Sent: 2011-04-18 14:52:40 UTC (about 1 year ago)

From: Alexandre Prokoudine

new website

permalink

Hi,

So I spoke to Jimmac, and he told me he has no time to do the new
gimp.org. He also said *everything* has to be done. That's IMO a bit
of exxageration given some work already done in static-html5 branch.
However it still means we need somebody else to do the work, as both
Jimmac and Andy Fitzsimon have no time to lead the project, but are
ready to help with core stuff.

Earlier we agreed that we want the new site to be static + JS for
updating content. However I don't remember a discussion on exact
content, or maybe I just missed it, because I wasn't around all the
time at the meeting last LGM.

Hence three questions:

1. Do we want to stick to everything we have right now (with
subsequent cleanup of FAQ and tutorials) or do we want to let go of
something and/or do something new (videos?)

2. Do we already have specs for the new site? I.e. pyramidal approach
to layout and suchlike?

3. Which way of finding new webdesigner and webdev should we go?
Announce a contest like the one Inkscape did on DeviantArt? Just write
news on gimp.org and sort through everyone who sends anything?

The whole thing is likely to take months, so we need to start *now* if
we intend to have a visual refresh for 2.8.

Alexandre Prokoudine
http://libregraphicsworld.org

Sent: 2011-04-18 15:35:41 UTC (about 1 year ago)

From: Jason Simanek

new website

permalink

Where do I apply to help with the website? Where can I read more about
the goals of the new site?

Jason Simanek
Web Designer/Developer

Sent: 2011-04-18 15:54:41 UTC (about 1 year ago)

From: Alexandre Prokoudine

new website

permalink

On 4/18/11, Jason Simanek wrote:

> Where do I apply to help with the website?

You are already here, no? :)

> Where can I read more about the goals of the new site?

There isn't much beyond
https://lists.xcf.berkeley.edu/lists/gimp-developer/2010-June/024830.html
to the best of my knowledge.

I'd summarize it as:

1. Make a visual refresh, better present what GIMP can do
2. Make it easy to update content for maintainers.

We are also thinking about creating a blogs planet. We can reuse
aggregator used at graphicsplanet.org for that.

Alexandre Prokoudine
http://libregraphicsworld.org

Sent: 2011-04-19 02:55:17 UTC (about 1 year ago)

From: Carol Spears

new website

permalink

On Mon, Apr 18, 2011 at 06:52:40PM +0400, Alexandre Prokoudine wrote:
> Hi,
>
> So I spoke to Jimmac, and he told me he has no time to do the new
> gimp.org. He also said *everything* has to be done. That's IMO a bit
> of exxageration given some work already done in static-html5 branch.
> However it still means we need somebody else to do the work, as both
> Jimmac and Andy Fitzsimon have no time to lead the project, but are
> ready to help with core stuff.
>
you should ask Jimmac a really good question about the "core" stuff
like what to do about when raphael in 2004 doesn't remember where
the howto was stashed by raphael in 2002 because from what i could
very easily see, there is not anyone who knows how to use the
"core stuff" involved with the web site at this moment in time.

determining the reason that raphael from one year did not know what
raphael from another year did and if this is a good reason would be
a very good place to start.

before raphael in 2002 decided that the web site had been abandoned
(a determination which was not shared by all, btw) our plans for the
next version of the web site included things like apt-get for plug-in
installation and a registry that could manage interaction like this.

since this situation with the dual versions of raphael has been
running things, we instead have a registry that thinks that it contains
"extensions", among other misnomers and this is one example which is
perhaps the least of the problems.

how to handle the "core stuff" is an interesting question because the
web site is being used only to about 15% of its capacity (that it instantly
rebuilds upon a commit) so, at the least, out of respect for previous
versions of the site, if you could restrain yourself from suggesting that
knowlege of the how the "core stuff" works can be gained from anyone
operating (however innocently) in the post-raphael duality problem world
that exists today.

carol

Sent: 2011-04-19 03:42:19 UTC (about 1 year ago)

From: Carol Spears

new website

permalink

On Mon, Apr 18, 2011 at 06:52:40PM +0400, Alexandre Prokoudine wrote:
>
> However it still means we need somebody else to do the work, as both
> Jimmac and Andy Fitzsimon have no time to lead the project, but are
> ready to help with core stuff.
>
one more thing. to the best of my knowlege, Andy Fitzsimon had very little
to do with running the web site and the person who really really really
needs to answer questions about what has been going on with the site since
~2005 is Michael Schumacher. i have some very good questions about some of
the content changes that have occured since then.

carol

Sent: 2011-04-19 04:52:29 UTC (about 1 year ago)

From: Martin Nordholts

new website

permalink

On 04/18/2011 04:52 PM, Alexandre Prokoudine wrote:
> Hi,
>
> So I spoke to Jimmac, and he told me he has no time to do the new
> gimp.org. He also said *everything* has to be done. That's IMO a bit
> of exxageration given some work already done in static-html5 branch.
> However it still means we need somebody else to do the work, as both
> Jimmac and Andy Fitzsimon have no time to lead the project, but are
> ready to help with core stuff.
>
> Earlier we agreed that we want the new site to be static + JS for
> updating content. However I don't remember a discussion on exact
> content, or maybe I just missed it, because I wasn't around all the
> time at the meeting last LGM.
>
> Hence three questions:
>
> 1. Do we want to stick to everything we have right now (with
> subsequent cleanup of FAQ and tutorials) or do we want to let go of
> something and/or do something new (videos?)
>
> 2. Do we already have specs for the new site? I.e. pyramidal approach
> to layout and suchlike?
>
> 3. Which way of finding new webdesigner and webdev should we go?
> Announce a contest like the one Inkscape did on DeviantArt? Just write
> news on gimp.org and sort through everyone who sends anything?
>
> The whole thing is likely to take months, so we need to start *now* if
> we intend to have a visual refresh for 2.8.

I'd rather plan for a visual refresh for GIMP 3.0, that gives us more
time and aligns with a new era of GIMP

BR,
Martin

My GIMP Blog:
http://www.chromecode.com/
"GIMP 2.8 schedule on tasktaste.com"

Sent: 2011-04-19 05:25:34 UTC (about 1 year ago)

From: Jason Simanek

new website

permalink

Carol, Alexandre and anybody who's listening,

I'm not sure about the history of the site, but it looks like the
content of the primary gimp.org site is pretty straight-forward. It
sounds like the general thought about the site is to avoid a CMS like
WordPress. I'm not sure why, WordPress can be made to work very well,
especially with a good caching plugin. WordPress would also make
updating the site more accessible to other members of the Gimp
developer community.

However, if a static, PHP-driven site is preferable I can do that too.
But as important as how the site is built is, the design is equally
important. And . . . looks like Martin Nordholts just put in his two
cents about that detail . . .

> I'd rather plan for a visual refresh for GIMP 3.0, that gives us more
> time and aligns with a new era of GIMP
>
> BR,
> Martin

Redesigning the site could be a great way to prepare people for the
upcoming Gimp 3. Hell, I just started using the 2.7 version and can't
wait for 2.8!

In the last year I built a new site for the Gramps project (genealogy software):

Course, the design for the Gimp site would be more appropriate for
Gimp, but that site is a good one, I think (though we haven't yet
implemented a caching plugin). The support for internationalization is
great, if you have people to write translations. The blog section
where various contributors and developers can write posts is great.

Regardless, I am interested in helping out with the site. I'm always
relatively busy, but I like and use Gimp and Inkscape for all of my
freelance design work. I would love to have an opportunity to give
back.

-Jason Simanek

Sent: 2011-04-19 05:48:22 UTC (about 1 year ago)

From: Martin Nordholts

new website

permalink

2011/4/19 Jason Simanek :
> I'm not sure about the history of the site, but it looks like the
> content of the primary gimp.org site is pretty straight-forward. It
> sounds like the general thought about the site is to avoid a CMS like
> WordPress. I'm not sure why, WordPress can be made to work very well,
> especially with a good caching plugin. WordPress would also make
> updating the site more accessible to other members of the Gimp
> developer community.

I think the main issue is security concerns, but if we host the site
on a separate machine and take regular back ups, that shouldn't be an
issue. Making the site trivial to update is paramount. Maybe you could
setup something quick in wordpress that people could try?

Best regards,
Martin

Sent: 2011-04-19 06:18:41 UTC (about 1 year ago)

From: Alexia Death

new website

permalink

On Tuesday, April 19, 2011 08:48:22 Martin Nordholts wrote:
> I think the main issue is security concerns, but if we host the site
> on a separate machine and take regular back ups, that shouldn't be an
> issue. Making the site trivial to update is paramount. Maybe you could
> setup something quick in wordpress that people could try?
I run deveral wordpress sites for my family ;) It's so simple that both my
sociology master sister and artstudent sister manage to use theirs without my
help. However, there is one con that you dont see in a test installation. It's
comment section is a popular target for spammers(I dont know if the whole
facility can be turned off) and due to widespreaded use, if there is a
vulnerability it gets exploited very very fast. These are the two things you
need to be prepared for with wordpress, spam and constant updates to avoid
getting defaced. If updates can be automated somehow that would just leave
spam.

Another downside is separate usermanagement

Benefits are great however. You get a truly dynamic site that is easy to update
:)

Best,
Alexia

Sent: 2011-04-19 06:33:37 UTC (about 1 year ago)

From: Martin Nordholts

new website

permalink

2011/4/19 Alexia Death :
> On Tuesday, April 19, 2011 08:48:22 Martin Nordholts wrote:
>> I think the main issue is security concerns, but if we host the site
>> on a separate machine and take regular back ups, that shouldn't be an
>> issue. Making the site trivial to update is paramount. Maybe you could
>> setup something quick in wordpress that people could try?
> I run deveral wordpress sites for my family ;) It's so simple that both my
> sociology master sister and artstudent sister manage to use theirs without my
> help. However, there is one con that you dont see in a test installation. It's
> comment section is a popular target for spammers(I dont know if the whole
> facility can be turned off) and due to widespreaded use, if there is a
> vulnerability it gets exploited very very fast. These are the two things you
> need to be prepared for with wordpress, spam and constant updates to avoid
> getting defaced. If updates can be automated somehow that would just leave
> spam.

Doesn't even requiring a login or captchas prevent this spam? If not
and spam become a big problem, maybe we have to disable comments
altogether. At least we'll have a website that is easy to update.

/ Martin

--

My GIMP Blog:
http://www.chromecode.com/
"GIMP 2.8 schedule on tasktaste.com"

Sent: 2011-04-19 08:10:52 UTC (about 1 year ago)

From: Shlomi Fish

new website

permalink

Hi all,

On Tuesday 19 Apr 2011 08:25:34 Jason Simanek wrote:
> Carol, Alexandre and anybody who's listening,
>
> I'm not sure about the history of the site, but it looks like the
> content of the primary gimp.org site is pretty straight-forward. It
> sounds like the general thought about the site is to avoid a CMS like
> WordPress. I'm not sure why, WordPress can be made to work very well,
> especially with a good caching plugin. WordPress would also make
> updating the site more accessible to other members of the Gimp
> developer community.
>
> However, if a static, PHP-driven site is preferable I can do that too.
> But as important as how the site is built is, the design is equally
> important. And . . . looks like Martin Nordholts just put in his two
> cents about that detail . . .
>

And another option would be an almost 100% static HTML site generated from
templates, like I have on:

* http://www.shlomifish.org/

* http://perl-begin.org/

* http://better-scm.berlios.de/

Etc.

This will be less resource hungry than a PHP-based site and more secure
(because you cannot exploit a problem in a static HTML site per-se and because
many authors of PHP, and to a lesser extent other server-side solutions are
sloppy and are not aware of their many pitfalls - see:
http://www.shlomifish.org/open-source/anti/php/ ) .

I would strongly recommend against even getting near WordPress.com because not
only it is incredibly insecure, but it also has very poor usability. See:

http://www.mail-archive.com/discussions@hamakor.org.il/msg02475.html

> > I'd rather plan for a visual refresh for GIMP 3.0, that gives us more
> > time and aligns with a new era of GIMP
> >
> > BR,
> > Martin
>
> Redesigning the site could be a great way to prepare people for the
> upcoming Gimp 3. Hell, I just started using the 2.7 version and can't
> wait for 2.8!
>
> In the last year I built a new site for the Gramps project (genealogy
> software):
>
>
>
> Course, the design for the Gimp site would be more appropriate for
> Gimp, but that site is a good one, I think (though we haven't yet
> implemented a caching plugin). The support for internationalization is
> great, if you have people to write translations. The blog section
> where various contributors and developers can write posts is great.
>
> Regardless, I am interested in helping out with the site. I'm always
> relatively busy, but I like and use Gimp and Inkscape for all of my
> freelance design work. I would love to have an opportunity to give
> back.
>

Regards,

Shlomi Fish

-----------------------------------------------------------------
Shlomi Fish http://www.shlomifish.org/
The Case for File Swapping - http://shlom.in/file-swap

"My name is Inigo Montoya. You forced my father to write XSLT. Prepare to die!
And be thankful I don't force you to write XSLT."

Please reply to list if it's a mailing list post - http://shlom.in/reply .

Sent: 2011-04-19 11:42:21 UTC (about 1 year ago)

From: Alexandre Prokoudine

new website

permalink

On Tue, Apr 19, 2011 at 8:52 AM, Martin Nordholts wrote:

>> The whole thing is likely to take months, so we need to start *now* if
>> we intend to have a visual refresh for 2.8.
>
> I'd rather plan for a visual refresh for GIMP 3.0, that gives us more
> time and aligns with a new era of GIMP

I'd rather not use the word "plan" with regards to something we have
no idea when it's out :)

In my experience it's best to do things while we have the drive to do
them. Nothing prevents us from facelifting gimp.org at 3.0 as long as
everything's done by that time. But I'd really like the thread to
refocus on the questions asked above :)

Alexandre Prokoudine
http://libregraphicsworld.org

Sent: 2011-04-19 13:21:47 UTC (about 1 year ago)

From: Jason Simanek

new website

permalink

On 04/19/2011 03:10 AM, Shlomi Fish wrote:
> This will be less resource hungry than a PHP-based site and more secure
> (because you cannot exploit a problem in a static HTML site per-se and because
> many authors of PHP, and to a lesser extent other server-side solutions are
> sloppy and are not aware of their many pitfalls - see:
> http://www.shlomifish.org/open-source/anti/php/ ) .

There are many counter-arguments to your view of PHP. We don't need to
start a flame-war about server-side scripting languages. The issue with
having a static HTML site is that only very few people will be able to
update it. Using a solution like WordPress would make the site more
accessible to Gimp developers and community members so that they can
help update the content of the site.

> I would strongly recommend against even getting near WordPress.com because not
> only it is incredibly insecure, but it also has very poor usability. See:

I think an undeniable proof of poor usability is us having this
discussion about how to update the Gimp site because the last version,
which seems to be static HTML or built with some server-side scripting,
is apparently not easy to update.

All of these problems can be avoided by good site construction and
management. We will not need to create any custom plugins for what the
current site already does. As far as being "resource hungry", any
database-driven CMS will have that problem without a good caching solution.

> 2011/4/19 Alexia Death :
>> > On Tuesday, April 19, 2011 08:48:22 Martin Nordholts wrote:
>>> However, there is one con that you dont see in a test installation. It's
>> > comment section is a popular target for spammers(I dont know if the whole
>> > facility can be turned off) and due to widespreaded use, if there is a
>> > vulnerability it gets exploited very very fast. These are the two things you
>> > need to be prepared for with wordpress, spam and constant updates to avoid
>> > getting defaced.

> Doesn't even requiring a login or captchas prevent this spam? If not
> and spam become a big problem, maybe we have to disable comments
> altogether. At least we'll have a website that is easy to update.

You can set up the comments on a WordPress site in a variety of ways:

- Must be a registered user to comment
- Anyone can comment, but the first comment from a given email address
is held for approval. After that, all following comments are let go.
- Anyone can comment
- Anyone can comment, but Akismet, their comment-spam service, filters
your comments

We could also turn off comments completely and let people discuss
anything interesting elsewhere: Twitter, Facebook, the mailing list,
etc. Outside of blog or News entries, comments don't really have a place
on the primary site anyway. There certainly aren't any comments there now.

I think security through obscurity is not security at all, so that is a
weak argument against WordPress. It's very secure in that it is
constantly being assaulted and patched. And with comments turned off you
are severely limiting the points of access.

One other major bonus for WordPress is the ability to allow site members
to submit translations for each page. Here's info about the plugin that
we used for gramps-project.org:

Course, I see now that we'd have to buy a license to get it, but for $79
it's a cheap solution to support multiple languages. And having multiple
language support for gimp.org seems like a necessity.

Hope this didn't come-off as sounding defensive. I want a practical
solution for Gimp more than a theoretically-pleasing one. I think a CMS
would make the site easier to keep updated, whichever CMS that may be.

-Jason Simanek

Sent: 2011-04-20 07:35:27 UTC (about 1 year ago)

From: Shlomi Fish

new website

permalink

Hi Jason,

On Tuesday 19 Apr 2011 16:21:47 Jason Simanek wrote:
> On 04/19/2011 03:10 AM, Shlomi Fish wrote:
> > This will be less resource hungry than a PHP-based site and more secure
> > (because you cannot exploit a problem in a static HTML site per-se and
> > because many authors of PHP, and to a lesser extent other server-side
> > solutions are sloppy and are not aware of their many pitfalls - see:
> > http://www.shlomifish.org/open-source/anti/php/ ) .
>
> There are many counter-arguments to your view of PHP.

Like which ones?

> We don't need to start a flame-war about server-side scripting languages.

Maybe we shouldn't.

> The issue with
> having a static HTML site is that only very few people will be able to
> update it.

That's not entirely true if you are using a version control system (possibly a
DVCS) and keep the sources under version control.

> Using a solution like WordPress would make the site more
> accessible to Gimp developers and community members so that they can
> help update the content of the site.

True.

>
> > I would strongly recommend against even getting near WordPress.com
> > because not
>
> > only it is incredibly insecure, but it also has very poor usability. See:
> I think an undeniable proof of poor usability is us having this
> discussion about how to update the Gimp site because the last version,
> which seems to be static HTML or built with some server-side scripting,
> is apparently not easy to update.

Did you try to update it.

>
> All of these problems can be avoided by good site construction and
> management. We will not need to create any custom plugins for what the
> current site already does. As far as being "resource hungry", any
> database-driven CMS will have that problem without a good caching solution.

Yes, you are right. Of course, I suppose that most modern systems can handle
such a load pretty well.

>
> > 2011/4/19 Alexia Death :
> >> > On Tuesday, April 19, 2011 08:48:22 Martin Nordholts wrote:
> >>> However, there is one con that you dont see in a test installation.
> >>> It's
> >>>
> >> > comment section is a popular target for spammers(I dont know if the
> >> > whole facility can be turned off) and due to widespreaded use, if
> >> > there is a vulnerability it gets exploited very very fast. These are
> >> > the two things you need to be prepared for with wordpress, spam and
> >> > constant updates to avoid getting defaced.
> >
> > Doesn't even requiring a login or captchas prevent this spam? If not
> > and spam become a big problem, maybe we have to disable comments
> > altogether. At least we'll have a website that is easy to update.
>
> You can set up the comments on a WordPress site in a variety of ways:
>
> - Must be a registered user to comment
> - Anyone can comment, but the first comment from a given email address
> is held for approval. After that, all following comments are let go.
> - Anyone can comment
> - Anyone can comment, but Akismet, their comment-spam service, filters
> your comments

The problem with WordPress is that its comments does not have previews by
default (so you have to hope they look OK after submission) and that all
comment previews plugins require JavaScript, and that it has a tendency to eat
comments.

>
> We could also turn off comments completely and let people discuss
> anything interesting elsewhere: Twitter, Facebook, the mailing list,
> etc. Outside of blog or News entries, comments don't really have a place
> on the primary site anyway. There certainly aren't any comments there now.
>

Yes, I agree.

> I think security through obscurity is not security at all, so that is a
> weak argument against WordPress. It's very secure in that it is
> constantly being assaulted and patched. And with comments turned off you
> are severely limiting the points of access.
>

WordPress may not employ security by obscurity but that does not make it
secure. Lack of obscurity is a necessary condition towards security, but not a
sufficient condition. (see:
http://en.wikipedia.org/wiki/Necessary_and_sufficient_condition ). By all
means, the flood of serious security vulnerabilities (often quickly exploited)
for WordPress is a clear indication that it has been poorly written, with a
code of bad quality. Open source is not a panacea for security, and you cannot
use it as a figleaf.

WordPress is insecure and badly written. It's a fact.

> One other major bonus for WordPress is the ability to allow site members
> to submit translations for each page. Here's info about the plugin that
> we used for gramps-project.org:
>
>
>
> Course, I see now that we'd have to buy a license to get it, but for $79
> it's a cheap solution to support multiple languages. And having multiple
> language support for gimp.org seems like a necessity.

I am admently oppposed to using a proprietary plugin for a site of an open-
source project. That will make us look bad, will burden our users and is
completely unethical. For why proprietary software should not be trusted, see:

http://better-scm.berlios.de/docs/shlomif-evolution.html

>
> Hope this didn't come-off as sounding defensive. I want a practical
> solution for Gimp more than a theoretically-pleasing one. I think a CMS
> would make the site easier to keep updated, whichever CMS that may be.

Yes, but we can also use a wiki or something like that.

Regards,

Shlomi Fish

(Please reply to the list or don't reply at all.)

-----------------------------------------------------------------
Shlomi Fish http://www.shlomifish.org/
"Star Trek: We, the Living Dead" - http://shlom.in/st-wtld

My Commodore 64 is suffering from slowness and insufficiency of memory, and
its
display device is grievously short of pixels. Can anybody help? -- Omer Zak

Please reply to list if it's a mailing list post - http://shlom.in/reply .

Sent: 2011-04-20 08:26:15 UTC (about 1 year ago)

From: Martin Nordholts

new website

permalink

2011/4/20 Shlomi Fish :
>> The issue with
>> having a static HTML site is that only very few people will be able to
>> update it.
>
> That's not entirely true if you are using a version control system (possibly a
> DVCS) and keep the sources under version control.

We currently version control gimp-web with git, but updating the site
is not easy enough. First, we don't want to go through GNOME git
account procedure to allow a community member to update the site.
Second, when updating sources directly you have to worry about writing
proper HTML, when it should be as easy as just writing content, like a
news item. People updating site content should not have to worry about
making sure not to break the entire site.

/ Martin

--

My GIMP Blog:
http://www.chromecode.com/
"GIMP 2.8 schedule on tasktaste.com"

Sent: 2011-04-21 12:08:27 UTC (about 1 year ago)

From: Alexandre Prokoudine

new website

permalink

On 4/20/11, Shlomi Fish wrote:

>> The issue with having a static HTML site is that only very few people will
>> be able to update it.
>
> That's not entirely true if you are using a version control system (possibly
> a DVCS) and keep the sources under version control.

I was kinda hoping for this discussion not to happen, but since you insist...

First of all, static HTML site is what we have *already* agreed on.
The agreement comes from status quo, which is: very few people
*interested* in maintaining the website and related services. If we
don't have 24/7/365 maintenance of the website, exposing it to people
who know about secuity more than we do is a no go. Period.

The website is barely maintained not because it's difficult to update
it, but because nobody wants to do it. You can argue about that till
you are blue in the face, but it's been that way for years: noone ever
wants to take responsibility. CMS is not a solution to that, nor DVCS
is (especially _D_VCS). This is a social issue, CMS is of no help
there.

So the first prerequisite for a CMS is a world-wide team that takes
responsibility for round-the-clock availability of the website.

The second prerequisite when CMS is something that might and should
happen is providing additional services: a resources catalog, an
improved version of registry.gimp.org etc.

It's been years since we have easy installation of resources and
extensions on our todo list, we even have a failed GSoC 2006 project
on that. I'm a realist and I don't see that changing any time soon. We
can't install an arbitrary CMS and hope that it might turn out good
enough if and when we have people to work on the other stuff. Was
Drupal a good choice for extensions registry? Clearly no.

Creating a new GIMP.org with new registry, resources catalog and
whatnot is a job for several experienced web developers backed by
several GIMP developers who would work on the software part. And it's
months of work, since it's about volunteering in spare time, and these
months of work won't start any time soon, because core team is busy
*all the time* with more important things. Does anybody see it
changing? I don't.

Hence the plan for static HTML5 + JavaScript for updating content.

Now once again I'd like to return your attention back to the questions
asked in the mail that started the thread. Thank you in advance :)

Alexandre Prokoudine
http://libregraphicsworld.org

Sent: 2011-04-22 18:33:55 UTC (about 1 year ago)

From: yahvuu

new website

permalink

Hi all,

please pardon me for arguing despite a long proven record
of not helping out on the GIMP website -- but i want to make
shure the following solutions model has been discussed.

On Thu, Apr 21, 2011 at 2:08 PM, Alexandre Prokoudine
wrote:
> [..] status quo, which is: very few people [are]
> *interested* in maintaining the website and related services.

true enough -- regarding the _technical_ part of the website.

> The website is barely maintained not because it's difficult to update
> it, but because nobody wants to do it.

I simply refuse to believe that not a single one among the 13 gimp users
who are into web design will find time and pride to show off her graphics
tool of choice and thereby showcase his personal web design skills.

(please substitute "13" by a realistic number of your choice).

> [..] it's been that way for years: noone ever
> wants to take responsibility.

Regarding the web site's _content_, IMHO this is a false conclusion:
No one new is ever _allowed_ to take responsibility. All you can do
is send in a patch for review. In fact, the real responsibility is in the
hands of those who apply the patches (but no patches get sent in ever).

I strongly suggest to completely revert the process:
Anybody showing interest in contributing to the web site
should be _urged_ to do so. Without restrictions and with
access as simple as possible. People tend to take responsibility
when they are given the powers to make a difference.

And i'd call it progress in case we face edit wars in gimp.org's
tutorials section.

There's a difference between code and web site content,
as much as is between the people(s) who usually have
fun in creating one of the two. The patch review process
is tried and proven for coding, but it is overkill for web site
content and repels people willing to contribute there.

The crucial part of the website, say starting page + downloads,
may remain under the safe control of the board of developers.
But for the rest of the site: what is the worst damage a
drunken/malicious user can cause? That will not get reverted
a few hours later?

Proposed standard reply for someone sending in a typo/spelling correction:
"Thanks for your contribution. Please correct the mistake yourself
on http://cms.gimp.org/page_affected. Be shure to add yourself to the
AUTHORS list on xxyyzz afterwards. I have just created a new account
for you with user:xy pass:yy.

In case you're interested to help out, our most needing page is xyz.
If you have any questions, do not hesitate to contact our community here: xyy"

> CMS is not a solution to that, nor DVCS is (especially _D_VCS).
> This is a social issue, CMS is of no help there.

The only way to prove your hypothesis is to experimentally demonstrate that
a CMS does not improve the situation...

thanks for you consideration,
yahvuu

Sent: 2011-04-22 20:10:14 UTC (about 1 year ago)

From: Alexandre Prokoudine

new website

permalink

On 4/22/11, yahvuu wrote:

>> CMS is not a solution to that, nor DVCS is (especially _D_VCS).
>> This is a social issue, CMS is of no help there.
>
> The only way to prove your hypothesis is to experimentally demonstrate that
> a CMS does not improve the situation...

You know, I had a rather long reply to each of your points, but this
last bit made me erase most of it, because your stubborness is trying
to rival mine, and I'm _not_ having any of that :)

The fact that the team is in need for help is not a mistery. It was
never hushed up by government or men in black backed up by helicopters
of likewise colour. It wasn't communicated to the user base well
enough, but even now when it _is_ communicated to user base, very,
very few people care to make a difference.

Over past four months ever since the "Why GIMP 2.8 is delayed" thing
began only a handful of developers cared to cook up a few patches and
none of them sticked to the project in the end (I'm personally
thankful to them even for that). Despite of the community knowing abou
all the issues I know just one person who keeps contibuting here and
there, but he's been around for years. Apart from that what we mostly
got was a temporary raise of donations, which we still haven't found a
use for to make 2.8 happen sooner.

Please learn facts before you start talking about hypotheses.

There are still things to be done to improve the situation, but _once
again_ they are mostly social and less technical. I'll just list a
few:

- we are not making it easy to get into development (partially solved
thanks to Alexia and Barak)
- we are not promoting new core features
- we are not promoting 3rd party development
- we are not promoting educational materials, courses, workshops and suchlike

All of the things above make a huge difference when you are trying to
create a rewarding community. In short, we are not even attempting to
build an ecosystem to make the project support itself. This plays a
huge role in the fact that the team is always outnumbered.

There is no silver bullet, there is only patience, passion and
dedication. If people don't have it, no CMS will substitute that. You
want to argue against that? Fine with me, but I'm not participating.
I'd like to have something done instead, thank you very much.

Now that I've done my shouting, may I once again try to get the
discussion back to the topic of new design? I quite like Jason's take
at GRAMPS website's design (I remember all its previous versions), but
I'm interested in what else people can come up with, and that sounds
like contest to me.

Could we please have a voting of some kind whether we do a contest or not?

Alexandre Prokoudine
http://libregraphicsworld.org

Sent: 2011-04-22 20:16:25 UTC (about 1 year ago)

From: Alexandre Prokoudine

new website

permalink

On 4/23/11, Alexandre Prokoudine wrote:

> I'd like to have something done instead, thank you very much.

Meaning "I'd like to get something done instead"

Truly yours,
Freudean slipmeister

Sent: 2011-04-22 21:36:06 UTC (about 1 year ago)

From: Ronald Chmara

new website

permalink

On Fri, Apr 22, 2011 at 1:10 PM, Alexandre Prokoudine
wrote:
> - we are not making it easy to get into development (partially solved
> thanks to Alexia and Barak)
>....
> There is no silver bullet, there is only patience, passion and
> dedication. If people don't have it, no CMS will substitute that.

These two thoughts seem to be somewhat in conflict.

A CMS arguably makes it easier to get into website development for
people who don't have the "patience, passion and dedication" to use
fairly convoluted workflows to do something as simple as updating,
managing, and contributing web content.

That being said, a CMS does not produce volunteers.... It merely
reduces the barriers a prospective volunteer has to cross in order to
become an active participant, so perhaps the conversation is not so
much about CMS vs. VCS vs. (whatever), but about reducing the barriers
to entry.

So, with that in mind, how many steps are required to fix a typo or
add content to the website, and can that amount of steps be reduced?

> Could we please have a voting of some kind whether we do a contest or not?

+1 on a redesign contest

-Ronabop

Sent: 2011-04-23 13:48:38 UTC (about 1 year ago)

From: Alexandre Prokoudine

new website

permalink

On 4/23/11, Ronald Chmara wrote:

>> There is no silver bullet, there is only patience, passion and
>> dedication. If people don't have it, no CMS will substitute that.
>
> These two thoughts seem to be somewhat in conflict.
>
> A CMS arguably makes it easier to get into website development for
> people who don't have the "patience, passion and dedication" to use
> fairly convoluted workflows to do something as simple as updating,
> managing, and contributing web content.

Why would you want people without patience, passion and dedication to
be in charge of something you care about?

> That being said, a CMS does not produce volunteers.... It merely
> reduces the barriers a prospective volunteer has to cross in order to
> become an active participant, so perhaps the conversation is not so
> much about CMS vs. VCS vs. (whatever), but about reducing the barriers
> to entry.

Yes, but again -- we have noone to reduce them for.

> So, with that in mind, how many steps are required to fix a typo or
> add content to the website, and can that amount of steps be reduced?

With existing approach or with (possible) future static HTML + JavaScript? :)

I think I'm utterly misunderstood -- I'm not against CMS. I'm fine
with both static site and CMS. The question is whether we have someone
who is willing to take full responsibility for availability of the CMS
based website. With CMS the question is not _whether_ we'll be hacked,
but _when_. And when it happens, we shouldn't be running around like
headless chickens, not knowing what to do. We need someone who will
always be up to most recent security patches.

For Audacity we solved that by paying a community member who is
experienced in that and can take care of issues. audacityteam.org was
recently hacked, and he fixed that within half an hour.

For Inkscape we are delegating that to OSUOSL whose team member is
part of our community as well.

For GIMP -- I don't think we have anybody right now to delegate it to.

>> Could we please have a voting of some kind whether we do a contest or not?
>
> +1 on a redesign contest

Thanks :)

Alexandre Prokoudine
http://libregraphicsworld.org

Sent: 2011-04-24 10:47:32 UTC (about 1 year ago)

From: Ronald Chmara

new website

permalink

On Sat, Apr 23, 2011 at 6:48 AM, Alexandre Prokoudine
wrote:
> On 4/23/11, Ronald Chmara wrote:
>> A CMS arguably makes it easier to get into website development for
>> people who don't have the "patience, passion and dedication" to use
>> fairly convoluted workflows to do something as simple as updating,
>> managing, and contributing web content.
> Why would you want people without patience, passion and dedication to
> be in charge of something you care about?

Because "patience, passion and dedication" means nothing when it comes
to fixing a typo, posting a random tutorial, uploading a tip, etc.

This is basically the wikipedia argument. Again.

If you only trust proven expert volunteers, that's all you're going to
get. It's slow. It doesn't work without lots of money.

Pity that wikipedia was such a failure, because they didn't demand
proven experts.

(Do I need a sarcasm tag?)

>> That being said, a CMS does not produce volunteers.... It merely
>> reduces the barriers a prospective volunteer has to cross in order to
>> become an active participant, so perhaps the conversation is not so
>> much about CMS vs. VCS vs. (whatever), but about reducing the barriers
>> to entry.
> Yes, but again -- we have noone to reduce them for.

Hi. I'd like to add a GIMP mirror, but the page on the topic is so
hosed that I cannot figure out how to do so. I have at least 10 years
of experience in apache, PHP, (blah blah blah) but I don't even know
how to *apply* to fix it, or who to talk to.

Yes, I'm just one person. How many of me have been shut out?

>> So, with that in mind, how many steps are required to fix a typo or
>> add content to the website, and can that amount of steps be reduced?
> With existing approach or with (possible) future static HTML + JavaScript? :)

I really don't care about the approach, but others might.

Others who can help.

> I think I'm utterly misunderstood -- I'm not against CMS. I'm fine
> with both static site and CMS. The question is whether we have someone
> who is willing to take full responsibility for availability of the CMS
> based website. With CMS the question is not _whether_ we'll be hacked,
> but _when_.

LOL, use a webserver, you'll be exploited..

With all tech, it's a question of "when". Some are more hardened than
others, but it's always "when".

> And when it happens, we shouldn't be running around like
> headless chickens, not knowing what to do. We need someone who will
> always be up to most recent security patches.

Or we need a team.

The concept of "someone" leads to failure, over time. Putting it into
the hands of one person creates a failure point.

> For Audacity we solved that by paying a community member who is
> experienced in that and can take care of issues. audacityteam.org was
> recently hacked, and he fixed that within half an hour.
> For Inkscape we are delegating that to OSUOSL whose team member is
> part of our community as well.
> For GIMP -- I don't think we have anybody right now to delegate it to.

This is why a team is needed. Folks check out. They come back. Or they
don't. Team leader roles change. As long as enough people are
involved, though, there is a team.

-Ronabop

Sent: 2011-04-24 16:57:49 UTC (about 1 year ago)

From: Alexandre Prokoudine

new website

permalink

On 4/24/11, Ronald Chmara wrote:

>> Why would you want people without patience, passion and dedication to
>> be in charge of something you care about?
>
> Because "patience, passion and dedication" means nothing when it comes
> to fixing a typo, posting a random tutorial, uploading a tip, etc.

It's very simple to setup typo notification. The usual selection and
Ctrl+Enter trick.

Random tutorials? i don't think it's a good idea.

Tips? We don't have tips on the website.

>> Yes, but again -- we have noone to reduce them for.
>
> Hi. I'd like to add a GIMP mirror, but the page on the topic is so
> hosed that I cannot figure out how to do so.

*sigh*

http://www.gimp.org/webmasters.html

"You can also use Bugzilla to notify us if a mirror of ftp.gimp.org is
gone or if a new one has recently been added. If you want to set up a
new mirror, you may want to use one of the existing rsync mirrors to
update your server. Please include the country of your server in the
summary of the bug."

What's so hosed about this description? You don't know how to use a
bug tracker with your...

> at least 10 years of experience in apache, PHP, (blah blah blah)

???

> I don't even know how to *apply* to fix it, or who to talk to.

Did you really try to read the page?

> Yes, I'm just one person. How many of me have been shut out?

As far as I can see, the only mail from you before is here:

http://comments.gmane.org/gmane.comp.video.gimp.web/1412

You were pointed to very clear instructions, you got puzzled even so
and you _never_ mailed again to say you still needed clarifications.

I'm sorry, but it looks like _you_ shut yourself out.

>> And when it happens, we shouldn't be running around like
>> headless chickens, not knowing what to do. We need someone who will
>> always be up to most recent security patches.
>
> Or we need a team.

Exactly. That's what I've been talking about from the very beginning.

Alexandre Prokoudine
http://libregraphicsworld.org

Sent: 2011-04-25 16:35:39 UTC (about 1 year ago)

From: Akkana Peck

new website

permalink

Ronald Chmara writes:
> Because "patience, passion and dedication" means nothing when it comes
> to fixing a typo, posting a random tutorial, uploading a tip, etc.
>
> This is basically the wikipedia argument. Again.
>
> If you only trust proven expert volunteers, that's all you're going to
> get. It's slow. It doesn't work without lots of money.

The trade-off with a CMS is that it makes things easier for
casual contributors -- people fixing typos or minor errors --
while raising the bar for site maintainers. You no longer merely
need someone who can type "apt-get update && apt-get safe-upgrade";
now you need somehow who's familiar with the particular CMS and all
the specifics of its local installation.

At least that's been my experience with other organizations. Some
passionate advocate of Zope or Drupal or Moodle or whatever comes
in and sets things up. It sounds like a great idea at the time. A
year later, something breaks, or the CMS is out of date and giving
security warnings, but the original person has gotten busy with other
projects and there's no volunteer with a clue how to fix/upgrade it.

If GIMP.org does switch to a CMS, I'd recommend making sure that
whoever sets it up fully documents where everything is on the server
(software, content, CSS, plug-ins), how to upgrade it, any plug-ins
used and how to install or upgrade those plug-ins. And don't do it
unless you have at least two people (preferably in very different
time zones) committed to maintaining it: if you can't find a second
maintainer now, how will you find one when the site is down and you
need an emergency fix?

...Akkana

Sent: 2011-04-25 23:05:50 UTC (about 1 year ago)

From: Ronald Chmara

new website

permalink

On Sun, Apr 24, 2011 at 9:57 AM, Alexandre Prokoudine
wrote:
> On 4/24/11, Ronald Chmara wrote:
>>> Why would you want people without patience, passion and dedication to
>>> be in charge of something you care about?
>> Because "patience, passion and dedication" means nothing when it comes
>> to fixing a typo, posting a random tutorial, uploading a tip, etc.
> It's very simple to setup typo notification. The usual selection and
> Ctrl+Enter trick.
> Random tutorials? i don't think it's a good idea.
> Tips? We don't have tips on the website.

Yes, it's a thin site. Possibly because adding content is not simple.

>>> Yes, but again -- we have noone to reduce them for.
>> Hi. I'd like to add a GIMP mirror, but the page on the topic is so
>> hosed that I cannot figure out how to do so.
> *sigh*
> http://www.gimp.org/webmasters.html

See:
http://www.gimp.org/downloads/

Bottom of the page, where mirrors should be:
"[an error occurred while processing this directive]"

Been busted for a few days, at least,

> What's so hosed about this description? You don't know how to use a
> bug tracker with your...
>> at least 10 years of experience in apache, PHP, (blah blah blah)
> ???

Bug filed now.
https://bugzilla.gnome.org/show_bug.cgi?id=648650

>> I don't even know how to *apply* to fix it, or who to talk to.
> Did you really try to read the page?

Different page, different problem.

>> Yes, I'm just one person. How many of me have been shut out?
> As far as I can see, the only mail from you before is here:
> http://comments.gmane.org/gmane.comp.video.gimp.web/1412

Wow, that takes me back. Whole different company. Not sure if I ever
got a gimp mirror there. Thanks for finding it, though.

Here is what I was referring to:
https://lists.xcf.berkeley.edu/lists/gimp-web/2011-April/000968.html

> You were pointed to very clear instructions, you got puzzled even so
> and you _never_ mailed again to say you still needed clarifications.
> I'm sorry, but it looks like _you_ shut yourself out.

That will happen too.

-Ronabop

Sent: 2011-04-26 05:14:21 UTC (about 1 year ago)

From: Carol Spears

new website

permalink

On Sun, Apr 24, 2011 at 08:57:49PM +0400, Alexandre Prokoudine wrote:

Alexandre, i have a question to ask you and i think it is a question that
should be answered in a semi-public venue -- this mail list should suffice...

while i was online recently and chatting with you in private, you asked me
if i might "be Akkana Peck". now that she is responding on this mail list
(I have not read that mail yet, btw) i am wondering if indeed you have found
someone to "be Akkana Peck".

i met Akkana or a person "being Akkana" in 2004. i have not seen that person
since but still have photographs and a memory of a gift that was exchanged. i
might be very very wrong about this, but the person who was "being Akkana Peck"
in 2003 on the internet and the person who was "being Akkana Peck" who i spoke
with in person in 2004 would never ever have 1) put me on /ignore or
2) continued to live locally and not have contacted me again.

it is a shame and one might say, even a tragedy that her book seems to not be
stocked by bookstores locally -- perhaps the bay area has too many of such
authors; where i came from it seemed that local authors had the support of the
community bookstores, but again, i might be wrong about that.

perhaps the whole situation can be summed up by these words that the person
being "Akkana Peck's" husband said the last time i saw them in 2004. Those
words were "This will never work."

Alexandre Prokoudine, did you find some one to "be Akkana Peck"?

carol

Sent: 2011-04-26 15:28:24 UTC (about 1 year ago)

From: Michael Schumacher

new website

permalink

> Von: Ronald Chmara

> >> Yes, I'm just one person. How many of me have been shut out?
> > As far as I can see, the only mail from you before is here:
> > http://comments.gmane.org/gmane.comp.video.gimp.web/1412
>
> Wow, that takes me back. Whole different company. Not sure if I ever
> got a gimp mirror there. Thanks for finding it, though.

Most of my changes to the gimp-web module have been mirror additions:

http://git.gnome.org/browse/gimp-web/log/?qt=grep&q=mirror

> Here is what I was referring to:
> https://lists.xcf.berkeley.edu/lists/gimp-web/2011-April/000968.html

Recently I've reduced the number of changes to gimp-web, simply because it's a bit frustrating to wait for manual updates of the website (i.e. wait until someone runs the make command manually).

I'd guess that a simple cron-based approach could fix this (update once every 30 minutes, for example), even if this isn't as fast as the commit-event triggered updates we had previously.

Regards,
Michael

NEU: FreePhone - kostenlos mobil telefonieren und surfen!
Jetzt informieren: http://www.gmx.net/de/go/freephone