Home » Language IDEs » ServerTools (WTP) » Enough people to get started with
| Enough people to get started with [message #12833] |
Tue, 13 January 2004 11:56  |
Eclipse User |
|
|
|
I already offered my company to do some project management, if wanted, I
am willing to develop an initial roadmap for the project to develop
against.
Enough people want to get started with engineering and coding elements,
already other options are being searched to get it going (such as a
project on sourceforge).
Others wanted to put some money aside for this project to support
development or have people who can help with engineering and maintaining
the projects.
Where can we contact the Eclipse board so we can get started and discuss
some of the projects in mind?
Yours,
Sjoerd van Leent
Smart and Logic V.O.F., the Netherlands
|
|
| |
| Re: Enough people to get started with [message #12929 is a reply to message #12897] |
Tue, 13 January 2004 15:36  |
| Eclipse User |
|
|
|
Originally posted by: richardluong.msn.com
I think a roadmap might be good, but it might even be too early for a
roadmap. At least until we have a design and maybe some commitment from
volunteers.
One of the problems, I potentially see, is the "project". If we start
talking about a bunch of different projects merging their efforts into
one project, which one will it be? I would think that the only solution
might be to start a fresh new project, possibly named 'WebTools'. ;-)
This project would obviously credit all the contributors.
The next major hurdle would be the design. I'm sure not all the
projects have been built with the same vision. Also by reading the
posts in this newsgroup, the posters haven't seemed to agree on the same
vision either. Generally, the debate falls into two groups: basic web
or jsp. This is a potential deal-breaker. I think our webtools project
should support it all. Anything and everything. By building variety
into the tool, we build for whatever future technology might bring.
Now before everyone goes crazy on that thought, let me clarify. I think
it would be great if the project could itself be extended, either
through plug-ins or templates. That way, different language support
could be added to the plugin to the user's specific need. If you just
want basic web - HTML, XML, XSLT, CSS and Javascript then just install
the core plugin. If you want JSP, ASP, PHP, Ruby or whatever, then
install the appropriate extension. This way, everyone wins. If people
want Ruby, rather than writing a whole new plugin for Eclipse, they can
just write an extension for the webtools plugin. Hopefully that should
be a lot simpler!
I believe this is similar to the proposal in the "Web Tools vs. J2EE
Tools" posting from September 2003. Which I think will suit everyone's
needs. Although some people(including myself <-- J2EE guy), may want a
JSP editor off the bat, I think it should probably wait until at least
the basic infrastructure of the project and the core basic web tools are
complete. A JSP extension can be added later, but the extensions should
be taken into account when deciding on the architecture of the project.
So I guess the next steps would be to see who's willing to contribute,
and who's willing to make compromises to agree on a common design.
After that's done, maybe we could flesh out a project plan and then
based on the resources move it into a roadmap.
Hopefully not more than my two cents,
Richard.
Christopher Lenz wrote:
> Sjoerd van Leent wrote:
>
>> I already offered my company to do some project management, if wanted, I
>> am willing to develop an initial roadmap for the project to develop
>> against.
>>
>> Enough people want to get started with engineering and coding elements,
>> already other options are being searched to get it going (such as a
>> project on sourceforge).
>
>
> Which is actually not a bad thing. It's close to impossible to start a
> community-driven open-source project with no preexisting code base (I've
> personally never seen one succeed). I understand that the WSAD team has
> code they were planning to contribute, but I have no idea what kind of
> code that would be (except for the WSVT). There were a few others that
> announced they'd be willing to donate code. But little of the code is
> out there for anyone to evaluate.
>
> Building a foundation outside of Eclipse.org might very well be the only
> way to move forward.
>
> Cheers,
> Chris
> --
> /=/ cmlenz at gmx.de
|
|
|
| Re: Enough people to get started with [message #12998 is a reply to message #12929] |
Tue, 13 January 2004 16:01 |
| Eclipse User |
|
|
|
Richard Luong wrote:
> Now before everyone goes crazy on that thought, let me clarify. I think
> it would be great if the project could itself be extended, either
> through plug-ins or templates. That way, different language support
> could be added to the plugin to the user's specific need. If you just
> want basic web - HTML, XML, XSLT, CSS and Javascript then just install
> the core plugin. If you want JSP, ASP, PHP, Ruby or whatever, then
> install the appropriate extension. This way, everyone wins. If people
> want Ruby, rather than writing a whole new plugin for Eclipse, they can
> just write an extension for the webtools plugin. Hopefully that should
> be a lot simpler!
I think that some other posts also mentioned this problem. I think many
people agree that splitting the core web development (as you suggested,
the XML, HTML, XSLT and CSS bit) should be in the core project.
I think especially the XML plug-in but also other plug-ins should be
extensible enough to use them ase base for other plugins. XML could for
example well be used for XHTML, XSLT and JSP.
Subprojects would cover the development of specific target area's such as
J2EE, ASP(.NET), PHP or another web development environment.
If we sum this we would get a simple tree structure like the following:
Core project -> XML, CSS, HTML
|
+---- J2EE project -> JSPX, TAGX, Beans, Struts, JSF and so forth
|
+---- .NET -> ASPX
|
+---- . . . .
.
.
With the roadmap I mentioned before I just wanted to give knowledge that I
am willing to contribute the things most engineers (and people) in my
experience hate to do in order to get this project going.
|
|
| | | | | |
| Re: Enough people to get started with [message #13253 is a reply to message #13197] |
Wed, 14 January 2004 10:50 |
| Eclipse User |
|
|
|
Igor Malinin wrote:
> Christopher Lenz wrote:
> > Alex.Fitzpatrick wrote:
> >
> > I'd suggest the base name "web development tools" (wdt). So what about:
> >
> > wdt.sourceforge.net
> > eclipsewdt.sourceforge.net
> > eclipse-wdt.sourceforge.net
> > wdt4eclipse.sourceforge.net
> >
> Why don't you like SolarEclipse so much?
Who said they didn't like SolarEclipse?
As far as I can tell you're being invited to contribute like everyone else!
Merging a group of existing projects into a new name bears the advandage
of removing some egos.
If I donate all the JavaScript code (Of course we use the other one, if he
ever answers email) I will no longer be driving the project, I just become
another contributor. (Which is good, more eyes and hands in the code!)
We will not succeed if we continue to "own" the code, even emotionally.
> - it is a great name
Sorry, but SolarEclipse does not suggest "web development tools" to me.
[Even after a few pints. :-) ]
> - it is CPL, i.e. fully compatible with Eclipse
> - it is already established and popular
> - it was created exactly with that goal two years ago just after similar
> talks in eclipse newsgroups!!! Don't you think two years later we should
> create one more project and then again and again???
> - the project already in top of the SourceForge search for "eclipse"
So you're saying that it's a great candidate. Good.
> My vision is that project should be generic Web Tools (XML, XHTML, CSS,
> XSLT, JavaScript) and it should be very extensible. This is the first
> proprity! NB! All languages listed above are used in thin clients or
> web-browsers and it is one more reason (not main) for the choise of
> features.
So there are two visions here, either we need to merge the visions, or
create two projects.
> The one reference extensinon should be under same project and it should
> be Core J2EE tooling as most standart alternative (the same as with
> Eclipse Platform and JDT)... This is the second priority but it must be
> the part of whole project just because there could be no success for the
> first part without the second! And it is a good test for extensibility
> of the generic Web Tools.
> And as I can see many people share the same or very similar oppinion...
With a few small exceptions, those working the the HTML, CSS, JavaScript
etc. kind of world think that HTML, CSS, JavaScript etc. are critical and
that JSP/J2EE is a secondary concern, and the people working on J2EE apps
think the opposite.
Is there a way of blending these concerns into a single cohesive project,
or do we accept that they are different in some way and treat them as such?
--
Alex Fitzpatrick
|
|
|
| Re: Enough people to get started with [message #13281 is a reply to message #13253] |
Wed, 14 January 2004 12:14 |
| Eclipse User |
|
|
|
Alex Fitzpatrick wrote:
> Igor Malinin wrote:
>
>>Why don't you like SolarEclipse so much?
>
> Who said they didn't like SolarEclipse?
Sorry if I misunderstood someone...
> As far as I can tell you're being invited to contribute like everyone else!
To be precise...
I am the one of initial group who talked about Web Tools project and I
took the responsibility of registering/managing project and I made a
first contribution and currently 80% of the code is mine. So,
technically I am the owner of the SolarEclipse. But there are some other
people who made contributions during past two years. And big thank to
them for even very small contribution(s)...
> Merging a group of existing projects into a new name bears the advandage
> of removing some egos.
As I said SolarEclipse was created to unite developers... Initially I
even added as members anyone who asked! But changed this rule only
because most people newer contribute anything...
> If I donate all the JavaScript code (Of course we use the other one, if he
> ever answers email) I will no longer be driving the project, I just become
> another contributor. (Which is good, more eyes and hands in the code!)
There must be some owner. Or the project could be destroyed easily. But
project owner is different to code owner. Project owner is a person or
group who allow or disallow other people to contribute. Any project,
commercial or opensource, has an owner.
> We will not succeed if we continue to "own" the code, even emotionally.
Everyone continue to own their code. And only their code. As defined by
CPL. So, every contributor own some piece of code, but use other's code
in compliance with CPL license.
>>- it is a great name
>
> Sorry, but SolarEclipse does not suggest "web development tools" to me.
> [Even after a few pints. :-) ]
Every succesfull project I could remember has quite abstract name
(Eclipse as an example :))... I think it is because too specific name
limits possible future of the project... For example if you call your
project csstools or jsptools no one will expect say xml features from
this project, even if in the future project extended to have more
features. IMHO...
> With a few small exceptions, those working the the HTML, CSS, JavaScript
> etc. kind of world think that HTML, CSS, JavaScript etc. are critical and
> that JSP/J2EE is a secondary concern, and the people working on J2EE apps
> think the opposite.
It is the main reason why both should be part of one project, but
separate subprojects or components (or features in terms of Eclipse)
with J2EE feature dependent on generic WebTools feature.
> Is there a way of blending these concerns into a single cohesive project,
> or do we accept that they are different in some way and treat them as such?
XML/HTML/CSS are independent, but JSP is nothing without XML/HTML/CSS,
etc... It is quite natural dependency... And it must be one project.
Reasons is a
Imagine Eclipse without JDT/PDE. Would it then have same succes? No.
Why J2EE and not PHP or whatever? Just because eclipse iself is Java.
J2EE here is JSP/Servlets in first place - i.e. natural extention to
WebTools. EJB etc. could be be quite independent different projects, all
depends only on those who like contribute...
And... I think there is no problem with different priorities for
different people. J2EE Tools is not the first priority for me, but if
there is anyone who thinks different and
implement/contribute/maintain/own the J2EE component I will only happy!
I am probable user for that plugin too :)
|
|
|
| Re: Enough people to get started with [message #13307 is a reply to message #13197] |
Wed, 14 January 2004 12:17 |
| Eclipse User |
|
|
|
Hi Igor,
Igor Malinin wrote:
> Christopher Lenz wrote:
>> Alex.Fitzpatrick wrote:
>> I'd suggest the base name "web development tools" (wdt). So what about:
>>
>> wdt.sourceforge.net
>> eclipsewdt.sourceforge.net
>> eclipse-wdt.sourceforge.net
>> wdt4eclipse.sourceforge.net
>
> Why don't you like SolarEclipse so much?
>
> - it is a great name
> - it is CPL, i.e. fully compatible with Eclipse
> - it is already established and popular
> - it was created exactly with that goal two years ago just after similar
> talks in eclipse newsgroups!!! Don't you think two years later we should
> create one more project and then again and again???
> - the project already in top of the SourceForge search for "eclipse"
I've never said that I don't like SolarEclipse. I've used your stuff in the
past, and it has been nice. I certainly thought about donating the CSS
plugin code to it, and generally helping out in various areas. But as a
potential code base to contribute to, I have the following objections:
1) It hasn't been active for quite some time now, so when I hear
"SolarEclipse" I think of a community effort that has basically been a
failure (mainly in terms of building a community, not so much in terms of
the code produced).
2) The organization of functionality is a bit awkward. The CSS support is
hidden somewhere in the XML UI plugin (IIRC), which I consider a rather
weird decision. In general there doesn't seem to be enough separation
between the individual parts.
4) Actually, no, I don't think SolarEclipse is such a great name. Can't tell
you why, but I always felt it sounded somewhat "cheap & cheesy" ;-) If you
want an abstract name, choose one that is *really abstract* (like "Hyades"
or "Equinox" are, to choose example from Eclipse.org), otherwise just name
it after the function it performs.
3) The CVS repository is somewhat of a mess, with modules that contain typos
and so on.
Those are not problems that are impossible to fix, but I feel a fresh start
might actually be a good thing, both from a "psychological" and from a
technical point of view.
But that's just my opinion. The fact is that SolarEclipse hasn't attracted
developers for some time, even though separate work in the same problem
domain has been done. As you say yourself, it's currently a one man show.
I'd be interested in learning what you think the reasons for that are.
> My vision is that project should be generic Web Tools (XML, XHTML, CSS,
> XSLT, JavaScript) and it should be very extensible. This is the first
> proprity! NB! All languages listed above are used in thin clients or
> web-browsers and it is one more reason (not main) for the choise of
> features.
Okay, so we totally agree here.
> The one reference extensinon should be under same project and it should
> be Core J2EE tooling as most standart alternative (the same as with
> Eclipse Platform and JDT)... This is the second priority but it must be
In fact, JDT is simply necessary because Eclipse is written in Java, and
thus Java development tooling is a prerequisite of plugin development. That
doesn't mean that Java is "the most standard" development platform (it might
be, especially from IBM's perspective, but that's not the point).
> the part of whole project just because there could be no success for the
> first part without the second! And it is a good test for extensibility
> of the generic Web Tools.
I see your point about JSP support being a good way to provide a
proof-of-concept implementation. Actually, with such a wording, I'd be okay
to include JSP support in the project, as long as it is strictly separated.
Cheers,
Chris
--
cmlenz at gmx.de
|
|
| | | |
| Re: Enough people to get started with [message #13396 is a reply to message #13307] |
Wed, 14 January 2004 15:22 |
| Eclipse User |
|
|
|
Christopher Lenz wrote:
> I've never said that I don't like SolarEclipse. I've used your stuff in
> the past, and it has been nice. I certainly thought about donating the
> CSS plugin code to it, and generally helping out in various areas. But
> as a potential code base to contribute to, I have the following objections:
>
> 1) It hasn't been active for quite some time now, so when I hear
> "SolarEclipse" I think of a community effort that has basically been a
> failure (mainly in terms of building a community, not so much in terms
> of the code produced).
It hasn't been active because of a lack of interest from developers, not
users. It is not completely failure because some plugins still use
SolarEclipse as dependency, to name a few:
http://attrezzo.sourceforge.net/
http://www.improve-technologies.com/alpha/struts-config-edit or/
http://www.improve-technologies.com/alpha/wsdl-viewer/
There were some contributions:
Vasanth Dharmaraj - CSS Editor
Colin Savage - XSLT Debugger (through not completed but good starting point)
+ several small contributins
And I am practically ready for new release (through it has no any
visible features, but many bugfixes and internal improvements).
There are many projects that are really dead! And SolarEclipse is most
live opensource alternative. And statistics of visitors and downloads
tell for themselves.
> 2) The organization of functionality is a bit awkward. The CSS support
> is hidden somewhere in the XML UI plugin (IIRC), which I consider a
> rather weird decision. In general there doesn't seem to be enough
> separation between the individual parts.
CSS is an XML standard and placing it in one plugin is quite natural. I
don't think that creating many too fine-grained plugins is good. And CSS
is a very small piece. It could be separated to several packages through
(I already thought about this)...
> 4) Actually, no, I don't think SolarEclipse is such a great name. Can't
> tell you why, but I always felt it sounded somewhat "cheap & cheesy" ;-)
> If you want an abstract name, choose one that is *really abstract* (like
> "Hyades" or "Equinox" are, to choose example from Eclipse.org),
> otherwise just name it after the function it performs.
Heh... Strange order of numbers :)
What is your proposal of really abstract name?
> 3) The CVS repository is somewhat of a mess, with modules that contain
> typos and so on.
Just need to make a request to SourceForge support and that modules will
be deleted... You will get this kind of problem in any project with
several committrers...
> Those are not problems that are impossible to fix, but I feel a fresh
> start might actually be a good thing, both from a "psychological" and
> from a technical point of view.
I remember how much effort I need to spend to start project and make it
visible enough for potential users (and contributors too) so I don't
feel that doing this again is a good thing.
> But that's just my opinion. The fact is that SolarEclipse hasn't
> attracted developers for some time, even though separate work in the
> same problem domain has been done. As you say yourself, it's currently a
> one man show. I'd be interested in learning what you think the reasons
> for that are.
One problem is that I am not native English speaker, so I spend too much
time for writing single letter. And just cannot be very good coordinator
alone due to a lack of time. And I think that one active person is not
enough anyway. I think that having 3-4 active people is much more
comfortable! Not necessarily active coders, but maintaining
documentation and just coordination is too very important.
Second - the project was started when only 1.0 version of Eclipse was
released. Lack of features in platform was the serious stopper.
And third, very important - nobody starts comparable big part of work
that need serious analysis and much effort. But it is impossible to make
a good quality product without this part. The chicken-and-egg problem.
I am implemented one such part in SolarEclipse that hasn't done in any
other plugin I have seen - nested partitioning through combining
existing partitioners (good for JSP-like editors). And I think about
future extending it so partitions is linked like in AST to reflect tree
structure of the text. It should alow more flexibility for
context-dependent highlighting in XML dialects and some other languages.
And of course live XML outline (like in WSAD if you've seen it) almost
for free and immediately responsive to entered text.
Second important part is an XSLT debugger. It was newer released because
it is not finished by the original author (Colin Savage), but seems that
it is not that much of work to finish it... The main problem development
was suspended is (I think) very technical - buggy version of Xalan in
JDK 1.3. For 1.4 (requirment for future Eclipse 3.0 release) it should
be no problem anymore.
Fourth - frequent releases is very important. Every release gives back a
wave of interest to the project. But even with frequent releases you
need at least three months of headache to introduce your project in
every place where people could find it until people would find it by
themselvs. It is an experience from SolarEclipse and some other
web-projects. SolarEclipse already has known enough to the community so
any new release will attract attention. New project might be forgot in a
short period of time and need much effort to prevent this in quite long
period since creation. And it should be better than alternatives and now
it is harder than two years before when there was no any free or cheap
alternative - even more chances to fail.
>> The one reference extensinon should be under same project and it
>> should be Core J2EE tooling as most standart alternative (the same as
>> with Eclipse Platform and JDT)... This is the second priority but it
>> must be
>
> In fact, JDT is simply necessary because Eclipse is written in Java, and
> thus Java development tooling is a prerequisite of plugin development.
> That doesn't mean that Java is "the most standard" development platform
> (it might be, especially from IBM's perspective, but that's not the point).
Independent of the reasons Eclipse has JDT/PDE, Java is the primary
target platform in Eclipse. So Java is not most standard in general, but
definitely is standard in Eclipse. More than any other at least now and
near future.
> I see your point about JSP support being a good way to provide a
> proof-of-concept implementation. Actually, with such a wording, I'd be
> okay to include JSP support in the project, as long as it is strictly
> separated.
Completely agree.
|
|
| |
| Re: Enough people to get started with [message #13473 is a reply to message #13225] |
Wed, 14 January 2004 17:03 |
| Eclipse User |
|
|
|
Alex Fitzpatrick wrote:
> Christopher Lenz wrote:
>>(and if anyone knows an alternative to SourceForge that sucks less,
> please...)
>
> Hmmm, what's your objection to sourceforge? (Not that I think it's
> perfect, I'd just like you to explain...)
It's down rather frequently, and when it's not down, response time is pretty
bad. The issue tracker sucks, no admin access to the CVS repository, ViewCVS
lags behind the actual CVS repo by several hours at best, etc etc.
SourceForge is okay, but there's a lot of room for improvement :-P
Cheers,
Chris
--
cmlenz at gmx.de
|
|
|
| Re: Enough people to get started with [message #13491 is a reply to message #13335] |
Wed, 14 January 2004 18:05 |
| Eclipse User |
|
|
|
Originally posted by: richardluong.msn.com
Sjoerd van Leent wrote:
> A lot of answers again and I read them all.
>
> I also noticed that Eclipse is just in a transition of ownership (it is
> going to be switched from IBM to the Eclipse Foundation). This is why we
> should wait a bit longer before we make the move to sourceforge or
> something similar. Perhaps (because it would be a rather large project) it
> is needed in the future to have our own space on the web.
I would have to disagree on that point. I don't think we should wait
any longer. Yes, it would be great if we were backed by Eclipse and the
"official" project for webtools. However, I think the prior posts from
IBM have been saying, "We want a big company to get involved first, then
all the little guys can chime in." Maybe that'll change with the
Eclipse foundation, but I doubt it. And I don't think anymore time
should be wasted waiting.
Besides, it looks like we'll be using up some of the time deciding on
the actual project and it's design and scope. But from all the posts,
it seems as if everyone's on board with the base tool set and JSP stuff
being secondary as a subproject/extension/step-child.
Lastly, although I don't have any project codebase of my own, I would
like to cast my vote for starting a new project. I think it would
encourage more potential contributors to donate and help start a fresh
new project, rather than joining an older project that belongs to
someone else. Just my opinion.
Richard.
|
|
| |
| Re: Enough people to get started with [message #13533 is a reply to message #13473] |
Wed, 14 January 2004 18:23 |
| Eclipse User |
|
|
|
Christopher Lenz wrote:
> Alex Fitzpatrick wrote:
>
>> Christopher Lenz wrote:
>>
>>> (and if anyone knows an alternative to SourceForge that sucks less,
>>
>> please...)
>>
>> Hmmm, what's your objection to sourceforge? (Not that I think it's
>> perfect, I'd just like you to explain...)
>
>
> It's down rather frequently, and when it's not down, response time is
> pretty bad. The issue tracker sucks, no admin access to the CVS
> repository, ViewCVS lags behind the actual CVS repo by several hours at
> best, etc etc.
>
> SourceForge is okay, but there's a lot of room for improvement :-P
Good points, is there a better free open access CVS server?
--
Alex
|
|
| | | | |
| Re: Enough people to get started with [message #13682 is a reply to message #13533] |
Thu, 15 January 2004 08:07 |
| Eclipse User |
|
|
|
Alex.Fitzpatrick wrote:
> Christopher Lenz wrote:
>> Alex Fitzpatrick wrote:
>>> Hmmm, what's your objection to sourceforge? (Not that I think it's
>>> perfect, I'd just like you to explain...)
>>
>> It's down rather frequently, and when it's not down, response time is
>> pretty bad. The issue tracker sucks, no admin access to the CVS
>> repository, ViewCVS lags behind the actual CVS repo by several hours
>> at best, etc etc.
>>
>> SourceForge is okay, but there's a lot of room for improvement :-P
>
> Good points, is there a better free open access CVS server?
I don't think so. There are a couple of small sourceforges (using the
same underlying mess of scripts) and a couple of community-sites with a
narrower focus and higher requirements on new projects (such as tigris.org).
The best fit would probably be eclipse.org, anyway ;-P
But that just isn't going to happen any time soon, so we'll probably
have to live with SourceForge, which is still the best option for
launching a new project anyway.
Cheers,
Chris
--
cmlenz at gmx.de
|
|
| | | | |
| Re: Enough people to get started with [message #13824 is a reply to message #13449] |
Thu, 15 January 2004 12:47 |
| Eclipse User |
|
|
|
Igor Malinin wrote:
> JavaSctipt could not be totally replavced with CSS. Both have very
> different purpuse. CSS is a look, JavaScript is behavior.
I do think that the standards should be supplied. JavaScript indeed is
something different from CSS. However, it's main purpose is to validate
and check on errors when submitting fields for example. As XForms rolls in
the JavaScript validation rolls out.
There are indeed other purposes where JavaScript could be useful, however,
I do think that the use of JavaScript in any page should be as less as
possible, not due to the language, but due to the incompatibility.
We may disagree of the use of it, but we do agree that it should be
included, thus I do not think that we should fight eachother, we should
provide our ideas to create a wide base to engineer on.
|
|
| |
| Re: Enough people to get started with [message #13944 is a reply to message #13839] |
Thu, 15 January 2004 13:12 |
| Eclipse User |
|
|
|
Sjoerd van Leent wrote:
> Richard Luong wrote:
>>I would have to disagree on that point. I don't think we should wait
>>any longer. Yes, it would be great if we were backed by Eclipse and the
>>"official" project for webtools. However, I think the prior posts from
>>IBM have been saying, "We want a big company to get involved first, then
>>all the little guys can chime in." Maybe that'll change with the
>>Eclipse foundation, but I doubt it. And I don't think anymore time
>>should be wasted waiting.
>
> I am indeed somewhat conservative here. But I think that the sentence you
> quoted posted by IBM is a shame. Have small companies less rights or
> somethink like that in the USA? I think that a bunch of small companies
> are better than one large company.
This is, IIUC, a very political issue. My understanding is that
IBM/OTI/Eclipse.org wants to get as many as possible different tooling
companies into the boat, because otherwise the vision and image of
Eclipse as a vendor-neutral platform becomes nothing but a nice theory.
If Eclipse.org started a full-fledged Web Tools project without first
trying to get other big players involved, that would probably upset
quite a couple of potential consortium members. And if a big player
*does* take this up, we'll see how much of a community-driven project
this will become :-P
Cheers,
Chris
--
cmlenz at gmx.de
|
|
| |
| Re: Enough people to get started with [message #13993 is a reply to message #13473] |
Mon, 19 January 2004 03:58 |
| Eclipse User |
|
|
|
Originally posted by: jens.ja-web.de
Christopher Lenz wrote:
> Alex Fitzpatrick wrote:
>
>> Christopher Lenz wrote:
>>
>>> (and if anyone knows an alternative to SourceForge that sucks less,
>>
>> please...)
>>
>> Hmmm, what's your objection to sourceforge? (Not that I think it's
>> perfect, I'd just like you to explain...)
>
>
> It's down rather frequently, and when it's not down, response time is
> pretty bad. The issue tracker sucks, no admin access to the CVS
> repository, ViewCVS lags behind the actual CVS repo by several hours at
> best, etc etc.
>
> SourceForge is okay, but there's a lot of room for improvement :-P
what do people think about
http://savannah.gnu.org/ http://savannah.nongnu.org/
at least performance-wise it looks waaaay better than SF.net
Jens
|
|
| |
| Re: Enough people to get started with [message #14659 is a reply to message #13533] |
Wed, 21 January 2004 17:58 |
| Eclipse User |
|
|
|
Originally posted by: Joerg.remove-this.Schaible.gmx.remove-this.de
Alex.Fitzpatrick wrote:
> Christopher Lenz wrote:
>
>> Alex Fitzpatrick wrote:
>>
>>> Christopher Lenz wrote:
>>>
>>>> (and if anyone knows an alternative to SourceForge that sucks less,
>>>
>>> please...)
>>>
>>> Hmmm, what's your objection to sourceforge? (Not that I think it's
>>> perfect, I'd just like you to explain...)
>>
>>
>> It's down rather frequently, and when it's not down, response time is
>> pretty bad. The issue tracker sucks, no admin access to the CVS
>> repository, ViewCVS lags behind the actual CVS repo by several hours at
>> best, etc etc.
>>
>> SourceForge is okay, but there's a lot of room for improvement :-P
>
> Good points, is there a better free open access CVS server?
http://developer.berlios.de/
Also an SF clone.
Regards,
Jörg
|
|
|
Goto Forum:
Current Time: Fri Jun 06 18:32:36 EDT 2025
Powered by FUDForum. Page generated in 0.06056 seconds |
Home