 
group for a short time, given that my work is being discussed.

As a matter of clarification: Jisp is not a database, it is a embedded 
engine that can be used to *create* databases.

I found this quote rather interesting:

 > FWIW: I think we should actively weed out one-man-effort dependencies.
 > People have the freedom to change their mind, but we shouldn't be a
 > victim of that.
Change the words "one-man" to "open source", and see how it reads. I've 
been in on debates about whether or not to use "open source" code... I 
suggest the Apache might want to take the high road here, and judge 
matters based on quality and applicability, not broad prejudices.
Scott,

there are many ways to describe the spirit of the apache community, but 
there is one that I like more than all the others: "we care about people 
more than we care about code".

While the FSF family of licenses is reciprocal (aka viral, but the term 
has a negative connotation and it's not appropriate to use) in respect 
of code freedom protection (and suggests ethical uses of branding), the 
Apache family of licenses is reciprocal in respect of brand protection 
(and suggests ethical code freedom).

Both are "free software" licenses in Richard Stallman's sense of the term.

But because you are redistributing software (and not brands!), the GPL 
replicates in code, while the ASL doesn't.

This makes it possible to include ASL software in the GPL, while the 
opposite is not possible.

And will never be.

By design.

Now. You, as the author and copyright holder, have the freedom to modify 
licensing at any time and for whatever reason. Here, nobody will tell 
you what's good or bad for you, you know that better than we all do.

But here I'm going to tell you what's good and bad for us, so that you 
can have enough information to act in the way that you consider most 
appropriate.

As I stated before, I'm willing to provide Apache with a specific 
license if it makes everyone's life easier.
A specific licensing to the ASF won't be acceptable unless the ASF is 
able to redistribute and sublicense with the same terms. This in fact, 
means dual licensing. The mozilla foundation choose this path.

Be aware though, that the ASF is against dual licensing because it 
leaves the door open for SCO-like attacks.

In short, if Jisp is GPLed, we will be forced to look someplace else or 
fork the current 2.x codebase (which is ethically bad, but legally 
allowed by the licensing terms under which Jisp 2.x was distributed)

- o -

So much for licensing.

Now for the one-man show.

The ASF is respected in the software industry not only for the quality 
of the code it produces but for the long-term stability of the 
communities that maintain a software.

As you show rather evidently, a single person cannot reach enough 
critical mass to run a project and maintain it in time. It's an unstable 
situation: either the software dies, or it becomes commercial or a 
community develops around it.

Just like the strenght of a chain is given by its weakest ring, a 
software project that wishes to give long stability as a feature cannot 
affort to base itself on projects that don't exhibit the same trends.

Open Source licenses allows forking to be a way out (this is critical 
requirements for OSI certification of a license), but ethically, it is 
considered a dramatic event and it's highly discouradged, given the 
amount of negative energy that brings with it.

Now, this is *NOT* a critique against you, your way of doing stuff or 
your licensing choices, but it's just stating facts on how software 
licensing and project collaborations work.

Be aware of the fact that we are debating the same issues with other 
libraries that are core for us and this is because we have been burned 
in the past and don't want to get burned again.

 - o -

So much for one-man shows.

Now for apache pretending and not giving back.

telnet www.coyotegulch.com 80
Trying 64.70.152.229...
Connected to coyotegulch.com.
Escape character is '^]'.
HEAD / HTTP/1.0
HTTP/1.1 200 OK
Date: Fri, 20 Feb 2004 22:11:31 GMT
Server: Apache
Connection: close
Content-Type: text/html
--
Stefano.

smime.p7s
Description: S/MIME Cryptographic Signature

Re: The (new) Jisp 3.0 license

On 20 Feb 2004, at 18:33, Scott Robert Ladd wrote:

Steven Noels wrote:

let me first start with thanking you for a wonderful set of tools, 
which we have eagerly been using over the years.
I'm glad to know it was useful.
We hope it still can be. :)



Quite to our surprise however, we just found out that Jisp 3 now 
comes with the MySQL-like combo-license: GPL + commercial licensing 
possibility.
Of course, we cannot and will not have any say into which license you 
should use for your products, since it's your blood and tears that 
went into it. As open source developers however, we have some 
considerations about licenses which we'd like to confront you with.
I appreciate your talking to me about this. Please indulge me in a bit
of background story, so you understand why I changed licenses.
I'm not a twenty-something college student who can afford to hack free
software for free; I am an independent business man and forty-something
father of three. My first "free" products appeared back in the 1980s,
and I've written several programming books, so I'm a long-time believer
in sharing code with the community. A society can not progress if
knowledge is kept secret.
I guess quite few of us were already aware of the fact that you were 
not a youngster, doing Jisp just to be "kewl". People don't write stuff 
like Jisp to increase Slashdot karma. :-)

Many of the active project participants on the Cocoon projects fit in 
your profile as well: independent freelancers or tiny businesses 
working on Cocoon collaboratively, if not mainly for their own private 
business endeavours, then very often because of a need or willingness 
to share knowledge and learn in the process.

As such, I originally released Jisp under a libpng-style license.

In the last year or so, I've been hounded by people who "demand" 
support
for Jisp -- including those who use Jisp via Apache. While I politely
inform them that my time isn't free, the frequency of such requests is
increasing. When I request some form of renumeration for providing
support, they act as if I've suggested concourse with a devil.
We (the Cocoon developers) understand this - since quite often we are 
contacted directly - offlist - for support questions. We do have a 
users list to refer users to, and sometimes (not that often) business 
can grow out of these direct contacts. I think our situation is 
slightly different from yours since we serve not only a community of 
users, but we are also a strongly intertwingled group of developers and 
committers, so the "burden" of support is shared across us all. We're 
not perfect though, and questions (about Cocoon) remain unanswered. 
When that happens, we see users stepping in to help each other. But 
that's the big difference between the cathedral and the bazaar, I 
guess. We believe our open source project is composed of both community 
and code. The community gives longevity to our code, so that people can 
come and go. The ASF is always open for new open source projects, BTW, 
and the ASF brand does make a huge difference.

I'm intrigued though by the questions you receive about Jisp from the 
Apache side: do these originate from Cocoon users? While Jisp is 
considered a core (though switchable) part of Cocoon (it's the default 
of two cache implementations), its existence is shielded from the 
passing eye of a normal user. Also, we would expect that Jisp issues 
within Cocoon would be reported on the Cocoon lists, which happens, but 
not regularly enough to suspect people to come and nag you directly. Or 
is it because of other ASF projects using Jisp?

My frustration lies with people who use and praise Jisp without any
feeling of obligation to me, other than a hearty handshake (at best). 
My
frustration grows when I know that people are profiting from my code,
without so much as an acknowledgement.

I am not a greedy soul, and excess wealth is not one of my goals in
life; however, I have three daughters, a wife, and a managerie of
critters that need my financial support. These are dark economic times
in our industry; survival for an independent consultant (like me) is
tenuous even in the best of times.
For me, the change in license (to the GPL) is an attempt to find equity
in people's use of my code. "Fairness" is, of course, a matter of
perception. In my case, I have received nothing from the Apache project
beyond the "hearty handshake" level. It's not that I don't want Apache
to use Jisp -- it's that I know people are paid to work on Apache, and
that many people make money selling Apache support and services.
I cannot speak for other projects, but very few, if any at all of the 
developers involved in Cocoon are getting paid full time by their 
employer to work on the project - all of them are here because of a 
passion for the project and its community of users and developers. 
Quite a few of us are effectively setting up a consulting shop around 
Cocoon-related services, but we don't consider the liberal Apache 
l

Re: Of One-man Efforts and the like

I would *love* to spend time working on projects like Apache -- but I 
haven't the luxury of free time. Do I think people can make money from 
working on free software? Certainly -- my primary contract right now is 
to write free software for a big British company (heh, they're 
outsourcing to America).

And while money is nice, I'm also willing to consider various designs 
for mutually-beneficial arrangements.
So the bottom line probably is... at least the Apache Cocoon community
is very thankful for the existing free version. But instead of bugging
you we would like to help fixing bugs and support what we use.
We would also like to do this for the new version. But a GPL licence
makes it impossible for us to ship it with Cocoon. Even if your are
so generous offering a special Apache permission.
I guess a lot of people on this list are working on Cocoon for money
(besides for the fun). I don't know if this will work for Jisp the
same way and if you can make enough money out of that.
Noone will blame you if that's not enough for you. But IMHO these days
it's the know-how that still sells good. Consider seeing it like having 
free support for your product. ...but that would require Jisp to live
in a CVS (sourceforge,apache,your own,...) and for Cocooners to help
an Apache compatible licence ;)

As good as I know the people of this community there will surely be
no harsh feelings if you decide otherwise. But it's up to you if you
want us to help you as your software helps us.
...otherwise we probably need to move to something else
...or keep using the old version.
My 2 cents
--
Torsten

Re: The (new) Jisp 3.0 license

Scott Robert Ladd wrote:

Steven Noels wrote:

We kindly ask you to consider these issues and get back to us at your 
convenience. The Cocoon PMC list ([EMAIL PROTECTED]) is private 
to PMC members only, if you would like to make a public comment feel 
free to do that on [EMAIL PROTECTED]


I've CC'd this to the list above; I'd appreciate it if any discussion 
would be CC'd back to me as well. Please read the entire message 
before commenting; it may start out sounding a bit negative, but in 
the end, I think you get what you want. :)

let me first start with thanking you for a wonderful set of tools, 
which we have eagerly been using over the years.


I'm glad to know it was useful.

Cocoon has been using Jisp for quite some time, and it served us well 
as a migration from a filesystem based cache store. We have been 
hitting some Jisp bugs lately, which made us looking forward to 
migrate to the upcoming 3.0 version.


Older versions of Jisp did, indeed, have some problems; initially, 
Jisp was an algorithmic demonstration for one of my books, and it was 
not intended for "real world use".

Given the number of people who have used Jisp, I decided to refactor 
the package in a more professional form. Thus, Jisp 3.0, which is 
based on the code I use in commercial products.

Quite to our surprise however, we just found out that Jisp 3 now 
comes with the MySQL-like combo-license: GPL + commercial licensing 
possibility.

Of course, we cannot and will not have any say into which license you 
should use for your products, since it's your blood and tears that 
went into it. As open source developers however, we have some 
considerations about licenses which we'd like to confront you with.


I appreciate your talking to me about this. Please indulge me in a bit 
of background story, so you understand why I changed licenses.

I'm not a twenty-something college student who can afford to hack free 
software for free; I am an independent business man and 
forty-something father of three. My first "free" products appeared 
back in the 1980s, and I've written several programming books, so I'm 
a long-time believer in sharing code with the community. A society can 
not progress if knowledge is kept secret.

As such, I originally released Jisp under a libpng-style license.

In the last year or so, I've been hounded by people who "demand" 
support for Jisp -- including those who use Jisp via Apache. While I 
politely inform them that my time isn't free, the frequency of such 
requests is increasing. When I request some form of renumeration for 
providing support, they act as if I've suggested concourse with a devil.

My frustration lies with people who use and praise Jisp without any 
feeling of obligation to me, other than a hearty handshake (at best). 
My frustration grows when I know that people are profiting from my 
code, without so much as an acknowledgement.

I am not a greedy soul, and excess wealth is not one of my goals in 
life; however, I have three daughters, a wife, and a managerie of 
critters that need my financial support. These are dark economic times 
in our industry; survival for an independent consultant (like me) is 
tenuous even in the best of times.

For me, the change in license (to the GPL) is an attempt to find 
equity in people's use of my code. "Fairness" is, of course, a matter 
of perception. In my case, I have received nothing from the Apache 
project beyond the "hearty handshake" level. It's not that I don't 
want Apache to use Jisp -- it's that I know people are paid to work on 
Apache, and that many people make money selling Apache support and 
services.

It seems to me that those who find value in my creation should also be 
willing to reward me for my contribution.

The ASF policy is not to use, link or distribute software 
(components) which use the (L)GPL license, since these licenses 
prohibit the redistribution of ASF software linking, using or 
embedding such dependencies along the same terms as the Apache 
Software License (1 or 2) allows.


I am not a fan of Mr. Stallman's religion; however, his license does 
have the weight of support behind it at this point, and using GPL 
solves a number of problems I've had in being compatible with some 
other "free" projects.

With the new ASL, it is allowed that ASF software is (re)packaged as 
a new open source project which is distributed under the GPL. Adding 
onto that, we regard the new ASL as being compatible with the GPL, 
while the FSF decided the contrary, without tangible reason: a matter 
of policy rather than legal differences, so it seems. So we have a 
bit of an issue with the GPL, to put it frankly.


Ah, more of Mr. Stallman's pedantry -- one of the reasons I resisted 
the GPL for many years.

The license change of Jisp is most likely going to force us to 
replace Jisp by some other dbm. Other ASF projects which use Jisp 
will be forced to do the same. This is bad since we, apart from some 
small issues, were quite happy

Re: Of One-man Efforts and the like

Scott Robert Ladd wrote:

Hunsberger, Peter wrote:

Fair enough; reading between the lines, I guess you're partly saying
that prior versions of Jisp weren't robust enough to be readily adopted
by a community?


Jisp began life as an accidental product. People found it in my books, 
and asked to license it. The book code was demonstrative; it had never 
been tested. Over the course of several years, I patched the code as 
best I could, but finally decided that the underlying core had serious 
flaws. Thus I did a major rewrite used by my few commercial customers, 
but not released as free code.

As I've moved away from writing books (too much work for too little 
money), Jisp has taken on a new role as an advertisement. As such, I 
released the commercial version as free software (version 3.0), under 
the GPL and commercial license to emphasize that this is a product and 
not some hobbiest's weekend hack.

Wish I could put my money where my mouth is on this issue (so to speak);
at this point I have to drop out of the discussion... 


As is usual in life, the best people tend to have the least money. :)

I would *love* to spend time working on projects like Apache -- but I 
haven't the luxury of free time. Do I think people can make money from 
working on free software? Certainly -- my primary contract right now 
is to write free software for a big British company (heh, they're 
outsourcing to America).

And while money is nice, I'm also willing to consider various designs 
for mutually-beneficial arrangements.


Scott,

By and large the people here recognize your right (and everyone's right) 
to do what you want with your code.  I appreciate your quandry and 
position.  Apache projects (at least this one) have their own "religion" 
valuing community - but not all projects translate well to community 
ownership.  There may be some very real benefits to a community like 
Cocoon to avoid single-person projects because they have a risk of 
"disappearing" either by license change or by "retirement" etc.  Of 
course these or similar risks exist in community-based projects, as well 
as pure commercial projects.  I personally have been bitten by all three 
of these poisons as have many here.  All that to say, I hope we haven't 
gotten off on the wrong foot here.  Thanks for joining in the discussion 
directly, and don't let our liveliness turn you off.

I'll add a brief comment and suggestion on the license-related thread.

Geoff

Re: Jisp 3.0 moved to GPL licence

On 20.02.2004 21:18, Ryan Hoegg wrote:

- Prevayler (http://www.prevayler.org/wiki.jsp)
Also LGPL.

Joerg

Re: Jisp 3.0 moved to GPL licence

2004-02-20 Thread Ryan Hoegg

Also add to this list:

- Xindice Filer (mentioned in subsequent posts)
- Prevayler (http://www.prevayler.org/wiki.jsp)
--
Ryan Hoegg
ISIS Networks
http://www.isisnetworks.net/
Reinhard Poetz wrote:

As Antonio already said it's no emergency becaue we *do use* a former
version under a 'friendly' licency and we have time to evaluate all
possible options without getting panic. For now this is
- forking Jisp 2.6
- JCS
- OSCache from Opensymphonie
Some mentioned RDBMS systems or XMLDBs - of course an option but I don't
think they are designed for our caching needs (but maybe I'm wrong
here).
--
Reinhard

Re: Of One-man Efforts and the like

Hunsberger, Peter wrote:
Fair enough; reading between the lines, I guess you're partly saying
that prior versions of Jisp weren't robust enough to be readily adopted
by a community?
Jisp began life as an accidental product. People found it in my books, 
and asked to license it. The book code was demonstrative; it had never 
been tested. Over the course of several years, I patched the code as 
best I could, but finally decided that the underlying core had serious 
flaws. Thus I did a major rewrite used by my few commercial customers, 
but not released as free code.

As I've moved away from writing books (too much work for too little 
money), Jisp has taken on a new role as an advertisement. As such, I 
released the commercial version as free software (version 3.0), under 
the GPL and commercial license to emphasize that this is a product and 
not some hobbiest's weekend hack.

Wish I could put my money where my mouth is on this issue (so to speak);
at this point I have to drop out of the discussion... 
As is usual in life, the best people tend to have the least money. :)

I would *love* to spend time working on projects like Apache -- but I 
haven't the luxury of free time. Do I think people can make money from 
working on free software? Certainly -- my primary contract right now is 
to write free software for a big British company (heh, they're 
outsourcing to America).

And while money is nice, I'm also willing to consider various designs 
for mutually-beneficial arrangements.

--
Scott Robert Ladd
Coyote Gulch Productions (http://www.coyotegulch.com)
Software Invention for High-Performance Computing

Re: Of One-man Efforts and the like

2004-02-20 Thread Bertrand Delacretaz

Le Vendredi, 20 fév 2004, à 20:19 Europe/Zurich, Hunsberger, Peter a 
écrit :
...Now, exactly, how you go about building a community is another 
question.
But, it seems that perhaps some of the Cocoon project members might be
willing to help?..
After starting an Open-Source project myself (jfor, see the story [1]), 
which for several reasons failed to attract a community of developers, 
and later joining the very healthy Cocoon community,  I'm still 
wondering what it takes to build a community.

I think it has a lot to do with the kind of project: with Cocoon, being 
targeted to developers, it's way easier to build a community of 
developers than with projects targeted to end users (and also, 
obviously, once the project reaches a certain level of recognition it 
is even easier).

Users will come, use your stuff, maybe complain a bit or praise a bit, 
some of them might help a bit on the mailing lists but it's usually 
very little compared to the number of downloads.

Developers will come, use your stuff and find that they can improve it 
with the same tools and skills that they need to use it, and get better 
at their craft in doing it. This makes it much easier to contribute 
than if you have to learn a different skill set for contributing than 
the one needed to use the software.

Rather than "improve it" I should say "extend it" - improving in the 
sense of fixing bugs does not give you as much "fame" as contributing a 
completely new thing. This is also an important factor IMO.

Jisp might be somewhere in the middle, as developers can use it as a 
black box without knowing much about how it works. And contributions to 
Jisp, as I understand it, would be more in the area of "improve it" 
rather than "extend it".

Anyway it's not easy, and until a project reaches a critical mass it 
needs lots of investment, in time, effort and money (as you have to eat 
in the meantime), and as with all investments one can get out of steam 
at some point, it's very understandable.

My 2 cents!

-Bertrand

[1] http://codeconsult.ch/bertrand/archives/68.html

Re: Jisp 3.0 moved to GPL licence

On 20.02.2004 12:25, Antonio Gallardo wrote:
AFAIK, we must move away from jisp. That was an ugly move! :-(
I intended no ugliness; I never even considered that Apache would be 
offended by my change in license.

When I used libpng, I had people complaining I wasn't using GPL; when I 
use the GPL, people complain that I should use something else. I'm sure 
you can understand this, given the current controversy over changes in 
Apache's license.

I'm willing to work with Apache on this issue, so how am I being ugly?

..Scott

--
Scott Robert Ladd
Coyote Gulch Productions (http://www.coyotegulch.com)
Software Invention for High-Performance Computing

Re: LDAPTransformer

On 20.02.2004 15:38, [EMAIL PROTECTED] wrote:

I apologize DEEPLY for the horrible formatting of my previous note.  I'm 
on Lotus Notes here at work, and it doesn't seem to cooperate.
As long it's pure text we are satisfied with the format :)

Hi,

I have 2 "features" that I think would be of great value to the 
LDAPTransformer. I have already begun working on a patch to implement 
both.
I can not tell you much on LDAP as I never used it myself.

Comments on both of these are more than welcome.  And PLEASE let me know 
if anyone has already engaged either or both of these issues.  I did 
research both on Bugzilla and the cocoon-dev archives, and did not find 
anything addressing either of these.
If anybody works on this feature (or has done it in the past) he he is 
not doing it in public. At least I did not read anything recently of 
patches to LDAPTransformer. I only can remember patches provided by Yury 
Mikhienko. For the latest commits see 
http://marc.theaimsgroup.com/?l=xml-cocoon-cvs&w=4&r=1&s=LDAPTransformer&q=b 
or directly the CVS 
http://cvs.apache.org/viewcvs.cgi/cocoon-2.1/src/blocks/naming/java/org/apache/cocoon/transformation/LDAPTransformer.java.

Once I have my patch complete, I will submit it via Bugzilla.
We are looking forward to it.

Joerg

RE: Of One-man Efforts and the like

2004-02-20 Thread Hunsberger, Peter

Scott Robert Ladd <[EMAIL PROTECTED]> write:

 
> Hunsberger, Peter wrote:
> > I think you miss the issue here: there's certainly no equivalency 
> > between "one-man-effort" and "open source".
> 
> This last week has seen some intense license conflicts between very 
> large open source projects: XFree86, Apache, and FSF/GNU have 
> all made 
> choices that have lead to (or could lead to) changes in various 
> distributions. Changing licenses is not merely the province 
> of "one man" 
> operations.

Sure, but that's not the real issue (other than the fact I'd to see
everyone adopt a Apache friendly license)...
  
> > It seems that on one hand
> > you complain that you can't afford the time to provide free support 
> > for everyone using Jisp, then on the other you want people 
> to accept 
> > code that has only you supporting it.
> 
> One reason for Jisp 3.0, by the way, was to reduce my support 
> load. The 
> old version of Jisp was not "robust", to use a marketroid term.

Fair enough; reading between the lines, I guess you're partly saying
that prior versions of Jisp weren't robust enough to be readily adopted
by a community?
 
> > The solution to your (and the Cocoon
> > communities) dilemma is not to have a restrictive license 
> (as I think 
> > you agree), or to accept "one-man-efforts", but rather, to have a 
> > community supporting the code.  Once you have a community 
> supporting 
> > the code you will personally no longer be "pestered" to 
> provide free 
> > support.  (OTOH, as one of the primary developers of the 
> code base you 
> > will be seen as having value for anyone willing to put up consultant
> > dollars)
> 
> Perhaps.
> 
> > Now, exactly, how you go about building a community is another 
> > question. But, it seems that perhaps some of the Cocoon project 
> > members might be willing to help?
> 
> I'm open to concrete suggestions.

Wish I could put my money where my mouth is on this issue (so to speak);
at this point I have to drop out of the discussion...

Re: Of One-man Efforts and the like

Hunsberger, Peter wrote:
I think you miss the issue here: there's certainly no equivalency
between "one-man-effort" and "open source".
This last week has seen some intense license conflicts between very 
large open source projects: XFree86, Apache, and FSF/GNU have all made 
choices that have lead to (or could lead to) changes in various 
distributions. Changing licenses is not merely the province of "one man" 
operations.

It seems that on one hand
you complain that you can't afford the time to provide free support for
everyone using Jisp, then on the other you want people to accept code
that has only you supporting it.
One reason for Jisp 3.0, by the way, was to reduce my support load. The 
old version of Jisp was not "robust", to use a marketroid term.

The solution to your (and the Cocoon
communities) dilemma is not to have a restrictive license (as I think
you agree), or to accept "one-man-efforts", but rather, to have a
community supporting the code.  Once you have a community supporting the
code you will personally no longer be "pestered" to provide free
support.  (OTOH, as one of the primary developers of the code base you
will be seen as having value for anyone willing to put up consultant
dollars)
Perhaps.

Now, exactly, how you go about building a community is another question.
But, it seems that perhaps some of the Cocoon project members might be
willing to help?
I'm open to concrete suggestions.

..Scott

--
Scott Robert Ladd
Coyote Gulch Productions (http://www.coyotegulch.com)
Software Invention for High-Performance Computing

Re: Jisp 3.0 moved to GPL licence

On 20.02.2004 12:25, Antonio Gallardo wrote:

AFAIK, we must move away from jisp. That was an ugly move! :-(
IMO you can not convict somebody just because he chose a license that 
does not match your requirements. He probably did not do that to make us 
angry. Maybe you chose only a wrong word (as the English language is not 
the natural language of most of us) but 'ugly' is a really hard word in 
this context I think.

Joerg

RE: Of One-man Efforts and the like

2004-02-20 Thread Hunsberger, Peter

Scott Robert Ladd <[EMAIL PROTECTED]> suggests:

> I found this quote rather interesting:
> 
>  > FWIW: I think we should actively weed out one-man-effort 
> dependencies.  > People have the freedom to change their 
> mind, but we shouldn't be a  > victim of that.
> 
> Change the words "one-man" to "open source", and see how it 
> reads. I've 
> been in on debates about whether or not to use "open source" 
> code... I 
> suggest the Apache might want to take the high road here, and judge 
> matters based on quality and applicability, not broad prejudices.

I think you miss the issue here: there's certainly no equivalency
between "one-man-effort" and "open source".  It seems that on one hand
you complain that you can't afford the time to provide free support for
everyone using Jisp, then on the other you want people to accept code
that has only you supporting it.  The solution to your (and the Cocoon
communities) dilemma is not to have a restrictive license (as I think
you agree), or to accept "one-man-efforts", but rather, to have a
community supporting the code.  Once you have a community supporting the
code you will personally no longer be "pestered" to provide free
support.  (OTOH, as one of the primary developers of the code base you
will be seen as having value for anyone willing to put up consultant
dollars)

Now, exactly, how you go about building a community is another question.
But, it seems that perhaps some of the Cocoon project members might be
willing to help?

Of One-man Efforts and the like

After looking at the archives of recent messages, I decided to join this 
group for a short time, given that my work is being discussed.

As a matter of clarification: Jisp is not a database, it is a embedded 
engine that can be used to *create* databases.

I found this quote rather interesting:

> FWIW: I think we should actively weed out one-man-effort dependencies.
> People have the freedom to change their mind, but we shouldn't be a
> victim of that.
Change the words "one-man" to "open source", and see how it reads. I've 
been in on debates about whether or not to use "open source" code... I 
suggest the Apache might want to take the high road here, and judge 
matters based on quality and applicability, not broad prejudices.

As I stated before, I'm willing to provide Apache with a specific 
license if it makes everyone's life easier.

..Scott

--
Scott Robert Ladd
Coyote Gulch Productions (http://www.coyotegulch.com)
Software Invention for High-Performance Computing

Re: The (new) Jisp 3.0 license

Steven Noels wrote:
We kindly ask you to consider these issues and get back to us at your 
convenience. The Cocoon PMC list ([EMAIL PROTECTED]) is private to 
PMC members only, if you would like to make a public comment feel free 
to do that on [EMAIL PROTECTED]
I've CC'd this to the list above; I'd appreciate it if any discussion
would be CC'd back to me as well. Please read the entire message before
commenting; it may start out sounding a bit negative, but in the end, I
think you get what you want. :)
let me first start with thanking you for a wonderful set of tools, which 
we have eagerly been using over the years.
I'm glad to know it was useful.

Cocoon has been using Jisp for quite some time, and it served us well as 
a migration from a filesystem based cache store. We have been hitting 
some Jisp bugs lately, which made us looking forward to migrate to the 
upcoming 3.0 version.
Older versions of Jisp did, indeed, have some problems; initially, Jisp
was an algorithmic demonstration for one of my books, and it was not
intended for "real world use".
Given the number of people who have used Jisp, I decided to refactor the
package in a more professional form. Thus, Jisp 3.0, which is based on
the code I use in commercial products.
Quite to our surprise however, we just found out that Jisp 3 now comes 
with the MySQL-like combo-license: GPL + commercial licensing possibility.

Of course, we cannot and will not have any say into which license you 
should use for your products, since it's your blood and tears that went 
into it. As open source developers however, we have some considerations 
about licenses which we'd like to confront you with.
I appreciate your talking to me about this. Please indulge me in a bit
of background story, so you understand why I changed licenses.
I'm not a twenty-something college student who can afford to hack free
software for free; I am an independent business man and forty-something
father of three. My first "free" products appeared back in the 1980s,
and I've written several programming books, so I'm a long-time believer
in sharing code with the community. A society can not progress if
knowledge is kept secret.
As such, I originally released Jisp under a libpng-style license.

In the last year or so, I've been hounded by people who "demand" support
for Jisp -- including those who use Jisp via Apache. While I politely
inform them that my time isn't free, the frequency of such requests is
increasing. When I request some form of renumeration for providing
support, they act as if I've suggested concourse with a devil.
My frustration lies with people who use and praise Jisp without any
feeling of obligation to me, other than a hearty handshake (at best). My
frustration grows when I know that people are profiting from my code,
without so much as an acknowledgement.
I am not a greedy soul, and excess wealth is not one of my goals in
life; however, I have three daughters, a wife, and a managerie of
critters that need my financial support. These are dark economic times
in our industry; survival for an independent consultant (like me) is
tenuous even in the best of times.
For me, the change in license (to the GPL) is an attempt to find equity
in people's use of my code. "Fairness" is, of course, a matter of
perception. In my case, I have received nothing from the Apache project
beyond the "hearty handshake" level. It's not that I don't want Apache
to use Jisp -- it's that I know people are paid to work on Apache, and
that many people make money selling Apache support and services.
It seems to me that those who find value in my creation should also be
willing to reward me for my contribution.
The ASF policy is not to use, link or distribute software (components) 
which use the (L)GPL license, since these licenses prohibit the 
redistribution of ASF software linking, using or embedding such 
dependencies along the same terms as the Apache Software License (1 or 
2) allows.
I am not a fan of Mr. Stallman's religion; however, his license does
have the weight of support behind it at this point, and using GPL solves
a number of problems I've had in being compatible with some other "free"
projects.
With the new ASL, it is allowed that ASF software is (re)packaged as a 
new open source project which is distributed under the GPL. Adding onto 
that, we regard the new ASL as being compatible with the GPL, while the 
FSF decided the contrary, without tangible reason: a matter of policy 
rather than legal differences, so it seems. So we have a bit of an issue 
with the GPL, to put it frankly.
Ah, more of Mr. Stallman's pedantry -- one of the reasons I resisted the
GPL for many years.
The license change of Jisp is most likely going to force us to replace 
Jisp by some other dbm. Other ASF projects which use Jisp will be forced 
to do the same. This is bad since we, apart from some small issues, were 
quite happy with Jisp and had no technical reason to move away. And 
since we are a volunteer orga

Re: The (new) Jisp 3.0 license

On 20 Feb 2004, at 17:51, Scott Robert Ladd wrote:

Steven Noels wrote:
We kindly ask you to consider these issues and get back to us at your 
convenience. The Cocoon PMC list ([EMAIL PROTECTED]) is private 
to PMC members only, if you would like to make a public comment feel 
free to do that on [EMAIL PROTECTED]
I've CC'd this to the list above; I'd appreciate it if any discussion 
would be CC'd back to me as well. Please read the entire message 
before commenting; it may start out sounding a bit negative, but in 
the end, I think you get what you want. :)
Thanks for stepping up in this public forum, Scott.

I need to go home now to take a look at my two boys, single daughter, 
single wife and no critters, but I will get back to you ASAP using this 
list with you on CC.


--
Steven Noelshttp://outerthought.org/
Outerthought - Open Source Java & XMLAn Orixo Member
Read my weblog athttp://blogs.cocoondev.org/stevenn/
stevenn at outerthought.orgstevenn at apache.org

Re: The (new) Jisp 3.0 license

Steven Noels wrote:
We kindly ask you to consider these issues and get back to us at your 
convenience. The Cocoon PMC list ([EMAIL PROTECTED]) is private to 
PMC members only, if you would like to make a public comment feel free 
to do that on [EMAIL PROTECTED]
I've CC'd this to the list above; I'd appreciate it if any discussion 
would be CC'd back to me as well. Please read the entire message before 
commenting; it may start out sounding a bit negative, but in the end, I 
think you get what you want. :)

let me first start with thanking you for a wonderful set of tools, which 
we have eagerly been using over the years.
I'm glad to know it was useful.

Cocoon has been using Jisp for quite some time, and it served us well as 
a migration from a filesystem based cache store. We have been hitting 
some Jisp bugs lately, which made us looking forward to migrate to the 
upcoming 3.0 version.
Older versions of Jisp did, indeed, have some problems; initially, Jisp 
was an algorithmic demonstration for one of my books, and it was not 
intended for "real world use".

Given the number of people who have used Jisp, I decided to refactor the 
package in a more professional form. Thus, Jisp 3.0, which is based on 
the code I use in commercial products.

Quite to our surprise however, we just found out that Jisp 3 now comes 
with the MySQL-like combo-license: GPL + commercial licensing possibility.

Of course, we cannot and will not have any say into which license you 
should use for your products, since it's your blood and tears that went 
into it. As open source developers however, we have some considerations 
about licenses which we'd like to confront you with.
I appreciate your talking to me about this. Please indulge me in a bit 
of background story, so you understand why I changed licenses.

I'm not a twenty-something college student who can afford to hack free 
software for free; I am an independent business man and forty-something 
father of three. My first "free" products appeared back in the 1980s, 
and I've written several programming books, so I'm a long-time believer 
in sharing code with the community. A society can not progress if 
knowledge is kept secret.

As such, I originally released Jisp under a libpng-style license.

In the last year or so, I've been hounded by people who "demand" support 
for Jisp -- including those who use Jisp via Apache. While I politely 
inform them that my time isn't free, the frequency of such requests is 
increasing. When I request some form of renumeration for providing 
support, they act as if I've suggested concourse with a devil.

My frustration lies with people who use and praise Jisp without any 
feeling of obligation to me, other than a hearty handshake (at best). My 
frustration grows when I know that people are profiting from my code, 
without so much as an acknowledgement.

I am not a greedy soul, and excess wealth is not one of my goals in 
life; however, I have three daughters, a wife, and a managerie of 
critters that need my financial support. These are dark economic times 
in our industry; survival for an independent consultant (like me) is 
tenuous even in the best of times.

For me, the change in license (to the GPL) is an attempt to find equity 
in people's use of my code. "Fairness" is, of course, a matter of 
perception. In my case, I have received nothing from the Apache project 
beyond the "hearty handshake" level. It's not that I don't want Apache 
to use Jisp -- it's that I know people are paid to work on Apache, and 
that many people make money selling Apache support and services.

It seems to me that those who find value in my creation should also be 
willing to reward me for my contribution.

The ASF policy is not to use, link or distribute software (components) 
which use the (L)GPL license, since these licenses prohibit the 
redistribution of ASF software linking, using or embedding such 
dependencies along the same terms as the Apache Software License (1 or 
2) allows.
I am not a fan of Mr. Stallman's religion; however, his license does 
have the weight of support behind it at this point, and using GPL solves 
a number of problems I've had in being compatible with some other "free" 
projects.

With the new ASL, it is allowed that ASF software is (re)packaged as a 
new open source project which is distributed under the GPL. Adding onto 
that, we regard the new ASL as being compatible with the GPL, while the 
FSF decided the contrary, without tangible reason: a matter of policy 
rather than legal differences, so it seems. So we have a bit of an issue 
with the GPL, to put it frankly.
Ah, more of Mr. Stallman's pedantry -- one of the reasons I resisted the 
GPL for many years.

The license change of Jisp is most likely going to force us to replace 
Jisp by some other dbm. Other ASF projects which use Jisp will be forced 
to do the same. This is bad since we, apart from some small issues, were 
quite happy with Jisp and had no technical reason to

Re: Xindice filer, Re: Jisp 3.0 moved to GPL licence

Torsten Curdt wrote:

Does that answer mean you think a XIndice persistent store 
implementation would be a good fit?  As you're involved heavily in 
both projects, you'd be the best to comment probably...




It would work. Take a look at the linked class, as well as at Filer 
interface. What's your opinion?


That's what I think. It would work. The issue in my mind is more to 
do with performance. Could it compete with jisp? Do you have any 
thoughts there Vadim?


Hm... A full blown XIndice because we want the persitant
store? Do you really think that's a good idea?


Actually, haven't we considered doing this anyway?  I have a dim memory 
of us wanting to put an xml fragment store in the core for some 
performance miracle...

I'd rather see this part ripped out of XIndice and put
into Avalon and share it that way instead of using
XIndice.


Sounds like Vadim already suggested a separate jar...

Geoff

Re: Xindice filer, Re: Jisp 3.0 moved to GPL licence

Does that answer mean you think a XIndice persistent store 
implementation would be a good fit?  As you're involved heavily in 
both projects, you'd be the best to comment probably...




It would work. Take a look at the linked class, as well as at Filer 
interface. What's your opinion?


That's what I think. It would work. The issue in my mind is more to do 
with performance. Could it compete with jisp? Do you have any thoughts 
there Vadim?
Hm... A full blown XIndice because we want the persitant
store? Do you really think that's a good idea?
I'd rather see this part ripped out of XIndice and put
into Avalon and share it that way instead of using
XIndice.
*shrug*
--
Torsten

RE: Xindice filer, Re: Jisp 3.0 moved to GPL licence


From: Vadim Gritsenko
 
> Upayavira wrote:
> 
> > Vadim Gritsenko wrote:
> >
> >> Geoff Howard wrote:
> >>
> >>> Vadim Gritsenko wrote:
> >>>
>  Xindice has a Filer abstraction, and there is BTreeFiler
>  implementation. It stores binary objects under an 
> arbitrary binary 
>  key, and keys are organized into the BTree for fast 
>  store/retrieval. See test for filer here:
> 
> 
>  
> http://cvs.apache.org/viewcvs.cgi/xml-xindice/java/tests/src/o
rg/ap
 ache/xindice/core/filer/FilerTestBase.java?view=auto


 Filer uses several RandomAccessFile descriptors to provide
 concurrent reads / writes to the file.
>>>
>>>
>>>
>>>
>>> Does that answer mean you think a XIndice persistent store
>>> implementation would be a good fit?  As you're involved heavily in 
>>> both projects, you'd be the best to comment probably...
>>
>>
>>
>>
>> It would work. Take a look at the linked class, as well as at Filer
>> interface. What's your opinion?
>
>
> That's what I think. It would work. The issue in my mind is more to do
> with performance. Could it compete with jisp? Do you have any thoughts

> there Vadim?


> Well, this I don't know (and I never looked into Jisp source code - so

> can't comment on its workings). Do you want to test it?
> 
> What I know is that filer is ~ 100kb of Java source code, so it would
be 
> easier to find / fix bugs, if any. In compiled form it will be even 
> less. To address Reinhard's concern, it could be packaged into
separate 
> jar, of Jisp size or so. So, if Jisp sticks to GPL, we still would
have 
> several options - from JCS to Xindice Filer.

This sounds good to me :-)
 
> PS What's JCS size / performance / etc?

*If* it is necessary to switch I think we need some tests ...

--
Reinhard

Re: Xindice filer, Re: Jisp 3.0 moved to GPL licence

2004-02-20 Thread Vadim Gritsenko

Upayavira wrote:

Vadim Gritsenko wrote:

Geoff Howard wrote:

Vadim Gritsenko wrote:

Xindice has a Filer abstraction, and there is BTreeFiler
implementation. It stores binary objects under an arbitrary binary
key, and keys are organized into the BTree for fast
store/retrieval. See test for filer here:

http://cvs.apache.org/viewcvs.cgi/xml-xindice/java/tests/src/org/apache/xindice/core/filer/FilerTestBase.java?view=auto

Filer uses several RandomAccessFile descriptors to provide
concurrent reads / writes to the file.

Does that answer mean you think a XIndice persistent store
implementation would be a good fit? As you're involved heavily in
both projects, you'd be the best to comment probably...

It would work. Take a look at the linked class, as well as at Filer
interface. What's your opinion?

That's what I think. It would work. The issue in my mind is more to do
with performance. Could it compete with jisp? Do you have any thoughts
there Vadim?

Well, this I don't know (and I never looked into Jisp source code - so
can't comment on its workings). Do you want to test it?

What I know is that filer is ~ 100kb of Java source code, so it would be
easier to find / fix bugs, if any. In compiled form it will be even
less. To address Reinhard's concern, it could be packaged into separate
jar, of Jisp size or so. So, if Jisp sticks to GPL, we still would have
several options - from JCS to Xindice Filer.

PS What's JCS size / performance / etc?

Vadim

RE: Xindice filer, Re: Jisp 3.0 moved to GPL licence


From: Vadim Gritsenko [mailto:[EMAIL PROTECTED] 
 
> Geoff Howard wrote:
> 
> > Vadim Gritsenko wrote:
> >
> >> Upayavira wrote:
> >>
> >>> Antonio Gallardo wrote:
> >>>
>  Maybe this is a crazy idea but: Is posible to replace jisp with 
>  Apache Xindice? Mainly because I have concerns for another ugly 
>  move (as jisp
>  did) if we choose a solution from a 3rd party again. If 
> we use Apache
>  Xindice I think this cannot happen again.
> 
> >>> Could do. How efficient is XIndice? It would need to be pretty
> >>> efficient on binary data too, as that is our primary use case.
> >>
> >>
> >>
> >> Xindice has a Filer abstraction, and there is BTreeFiler
> >> implementation. It stores binary objects under an arbitrary binary 
> >> key, and keys are organized into the BTree for fast 
> store/retrieval. 
> >> See test for filer here:
> >>
> >>
> >> 
> http://cvs.apache.org/viewcvs.cgi/xml->
xindice/java/tests/src/org/apac
> >> he/xindice/core/filer/FilerTestBase.java?view=auto
> >>
> >>
> >> Filer uses several RandomAccessFile descriptors to provide 
> concurrent
> >> reads / writes to the file.
> >
> >
> >
> > Does that answer mean you think a XIndice persistent store
> > implementation would be a good fit?  As you're involved heavily in 
> > both projects, you'd be the best to comment probably...
> 
> 
> It would work. Take a look at the linked class, as well as at Filer 
> interface. What's your opinion?

Using XMLDB would mean another ~500KB more in Cocoon core ... :-(

--
Reinhard

Re: Xindice filer, Re: Jisp 3.0 moved to GPL licence

2004-02-20 Thread Upayavira

Vadim Gritsenko wrote:

Geoff Howard wrote:

Vadim Gritsenko wrote:

Upayavira wrote:

Antonio Gallardo wrote:

Maybe this is a crazy idea but: Is posible to replace jisp with 
Apache
Xindice? Mainly because I have concerns for another ugly move (as 
jisp
did) if we choose a solution from a 3rd party again. If we use Apache
Xindice I think this cannot happen again.

Could do. How efficient is XIndice? It would need to be pretty 
efficient on binary data too, as that is our primary use case.




Xindice has a Filer abstraction, and there is BTreeFiler 
implementation. It stores binary objects under an arbitrary binary 
key, and keys are organized into the BTree for fast store/retrieval. 
See test for filer here:

   
http://cvs.apache.org/viewcvs.cgi/xml-xindice/java/tests/src/org/apache/xindice/core/filer/FilerTestBase.java?view=auto 

Filer uses several RandomAccessFile descriptors to provide 
concurrent reads / writes to the file.




Does that answer mean you think a XIndice persistent store 
implementation would be a good fit?  As you're involved heavily in 
both projects, you'd be the best to comment probably...


It would work. Take a look at the linked class, as well as at Filer 
interface. What's your opinion?
That's what I think. It would work. The issue in my mind is more to do 
with performance. Could it compete with jisp? Do you have any thoughts 
there Vadim?

Upayavira

Re: Xindice filer, Re: Jisp 3.0 moved to GPL licence

2004-02-20 Thread Vadim Gritsenko

Geoff Howard wrote:

Vadim Gritsenko wrote:

Upayavira wrote:

Antonio Gallardo wrote:

Maybe this is a crazy idea but: Is posible to replace jisp with Apache
Xindice? Mainly because I have concerns for another ugly move (as jisp
did) if we choose a solution from a 3rd party again. If we use Apache
Xindice I think this cannot happen again.
Could do. How efficient is XIndice? It would need to be pretty 
efficient on binary data too, as that is our primary use case.


Xindice has a Filer abstraction, and there is BTreeFiler 
implementation. It stores binary objects under an arbitrary binary 
key, and keys are organized into the BTree for fast store/retrieval. 
See test for filer here:

   
http://cvs.apache.org/viewcvs.cgi/xml-xindice/java/tests/src/org/apache/xindice/core/filer/FilerTestBase.java?view=auto 

Filer uses several RandomAccessFile descriptors to provide concurrent 
reads / writes to the file.


Does that answer mean you think a XIndice persistent store 
implementation would be a good fit?  As you're involved heavily in 
both projects, you'd be the best to comment probably...


It would work. Take a look at the linked class, as well as at Filer 
interface. What's your opinion?

Vadim

RE: Jisp 3.0 moved to GPL licence



> -Original Message-
> From: Stefano Mazzocchi [mailto:[EMAIL PROTECTED]
> Sent: Friday, February 20, 2004 3:55 PM
> To: [EMAIL PROTECTED]
> Subject: Re: Jisp 3.0 moved to GPL licence
> 
> 
> Steven Noels wrote:
> 
> > On 20 Feb 2004, at 12:52, Torsten Curdt wrote:
> > 
> >> hm... what do they do?
> >> maybe we should talk to them and coordinate efforts?
> > 
> > 
> > +1 - seems like the best plan to go forward without over-reacting.
> > 
> > FWIW: I think we should actively weed out one-man-effort
> dependencies.
> > People have the freedom to change their mind, but we shouldn't be a
> > victim of that.
> 
> +1000
> 
> But keep in mind that we are never locked in as we always
> have the right 
> to fork. So, let's keep reasonable and not panic.

+1
As Antonio already said it's no emergency becaue we *do use* a former
version under a 'friendly' licency and we have time to evaluate all
possible options without getting panic. For now this is

 - forking Jisp 2.6
 - JCS
 - OSCache from Opensymphonie

Some mentioned RDBMS systems or XMLDBs - of course an option but I don't
think they are designed for our caching needs (but maybe I'm wrong
here).

--
Reinhard

Re: Jisp 3.0 moved to GPL licence

Reinhard Poetz wrote:

As our store depends on Jisp - what does this mean for us? IMO we have
to look for a replacement. Any ideas/hints?
 

This is not stricly true, is it?  We use by default the JISP persistent 
store implementation, but it's a config entry away to switch to the file 
system store or any other.  IIRC from the discussion at the time (Vadim 
did this good work, no?) there was a performance improvement but I 
haven't yet located the discussion to determine how much.

Geoff

Re: Xindice filer, Re: Jisp 3.0 moved to GPL licence

Vadim Gritsenko wrote:

Upayavira wrote:

Antonio Gallardo wrote:

Maybe this is a crazy idea but: Is posible to replace jisp with Apache
Xindice? Mainly because I have concerns for another ugly move (as jisp
did) if we choose a solution from a 3rd party again. If we use Apache
Xindice I think this cannot happen again.
Could do. How efficient is XIndice? It would need to be pretty 
efficient on binary data too, as that is our primary use case.


Xindice has a Filer abstraction, and there is BTreeFiler 
implementation. It stores binary objects under an arbitrary binary 
key, and keys are organized into the BTree for fast store/retrieval. 
See test for filer here:

   
http://cvs.apache.org/viewcvs.cgi/xml-xindice/java/tests/src/org/apache/xindice/core/filer/FilerTestBase.java?view=auto 

Filer uses several RandomAccessFile descriptors to provide concurrent 
reads / writes to the file.


Does that answer mean you think a XIndice persistent store 
implementation would be a good fit?  As you're involved heavily in both 
projects, you'd be the best to comment probably...

Geoff

Xindice filer, Re: Jisp 3.0 moved to GPL licence

2004-02-20 Thread Vadim Gritsenko

Upayavira wrote:

Antonio Gallardo wrote:

Maybe this is a crazy idea but: Is posible to replace jisp with Apache
Xindice? Mainly because I have concerns for another ugly move (as jisp
did) if we choose a solution from a 3rd party again. If we use Apache
Xindice I think this cannot happen again.
Could do. How efficient is XIndice? It would need to be pretty 
efficient on binary data too, as that is our primary use case.


Xindice has a Filer abstraction, and there is BTreeFiler implementation. 
It stores binary objects under an arbitrary binary key, and keys are 
organized into the BTree for fast store/retrieval. See test for filer here:

   
http://cvs.apache.org/viewcvs.cgi/xml-xindice/java/tests/src/org/apache/xindice/core/filer/FilerTestBase.java?view=auto

Filer uses several RandomAccessFile descriptors to provide concurrent 
reads / writes to the file.

Vadim

Re: Jisp 3.0 moved to GPL licence

Steven Noels wrote:

On 20 Feb 2004, at 12:52, Torsten Curdt wrote:

hm... what do they do?
maybe we should talk to them and coordinate efforts?


+1 - seems like the best plan to go forward without over-reacting.

FWIW: I think we should actively weed out one-man-effort dependencies. 
People have the freedom to change their mind, but we shouldn't be a 
victim of that.
+1000

But keep in mind that we are never locked in as we always have the right 
to fork. So, let's keep reasonable and not panic.

--
Stefano.


smime.p7s
Description: S/MIME Cryptographic Signature

Re: Jisp 3.0 moved to GPL licence

Antonio Gallardo wrote:

Torsten Curdt dijo:

Antonio Gallardo wrote:


Carsten Ziegeler dijo:


What about JCS?

http://jakarta.apache.org/turbine/jcs/index.html


jisp is part of the dependencies. :-(
See:
http://jakarta.apache.org/turbine/jcs/dependencies.html
hm... what do they do?
maybe we should talk to them and coordinate efforts?
Hi Torsten!

Long time don't saw a post from you. Welcome back!

It is a good idea. I think this is the best we can do. Also I think we
will need a database to replace jisp, I thought another good candidate can
be hsqldb (already on out CVS) or - Axion
http://axion.tigris.org/quickstart.html
WDYT?

Best Regards,

Antonio Gallardo
Antonio, Jisp is not a database! is a binary object storage system. The 
closest thing I can think of is BerkeleyDB or, in short, a persistent 
hashtable.

That's all we need.

--
Stefano.


smime.p7s
Description: S/MIME Cryptographic Signature

Re: Jisp 3.0 moved to GPL licence

On 20 Feb 2004, at 15:44, Stefano Mazzocchi wrote:

We have the right to fork, keep this in mind.
I wouldn't advocate this as a matter of practice. But your point is 
perfectly valid.


--
Steven Noelshttp://outerthought.org/
Outerthought - Open Source Java & XMLAn Orixo Member
Read my weblog athttp://blogs.cocoondev.org/stevenn/
stevenn at outerthought.orgstevenn at apache.org

Re: Jisp 3.0 moved to GPL licence

Upayavira wrote:

Antonio Gallardo wrote:

Reinhard Poetz dijo:
 

Am 14:01 10.02.2004 -0500 schrieb Steve Krulewitz:


Does anyone know what happened to the jisp website?  The old URL
http://www.coyotegulch.com/jisp/index.html sends you to an
  
invalid link.

It's back up now: http://www.coyotegulch.com/jisp/
I never went there before it went down, but it now has a
version 3.0.0. That version isn't under the old license
anymore but GPLed (or commercial
for 2500$).
Old versions I can't find there..
gunnar

ps: sorry if you get this mail twice, steve. small mistake by me.

--
G. Brand - interface:projects GmbH

As our store depends on Jisp - what does this mean for us? IMO we have
to look for a replacement. Any ideas/hints?
  
AFAIK, we must move away from jisp. That was an ugly move! :-(

Maybe this is a crazy idea but: Is posible to replace jisp with Apache
Xindice? Mainly because I have concerns for another ugly move (as jisp
did) if we choose a solution from a 3rd party again. If we use Apache
Xindice I think this cannot happen again.
 

Could do. How efficient is XIndice? It would need to be pretty efficient 
on binary data too, as that is our primary use case.

With the new embedded mode driver for XIndice, it wouldn't be that hard 
to implement, so long as it is efficient enough.
I personally think it's a very bad idea. XIndice is not designed for 
these things. Carsten's suggestion seems *a lot* better!

--
Stefano.


smime.p7s
Description: S/MIME Cryptographic Signature

Re: Jisp 3.0 moved to GPL licence

Reinhard Poetz wrote:

Am 14:01 10.02.2004 -0500 schrieb Steve Krulewitz:

Does anyone know what happened to the jisp website?  The old URL
http://www.coyotegulch.com/jisp/index.html sends you to an 
invalid link.

It's back up now: http://www.coyotegulch.com/jisp/
I never went there before it went down, but it now has a 
version 3.0.0. That version isn't under the old license 
anymore but GPLed (or commercial 
for 2500$).
Old versions I can't find there..

gunnar

ps: sorry if you get this mail twice, steve. small mistake by me.

--
G. Brand - interface:projects GmbH


As our store depends on Jisp - what does this mean for us? IMO we have
to look for a replacement. Any ideas/hints?
We have the right to fork, keep this in mind.

--
Stefano.


smime.p7s
Description: S/MIME Cryptographic Signature

RE: LDAPTransformer

2004-02-20 Thread DAYD3

I apologize DEEPLY for the horrible formatting of my previous note.  I'm 
on Lotus Notes here at work, and it doesn't seem to cooperate.
Let me try again:

Hi,

I have 2 "features" that I think would be of great value to the 
LDAPTransformer. I have already begun working on a patch to implement 
both.

1) In order to use SSL for LDAP (ldaps) while using a JDK prior to 1.4.2, 
I suggest that an additional optional attribute be added - 
"securityprotocol".

Please reference: 

http://java.sun.com/products/jndi/tutorial/ldap/security/ssl.html

This outlines 2 methods of making an ssl connects, one using the "ldaps" 
protocol, and the other using the "SSL Socket Property", which is not 
currently available in the LDAPTransformer.  Also please note on the page, 
the section on "Using the LDAPS URL" states explicitly that "LDAPS URLs 
are supported only in Java 2 SDK, v1.4.2 and later releases."

If this attribute were set, then it would get set in the env, i.e.

if (securityprotocol != null) {
env.put(Context.SECURITY_PROTOCOL, securityprotocol);
}

2)  Each entry returned needs to have its DN included.  You can not 
reliably reconstruct the dn from the searchbase and an entry's attribute. 
In order to obtain the true dn, at the code level you have to do something 
like:

String dn = sr.getName() + "," + ldapSearchBase;

What I'm still pondering is whether the dn should simply be returned as an 
another attribute (i.e. a sub-element of row-element), or if it should be 
elevated to an attribute of the row-element.

Comments on both of these are more than welcome.  And PLEASE let me know 
if anyone has already engaged either or both of these issues.  I did 
research both on Bugzilla and the cocoon-dev archives, and did not find 
anything addressing either of these.

Once I have my patch complete, I will submit it via Bugzilla.

Regards,
David Day

LDAPTransformer

2004-02-20 Thread DAYD3

Hi,

I have 2 "features" that I think would be of great value to the 
LDAPTransformer.
I have already begun working on a patch to implement both.

1) In order to use SSL for LDAP (ldaps) while using a JDK prior to 1.4.2, 
I 
suggest that an additional optional attribute be added - 
"securityprotocol".

Please reference: 

http://java.sun.com/products/jndi/tutorial/ldap/security/ssl.html

This outlines 2 methods of making an ssl connects, one using the "ldaps" 
protocol, and the other using the "SSL Socket Property", which is not 
currently 
available in the LDAPTransformer.  Also please note on the page, the 
section on 
"Using the LDAPS URL" states explicitly that "LDAPS URLs are supported 
only in 
Java 2 SDK, v1.4.2 and later releases."

If this attribute were set, then it would get set in the env, i.e.

if (securityprotocol != null) {
env.put(Context.SECURITY_PROTOCOL, securityprotocol);
}

12345678901234567890123456789012345678901234567890123456789012345678901234567890
2)  Each entry returned needs to have its DN included.  You can not 
reliably
reconstruct the dn from the searchbase and an entry's attribute.  In order 
to 
obtain the true dn, at the code level you have to do something like:

String dn = sr.getName() + "," + ldapSearchBase;

What I'm still pondering is whether the dn should simply be returned as an 

another attribute (i.e. a sub-element of row-element), or if it should be 
elevated to an attribute of the row-element.

Comments on both of these are more than welcome.  And PLEASE let me know 
if 
anyone has already engaged either or both of these issues.  I did research 
both 
on Bugzilla and the cocoon-dev archives, and did not find anything 
addressing 
either of these.

Once I have my patch complete, I will submit it via Bugzilla.

Regards,
David Day

Re: Jisp 3.0 moved to GPL licence

On 20 Feb 2004, at 13:36, Steven Noels wrote:

I will contact them.
Just done so. As I wanted to give Scott the possibility to vent about 
this issue in private, I only copied the PMC list. Hope you understand. 
Of course, we'll get back with the conclusion on the dev list as well.


--
Steven Noelshttp://outerthought.org/
Outerthought - Open Source Java & XMLAn Orixo Member
Read my weblog athttp://blogs.cocoondev.org/stevenn/
stevenn at outerthought.orgstevenn at apache.org

Re: Jisp 3.0 moved to GPL licence

2004-02-20 Thread Bertrand Delacretaz

Le Vendredi, 20 fév 2004, à 14:16 Europe/Zurich, Antonio Gallardo a 
écrit :

...Good news!
I already read some info of JCS and looks like the jisp usage is 
optional:
Cool - moving to an ASF thing would certainly be good.

-Bertrand

Re: Jisp 3.0 moved to GPL licence

Steven Noels dijo:
> On 20 Feb 2004, at 12:52, Torsten Curdt wrote:
>
>> hm... what do they do?
>> maybe we should talk to them and coordinate efforts?
>
> +1 - seems like the best plan to go forward without over-reacting.

Did we need a VOTE?

Good news!
I already read some info of JCS and looks like the jisp usage is optional:

http://jakarta.apache.org/turbine/jcs/JCSPackageInformation.html

In this document under "Auxiliary caches":

o.a.s.j.auxiliary.disk
-
The primary disk auxiliary. Objects are serialized to a file on disk. This
implementation uses memory keys and performs quite well. Recomended for
most cases.

o.a.s.j.auxiliary.disk.jisp
---
Disk cache implemented with the Java Indexed Serialization Package , which
allows serialization of objects to B-Tree indexed tables on disk. This is
quite slow currently.

o.a.s.j.auxiliary.disk.hsql
---
A disk cache using Hypersonic SQL to serialize the contained objects.

Hope this help.

Best Regards,

Antonio Gallardo

Re: Jisp 3.0 moved to GPL licence

Hi Torsten!

Long time don't saw a post from you. Welcome back!
Geez ...that's nice welcome :D
(I was only lurking for few days)
God, I love this people here :)

It is a good idea. I think this is the best we can do. Also I think we
will need a database to replace jisp, I thought another good candidate can
be hsqldb (already on out CVS) or - Axion
http://axion.tigris.org/quickstart.html
(Any particular reason to switch here, too?)

I don't have a very good overview
about what's out there but I like
JCS for several reasons:
a) it's an apache project
b) caching is it's primary project goal
c) server clustering and remote synchronization
Of course they have the same problem with
the licence change (in the future) ...but
they *have* to deal with it too. Why not help
them and give us some more features at the
same time? BUT!
Unfortunately looking at the changelog
and the list of project members the
project seems to be... well...
not very active :-/

Hm...
--
Torsten

Re: Jisp 3.0 moved to GPL licence

Steven Noels dijo:
> On 20 Feb 2004, at 12:52, Torsten Curdt wrote:
>
>> hm... what do they do?
>> maybe we should talk to them and coordinate efforts?
>
> +1 - seems like the best plan to go forward without over-reacting.

I will try to contact the Apache JCS community. Seems like the development
stalled since 2003-08-21:
http://jakarta.apache.org/turbine/jcs/changelog-report.html

There are 2 committers:

http://jakarta.apache.org/turbine/jcs/team-list.html

Looks like we will need to give them a hand. :-D

Best Regards,

Antonio Gallardo

Re: Jisp 3.0 moved to GPL licence

On 20 Feb 2004, at 12:52, Torsten Curdt wrote:

hm... what do they do?
maybe we should talk to them and coordinate efforts?
+1 - seems like the best plan to go forward without over-reacting.

FWIW: I think we should actively weed out one-man-effort dependencies. 
People have the freedom to change their mind, but we shouldn't be a 
victim of that.


--
Steven Noelshttp://outerthought.org/
Outerthought - Open Source Java & XMLAn Orixo Member
Read my weblog athttp://blogs.cocoondev.org/stevenn/
stevenn at outerthought.orgstevenn at apache.org

Re: Jisp 3.0 moved to GPL licence

On 20 Feb 2004, at 13:25, Bertrand Delacretaz wrote:

Hmm..I'm not too excited by http://www.coyotegulch.com/jisp/index.html 
mentioning "the Apache project" as a reference user of their software 
and at the same time changing the license in a way that will prevent 
us from using Jisp in the future.
I will contact them.


--
Steven Noelshttp://outerthought.org/
Outerthought - Open Source Java & XMLAn Orixo Member
Read my weblog athttp://blogs.cocoondev.org/stevenn/
stevenn at outerthought.orgstevenn at apache.org

Re: Jisp 3.0 moved to GPL licence

2004-02-20 Thread Bertrand Delacretaz

Le Vendredi, 20 fév 2004, à 11:52 Europe/Zurich, Reinhard Poetz a écrit 
:
...As our store depends on Jisp - what does this mean for us? IMO we 
have
to look for a replacement. Any ideas/hints?
Hmm..I'm not too excited by http://www.coyotegulch.com/jisp/index.html 
mentioning "the Apache project" as a reference user of their software 
and at the same time changing the license in a way that will prevent us 
from using Jisp in the future.

Maybe it is still possible for them change their mind about this?
Cocoon can certainly be a good proof of Jisp's quality, if I were them 
I'd try to do something to make it possible for the ASF to continue 
using Jisp.

Anyone from CoyoteGulch listening here?

-Bertrand

Re: Jisp 3.0 moved to GPL licence

Torsten Curdt dijo:
> Antonio Gallardo wrote:
>
>> Carsten Ziegeler dijo:
>>
>>>What about JCS?
>>>
>>>http://jakarta.apache.org/turbine/jcs/index.html
>>
>>
>> jisp is part of the dependencies. :-(
>> See:
>> http://jakarta.apache.org/turbine/jcs/dependencies.html
>
> hm... what do they do?
> maybe we should talk to them and coordinate efforts?
Hi Torsten!

Long time don't saw a post from you. Welcome back!

It is a good idea. I think this is the best we can do. Also I think we
will need a database to replace jisp, I thought another good candidate can
be hsqldb (already on out CVS) or - Axion
http://axion.tigris.org/quickstart.html

WDYT?

Best Regards,

Antonio Gallardo

RE: Jisp 3.0 moved to GPL licence


From: Gunnar Brand [mailto:[EMAIL PROTECTED] 

> Am 05:39 20.02.2004 -0600 schrieb Antonio Gallardo:
> >Carsten Ziegeler dijo:
> > > What about JCS?
> > >
> > > http://jakarta.apache.org/turbine/jcs/index.html
> >
> >jisp is part of the dependencies. :-(
> >See:
> >http://jakarta.apache.org/turbine/jcs/dependencies.html
> >
> >Best Regards,
> >
> >Antonio Gallardo
> 
> I wrote that version 3.0.0 is GPLed.
> 
> If you read 
> http://www.coyotegulch.com/jisp/license> _policy.html 
> carefully, 
> you will notice that it says
> "Older Versions
> Older versions of Jisp use different licenses; those licenses 
> only apply to 
> Jisp version 2.6 and earlier. Jisp 3.0.0 is licensed under 
> new terms, and 
> old terms from the previous version may not be applied to 
> Jisp version 
> 3.0.0 or later."
> 
> IMHO as long as the old jisp (v2.6 or lower) is used, nothing 
> needs to 
> change. No updates anymore, though.

That's right for the near future but this also means we don't get any
upgrades/patches and this is IMO no good thing. Hence I think that we
should find a replacement.

--
Reinhard

RE: Jisp 3.0 moved to GPL licence

Hi Gunnar:

We understand this is not an emergency. Instead, this is an alert that we
need to change. Of course we will evaluate the best replacement and do
switch as quick as we can. AFAIK, almost every piece of software is far to
be perfect and it is improved or have bug fixes on veery new release. In
that way, it is clear to me that we cannot continue the use of jisp in
Cocoon.

Best Regards,

Antonio Gallardo.

Gunnar Brand dijo:
> Am 05:39 20.02.2004 -0600 schrieb Antonio Gallardo:
>>Carsten Ziegeler dijo:
>> > What about JCS?
>> >
>> > http://jakarta.apache.org/turbine/jcs/index.html
>>
>>jisp is part of the dependencies. :-(
>>See:
>>http://jakarta.apache.org/turbine/jcs/dependencies.html
>>
>>Best Regards,
>>
>>Antonio Gallardo
>
> I wrote that version 3.0.0 is GPLed.
>
> If you read http://www.coyotegulch.com/jisp/license_policy.html carefully,
> you will notice that it says
> "Older Versions
> Older versions of Jisp use different licenses; those licenses only apply
> to
> Jisp version 2.6 and earlier. Jisp 3.0.0 is licensed under new terms, and
> old terms from the previous version may not be applied to Jisp version
> 3.0.0 or later."
>
> IMHO as long as the old jisp (v2.6 or lower) is used, nothing needs to
> change. No updates anymore, though.
>
> gunnar
>
>
>
> --
> G. Brand - interface:projects GmbH
> Tolkewitzer Strasse 49
> D-01277 Dresden
> mail:   [EMAIL PROTECTED]
> tel:++49-351-3 18 09 - 41
>
> Ein Unternehmen der interface:business-Gruppe
>

Re: Jisp 3.0 moved to GPL licence

Antonio Gallardo wrote:

Carsten Ziegeler dijo:

What about JCS?

http://jakarta.apache.org/turbine/jcs/index.html


jisp is part of the dependencies. :-(
See:
http://jakarta.apache.org/turbine/jcs/dependencies.html
hm... what do they do?
maybe we should talk to them and coordinate efforts?
--
Torsten

Re: Jisp 3.0 moved to GPL licence

2004-02-20 Thread Ugo Cei

Carsten Ziegeler wrote:
What about JCS?

http://jakarta.apache.org/turbine/jcs/index.html

Carsten 
I know that Hibernate moved away from JCS because of some bugs, but I 
cannot recall exactly what problems it had.

Another candidate might be OSCache [1].

	Ugo

[1]:

RE: Jisp 3.0 moved to GPL licence

2004-02-20 Thread Gunnar Brand

Am 05:39 20.02.2004 -0600 schrieb Antonio Gallardo:
Carsten Ziegeler dijo:
> What about JCS?
>
> http://jakarta.apache.org/turbine/jcs/index.html
jisp is part of the dependencies. :-(
See:
http://jakarta.apache.org/turbine/jcs/dependencies.html
Best Regards,

Antonio Gallardo
I wrote that version 3.0.0 is GPLed.

If you read http://www.coyotegulch.com/jisp/license_policy.html carefully, 
you will notice that it says
"Older Versions
Older versions of Jisp use different licenses; those licenses only apply to 
Jisp version 2.6 and earlier. Jisp 3.0.0 is licensed under new terms, and 
old terms from the previous version may not be applied to Jisp version 
3.0.0 or later."

IMHO as long as the old jisp (v2.6 or lower) is used, nothing needs to 
change. No updates anymore, though.

gunnar



--
G. Brand - interface:projects GmbH
Tolkewitzer Strasse 49
D-01277 Dresden
mail:   [EMAIL PROTECTED]
tel:++49-351-3 18 09 - 41
Ein Unternehmen der interface:business-Gruppe

Re: Jisp 3.0 moved to GPL licence

2004-02-20 Thread Upayavira

Antonio Gallardo wrote:

Reinhard Poetz dijo:
 

Am 14:01 10.02.2004 -0500 schrieb Steve Krulewitz:
 

Does anyone know what happened to the jisp website?  The old URL
http://www.coyotegulch.com/jisp/index.html sends you to an
   

invalid link.

It's back up now: http://www.coyotegulch.com/jisp/
I never went there before it went down, but it now has a
version 3.0.0. That version isn't under the old license
anymore but GPLed (or commercial
for 2500$).
Old versions I can't find there..
gunnar

ps: sorry if you get this mail twice, steve. small mistake by me.

--
G. Brand - interface:projects GmbH
 

As our store depends on Jisp - what does this mean for us? IMO we have
to look for a replacement. Any ideas/hints?
   

AFAIK, we must move away from jisp. That was an ugly move! :-(

Maybe this is a crazy idea but: Is posible to replace jisp with Apache
Xindice? Mainly because I have concerns for another ugly move (as jisp
did) if we choose a solution from a 3rd party again. If we use Apache
Xindice I think this cannot happen again.
 

Could do. How efficient is XIndice? It would need to be pretty efficient 
on binary data too, as that is our primary use case.

With the new embedded mode driver for XIndice, it wouldn't be that hard 
to implement, so long as it is efficient enough.

Regards, Upayavira

RE: Jisp 3.0 moved to GPL licence

Carsten Ziegeler dijo:
> What about JCS?
>
> http://jakarta.apache.org/turbine/jcs/index.html

jisp is part of the dependencies. :-(
See:
http://jakarta.apache.org/turbine/jcs/dependencies.html

Best Regards,

Antonio Gallardo

RE: Jisp 3.0 moved to GPL licence

Carsten Ziegeler dijo:
> What about JCS?
>
> http://jakarta.apache.org/turbine/jcs/index.html

Thanks Carsten!

Looks like the right tool for the right work! :-DDD

In short, here is my: +1

Best Regards,

Antonio Gallardo

DO NOT REPLY [Bug 27061] - CSV transformation fail on successive commas as produced by Excell

2004-02-20 Thread bugzilla

DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND 
INSERTED IN THE BUG DATABASE.

http://nagoya.apache.org/bugzilla/show_bug.cgi?id=27061

CSV transformation fail on successive commas as produced by Excell

[EMAIL PROTECTED] changed:

   What|Removed |Added

 Status|NEW |RESOLVED
 Resolution||FIXED



--- Additional Comments From [EMAIL PROTECTED]  2004-02-20 11:33 ---
Updated grammar by the grammar of the Chaperon repository. I added 
your sample to the examples of the Chaperon block, to test against it. (If
you not agree, please complain). Empty cell should now work.

Stephan Michels.

RE: Jisp 3.0 moved to GPL licence

What about JCS?

http://jakarta.apache.org/turbine/jcs/index.html

Carsten 

> -Original Message-
> From: Reinhard Poetz [mailto:[EMAIL PROTECTED] 
> Sent: Friday, February 20, 2004 11:53 AM
> To: [EMAIL PROTECTED]
> Subject: Jisp 3.0 moved to GPL licence
> 
> 
> > Am 14:01 10.02.2004 -0500 schrieb Steve Krulewitz:
> > >Does anyone know what happened to the jisp website?  The old URL 
> > >http://www.coyotegulch.com/jisp/index.html sends you to an
> > invalid link.
> > 
> > It's back up now: http://www.coyotegulch.com/jisp/ I never 
> went there 
> > before it went down, but it now has a version 3.0.0. That version 
> > isn't under the old license anymore but GPLed (or commercial for 
> > 2500$).
> > Old versions I can't find there..
> > 
> > gunnar
> > 
> > ps: sorry if you get this mail twice, steve. small mistake by me.
> > 
> > --
> > G. Brand - interface:projects GmbH
> 
> 
> As our store depends on Jisp - what does this mean for us? IMO we have
> to look for a replacement. Any ideas/hints?
> 
> --
> Reinhard
> 
>

Re: Jisp 3.0 moved to GPL licence

Reinhard Poetz dijo:
>
>> Am 14:01 10.02.2004 -0500 schrieb Steve Krulewitz:
>> >Does anyone know what happened to the jisp website?  The old URL
>> >http://www.coyotegulch.com/jisp/index.html sends you to an
>> invalid link.
>>
>> It's back up now: http://www.coyotegulch.com/jisp/
>> I never went there before it went down, but it now has a
>> version 3.0.0. That version isn't under the old license
>> anymore but GPLed (or commercial
>> for 2500$).
>> Old versions I can't find there..
>>
>> gunnar
>>
>> ps: sorry if you get this mail twice, steve. small mistake by me.
>>
>> --
>> G. Brand - interface:projects GmbH
>
>
> As our store depends on Jisp - what does this mean for us? IMO we have
> to look for a replacement. Any ideas/hints?
AFAIK, we must move away from jisp. That was an ugly move! :-(

Maybe this is a crazy idea but: Is posible to replace jisp with Apache
Xindice? Mainly because I have concerns for another ugly move (as jisp
did) if we choose a solution from a 3rd party again. If we use Apache
Xindice I think this cannot happen again.

WDYT?

Best Regards,

Antonio Gallardo.

Re: Jisp 3.0 moved to GPL licence

2004-02-20 Thread Upayavira

Joerg Heinicke wrote:

On 20.02.2004 11:52, Reinhard Poetz wrote:

I never went there before it went down, but it now has a version 
3.0.0. That version isn't under the old license anymore but GPLed 
(or commercial for 2500$).
Old versions I can't find there..


As our store depends on Jisp - what does this mean for us? IMO we have
to look for a replacement. Any ideas/hints?


On the thread about the announcement of Momento there was mentioned a 
"journaling file data structure" [1], Stefano asked if it is a 
replacment for JISP [2] and in Alan's opinion it might be possible [3].

As I don't know the requirements I can not evaluate it though.
I did think of this. But, it seems that Momento is all about XML, whilst 
JISP is a generic 'binary' store. Whilst we do cache XML, we also cache 
serialized content, i.e. binary data, and I would say that that is our 
primary use case. So, Momento might be able to do that, but it doesn't 
seem to be the main problem it is trying to sove.

Regards, Upayavira

[1] http://marc.theaimsgroup.com/?l=xml-cocoon-dev&m=107700015102265&w=4
[2] http://marc.theaimsgroup.com/?l=xml-cocoon-dev&m=107711295731173&w=4
[3] http://marc.theaimsgroup.com/?l=xml-cocoon-dev&m=107712768202660&w=4
[4] complete thread: 
http://marc.theaimsgroup.com/?t=10769599726&r=1&w=4

Re: Jisp 3.0 moved to GPL licence

On 20.02.2004 11:52, Reinhard Poetz wrote:

I never went there before it went down, but it now has a 
version 3.0.0. That version isn't under the old license 
anymore but GPLed (or commercial 
for 2500$).
Old versions I can't find there..


As our store depends on Jisp - what does this mean for us? IMO we have
to look for a replacement. Any ideas/hints?
On the thread about the announcement of Momento there was mentioned a 
"journaling file data structure" [1], Stefano asked if it is a 
replacment for JISP [2] and in Alan's opinion it might be possible [3].

As I don't know the requirements I can not evaluate it though.

Joerg

[1] http://marc.theaimsgroup.com/?l=xml-cocoon-dev&m=107700015102265&w=4
[2] http://marc.theaimsgroup.com/?l=xml-cocoon-dev&m=107711295731173&w=4
[3] http://marc.theaimsgroup.com/?l=xml-cocoon-dev&m=107712768202660&w=4
[4] complete thread: http://marc.theaimsgroup.com/?t=10769599726&r=1&w=4

Jisp 3.0 moved to GPL licence


> Am 14:01 10.02.2004 -0500 schrieb Steve Krulewitz:
> >Does anyone know what happened to the jisp website?  The old URL
> >http://www.coyotegulch.com/jisp/index.html sends you to an 
> invalid link.
> 
> It's back up now: http://www.coyotegulch.com/jisp/
> I never went there before it went down, but it now has a 
> version 3.0.0. That version isn't under the old license 
> anymore but GPLed (or commercial 
> for 2500$).
> Old versions I can't find there..
> 
> gunnar
> 
> ps: sorry if you get this mail twice, steve. small mistake by me.
> 
> -- 
> G. Brand - interface:projects GmbH


As our store depends on Jisp - what does this mean for us? IMO we have
to look for a replacement. Any ideas/hints?

--
Reinhard

Re: What happened to the jisp website?

2004-02-20 Thread Gunnar Brand

Am 14:01 10.02.2004 -0500 schrieb Steve Krulewitz:
Does anyone know what happened to the jisp website?  The old URL 
http://www.coyotegulch.com/jisp/index.html sends you to an invalid link.
It's back up now: http://www.coyotegulch.com/jisp/
I never went there before it went down, but it now has a version 3.0.0.
That version isn't under the old license anymore but GPLed (or commercial 
for 2500$).
Old versions I can't find there..

gunnar

ps: sorry if you get this mail twice, steve. small mistake by me.

--
G. Brand - interface:projects GmbH
Tolkewitzer Strasse 49
D-01277 Dresden
mail:   [EMAIL PROTECTED]
tel:++49-351-3 18 09 - 41
Ein Unternehmen der interface:business-Gruppe

RE: redirect handling, environment handling, etc.

Marco Rolappe wrote: 
> 
> > Unfortunately, an internal request/environment gets a copy of the 
> > parent object model, so it also gets the key in it's copy.
> > So, in this case startProcessing() is legal and *must* 
> overwrite the 
> > key stored in the object model.
> 
> but then I can't see the real point in protecting the 
> environment processing by a key. why not binding resource 
> looked up/used while processing/to be cleaned up to the 
> actual environment then?

Perhaps we are taking about different things here. The environment
implementation and especially the wrapper implementation didn't
allow it. For example, if you remove an attribute on the wrapper
(or the child env), then this attribute is removed at the
parent as well! This was done because of some flow stuff I think
but makes it really dificult to get it right.
Anyways, the basic idea is that the environment is independent
from the processing, so the environment should not know anything
about how the core organizes things.
The current implementation is not optional but is not that easy
changeable without loosing some compatibility. That's why we
rewrote all of that for 2.2.

> > In this aspect 2.2 is much better as only the key is stored in the 
> > object model, everything else is stored not reachable from 
> either the 
> > environment or the object model. So, it's much cleaner and safer.
> 
> with everything else you mean request, response, ... objects? 
> I guess I'm going to take a look... ;-)
> 
No, request etc. are of course still stored in the object model, but
some internal objects for processing the request are not part
of the object model any more.
I even hope to remove the processing key from the object model in 2.2.
I'm just waiting for some changes in the flow/redirector part and then
I will hopefully finally clean up everything.
I guess we have to rethink the forward/wrapper strategy in 2.2 as well.

Carsten

Re: [Vote] Move portlet environment into portal block

On 20 Feb 2004, at 09:41, Carsten Ziegeler wrote:

And perhaps such a move drives the block development...
Yep. Empathic +1.


--
Steven Noelshttp://outerthought.org/
Outerthought - Open Source Java & XMLAn Orixo Member
Read my weblog athttp://blogs.cocoondev.org/stevenn/
stevenn at outerthought.orgstevenn at apache.org

RE: [Vote] Move portlet environment into portal block

Nicola Ken Barozzi wrote: 

> Then there is no reason not to want to do this.
> 
> +1 to the change
> 
:)

Carsten

RE: [Vote] Move portlet environment into portal block