For sure the project is not stalled - you can watch the progress on http://www.gimpusers.com/ or you can read articles like this one http://libregraphicsworld.org/blog/entry/basic-gpu-side-rendering-and-processing-in-gimp-available on libre graphics world website.

But your point is true ..there are a lack of qualified manpower to speed-up the development and a daily growing list of requests from the userland.

2012/3/4 Paul Slocum :

I've been a serious professional GIMP user for about 10 years, but the project has seem stalled for quite some time now.  I was wondering how much money it would take to get the project back on track?  Would 25k, 50k, 100k to hire one or two programmers for several months make a substantial difference?

-paul
_______________________________________________ gimp-developer-list mailing list
gimp-developer-list@gnome.org
http://mail.gnome.org/mailman/listinfo/gimp-developer-list

So what kind of money do you think would it take to catch up? Or to at least make a significant step towards catching up?

Whatever it takes to get one or two of our lead developers to work on GIMP full time. Just hiring 2 blokes off the street will not do. Michael Natterer has done some paid for work for GTK through his company I believe, but as far as I know he is not interested in turning a hobby into a job... Because jobs come with a boss who can boss you around and that is no fun. Same may be true for others.

Having a paid developer on the team has been discussed before. there is no actual opposition to the idea, but nobody has stepped up to handle the paperwork and manager duties:P

--Alexia
_______________________________________________
gimp-developer-list mailing list
gimp-developer-list@gnome.org
http://mail.gnome.org/mailman/listinfo/gimp-developer-list

Hi

On Mon, Mar 05, 2012 at 08:41:08AM +0200, Alexia Death wrote:

So what kind of money do you think would it take to catch up?  Or to at least make a significant step towards catching up?

Whatever it takes to get one or two of our lead developers to work on GIMP full time. Just hiring 2 blokes off the street will not do. Michael Natterer has done some paid for work for GTK through his company I believe, but as far as I know he is not interested in turning a hobby into a job... Because jobs come with a boss who can boss you around and that is no fun. Same may be true for others.

Having a paid developer on the team has been discussed before. there is no actual opposition to the idea, but nobody has stepped up to handle the paperwork and manager duties:P

The paperwork and manager duties are straightforward.

I'd like to first see a serious proposal first on what an investor wants out of it, and what money they're willing to put in. Talking briefly about investing is one thing, but making schedules and projections for them takes time. If there is a serious interest, we can talk about it. Tell us about your background.

Some developers will indeed be interested in working on GIMP and GEGL full-time, and IMO, it'll be good for GIMP and GEGL's development generally. Other free software projects are prospering due to commercial development.

Mukund

On Mon, Mar 5, 2012 at 10:41 AM, Alexia Death wrote:

Having a paid developer on the team has been discussed before. there is no actual opposition to the idea, but nobody has stepped up to handle the paperwork and manager duties:P

I told you: SFC can do that :)

Alexandre Prokoudine http://libregraphicsworld.org

On Mon, Mar 5, 2012 at 9:37 AM, Alexandre Prokoudine wrote:

On Mon, Mar 5, 2012 at 10:41 AM, Alexia Death wrote:

Having a paid developer on the team has been discussed before. there is no actual opposition to the idea, but nobody has stepped up to handle the paperwork and manager duties:P

I told you: SFC can do that :)

Only if someone of infuluence from our side is willing to set it up. Currently that person would need to be mitch...

Paul Slocum wrote:

I've been a serious professional GIMP user for about 10 years, but the project has seem stalled for quite some time now. I was wondering how much money it would take to get the project back on track? Would 25k, 50k, 100k to hire one or two programmers for several months make a substantial difference?

can I point out a couple of things?

first, it has been pointed out for years that that would build a two-class society in GIMP: paid and very active (and contribution is power in free software), vs. unpaid and occasional contributor (or more tragic, unpaid and steady contributor; how long would that last?)

next, software engineering is only one single piece of the whole puzzle of shipping software. I have a lot of respect for the people who write the documentation; for the people who do triage in bugzilla; for those who run the SoC; for those who organise and do localisation; I probably forget some more who do similar hard, nagging work that involves quite a bit of managing processes. all of this is not seen as development, which is already a put-down for these people. add another one on top; that it would not speak for itself to pay to get this done?

then there is my team, the UI team. and related, the people undertaking quite a bit of usability research at this moment. As a professional, I know what all that is worth, both in what it delivers to the project and what it costs in the real world: a substantial amount. all of this is contributed at the moment with the understanding that there is no money going around in GIMP (donations are used for travel to bring contributors together and for servers and hosting).

I would not like to see that understanding being broken.

the reason our (m+mi works) contribution of years to openPrinting came to an end, was that I realised that everyone was paid to contribute to open source printing (_no_one_ work voluntarily in printing) except for us, the interaction design team, who were dragging printing out of the 1980s (kicking and screaming). meanwhile there was a lot of pressure on us from these paid folks to make progress, but not a dollar to make it happen. then something snapped.

I won't get fooled again. if there is money for the engineering of a project, then there better be real ($) appreciation of what interaction design is worth.

my conclusion is to let pandora's box of paid development closed.

--ps

founder + principal interaction architect man + machine interface works

http://blog.mmiworks.net: on interaction architecture

On Mon, Mar 5, 2012 at 3:56 PM, peter sikking wrote:

Paul Slocum wrote:

I've been a serious professional GIMP user for about 10 years, but the project has seem stalled for quite some time now. I was wondering how much money it would take to get the project back on track? Would 25k, 50k, 100k to hire one or two programmers for several months make a substantial difference?

can I point out a couple of things?

first, it has been pointed out for years that that would build a two-class society in GIMP: paid and very active (and contribution is power in free software), vs. unpaid and occasional contributor (or more tragic, unpaid and steady contributor; how long would that last?)

Blender Foundation has been able to deal with that for years and it's by far the most active free graphics project.

Whether we are able to organize things same efficiently is a whole different topic :)

next, software engineering is only one single piece of the whole puzzle of shipping software. I have a lot of respect for the people who write the documentation; for the people who do triage in bugzilla; for those who run the SoC; for those who organise and do localisation; I probably forget some more who do similar hard, nagging work that involves quite a bit of managing processes. all of this is not seen as development, which is already a put-down for these people. add another one on top; that it would not speak for itself to pay to get this done?

Well, I do some of what you mentioned and I have absolutely no problems with somebody else getting paid to develop GIMP.

then there is my team, the UI team. and related, the people undertaking quite a bit of usability research at this moment. As a professional, I know what all that is worth, both in what it delivers to the project and what it costs in the real world: a substantial amount. all of this is contributed at the moment with the understanding that there is no money going around in GIMP (donations are used for travel to bring contributors together and for servers and hosting).

I would not like to see that understanding being broken.

I'm not sure what you mean with money going around. If a hired programmer interacted with your team, would that be a bad thing for the understanding?

the reason our (m+mi works) contribution of years to openPrinting came to an end, was that I realised that everyone was paid to contribute to open source printing (_no_one_ work voluntarily in printing) except for us, the interaction design team, who were dragging printing out of the 1980s (kicking and screaming). meanwhile there was a lot of pressure on us from these paid folks to make progress, but not a dollar to make it happen. then something snapped.

My impression was that it was because nobody actually used your spec :)

I won't get fooled again. if there is money for the engineering of a project, then there better be real ($) appreciation of what interaction design is worth.

I can't see why both UX team and a programmer couldn't be paid.

my conclusion is to let pandora's box of paid development closed.

OK, here is my take.

It's been a 1+ year since I deliberately raised public awereness of lack of developers in the team. It got quite a coverage in the online press, including Phoronix. We've been mentioning lack of developers in the public all the time ever since then.

What's the outcome?

We haven't experienced any substantial raise of active developers. We haven't released 2.8 yet, after 3+ years of having it in works. We wouldn't get a lot of the new features in 2.8/2.10 without GSoC.

BTW, GSoC _already_ is paid development_and_ you already worked with at least one of the students :)

So, would you say that the project manages itself just fine? Because the way things are going we are not going to deliver. We've been failing expectations for years. This has got to stop.

Alexandre Prokoudine http://libregraphicsworld.org

gimp-developer-list mailing list
gimp-developer-list@gnome.org
http://mail.gnome.org/mailman/listinfo/gimp-developer-list

On Mon, Mar 5, 2012 at 9:14 PM, Paul Slocum wrote:

Personally, what I really need is CMYK

Everyone wants CMYK, few mention what they need specifically :)

and layer folders.

Already done.

Alexandre Prokoudine http://libregraphicsworld.org

Alexandre wrote:

Personally, what I really need is CMYK

Everyone wants CMYK, few mention what they need specifically :)

yeah, beware of the schmuck:

--ps

founder + principal interaction architect man + machine interface works

http://blog.mmiworks.net: on interaction architecture

Need easy tasks page

Libreoffice success not only because the paid developers but also because they create a wiki page which contains a lot of easy tasks in categories from beginners to advances so they encourage a lot of people to do small tasks which together become big advance for the libreoffice.

It much easier to commitment for a one task rather than consistently in a big project. I'm also contribute code to some projects like libreoffice and KDE just because the small tasks, and there are a many other programmers who can write code from time to time but don't want to commitment regularly and some of them may become commitment when they feel better with the code and so on.

I follow the commits of gimp and other projects and most of the commits are not implementing special algorithms and so on. tasks like converting units from one API to another, like porting to GTK 3 are good to be in those easy tasks.

and the most important issue is that the main developers can focus in the major gimp development like writing new features and so on

On 5 March 2012 19:40, peter sikking wrote:

Alexandre wrote:

Personally, what I really need is CMYK

Everyone wants CMYK, few mention what they need specifically :)

yeah, beware of the schmuck:

   --ps

       founder + principal interaction architect            man + machine interface works

       http://blog.mmiworks.net: on interaction architecture

_______________________________________________ gimp-developer-list mailing list
gimp-developer-list@gnome.org
http://mail.gnome.org/mailman/listinfo/gimp-developer-list

On Mon, Mar 5, 2012 at 11:19 AM, Alexandre Prokoudine wrote:

Personally, what I really need is CMYK

Everyone wants CMYK, few mention what they need specifically :)

Early binding. Exporting to CMYK (for publishers who only accept CMYK files). Rich black vs 100% K. "Hard" soft proof (for clients who wish to print 0,0,255 and need to be shown the inevitable color shift). Overprinting. I'm sure I could dig up a couple more ;)

On topic: I'm all for paying developers and/or hiring the right extra people. As others have pointed out, several FOSS projects have successfully integrated paid development. If I really knew C, I'd go part or full time in a heartbeat.

But I just lurk here - don't mind me ;)

Chris

With all due respect, your method of not paying anyone has resulted in 2 years without a stable release of GIMP. What's your point? It's not like things are just rosy and there aint nothing to fix.

On Mon, Mar 5, 2012 at 10:05 PM, Jeremy Morton wrote:

With all due respect, your method of not paying anyone has resulted in 2 years without a stable release of GIMP. What's your point? It's not like things are just rosy and there aint nothing to fix.

Failure to release in 2 years is not a problem curable with money. Long development cycle is the result of merging more than a few large features early on that took time to mature and stabilize. It's a result of going for "perfect completeness in one go", instead of taking small simple and above all, releasable steps. We will try to address this shortcoming in the next few cycles by trying to keep the main tree in a "releasable" state at all times and add new stuff in small self-contained functional but perhaps feature incomplete chunks. I know Peter disagrees with me and many others on this, because he wants to see his babies "complete", but that's exactly the thinking that got us this long cycle and exactly the thing we need to avoid. Adding more developers, specially paid ones without addressing the causes is not going to improve matters much unfortunately. there will be more things waiting around for those few touches to become releasable pushing the deadline ahead.

--Alexia
_______________________________________________
gimp-developer-list mailing list
gimp-developer-list@gnome.org
http://mail.gnome.org/mailman/listinfo/gimp-developer-list

2012/3/5 Chris Mohler

Early binding. Exporting to CMYK (for publishers who only accept CMYK files). Rich black vs 100% K. "Hard" soft proof (for clients who wish to print 0,0,255 and need to be shown the inevitable color shift). Overprinting. I'm sure I could dig up a couple more ;)

That could be addressed with an intermediate binding workflow, and the idea (already proposed) of creating special projections for other color spaces like CMYK would be compatible with that. Rich black, pure black, overprinting is more about using CMYK channels as spot channels rather than creating color managed separations. Some time ago I proposed some rather simple changes for the Separate+ plugin to make that possible (it's already possible to create pure blacks, overprints, even rich blacks and pure CMYK primaries with Separate+ but in a manual, tedious and error prone way using the pseudo-channels). So, back to the topic: some money sponsoring those changes in the plugin would give users a tool that would make it possible to create CMYK files with intermediate binding covering all those common needs (overprint, rich black, using CMYK primaries as spot channels, pure black/grays, etc.)

Someone wrote:

So what kind of money do you think would it take to catch up? Or to at least make a significant step towards catching up?

No amount of money will help if there is no one available with the skills, level of knowledge of GIMP's source code, and the time to work on the project. I think the people with enough knowledge of GIMP's internals and the who have the time to work on the project are already doing so.

On Sat, Mar 10, 2012 at 12:01 AM, Kevin Cozens wrote:

Someone wrote:

So what kind of money do you think would it take to catch up? Or to at least make a significant step towards catching up?

No amount of money will help if there is no one available with the skills, level of knowledge of GIMP's source code, and the time to work on the project. I think the people with enough knowledge of GIMP's internals and the who have the time to work on the project are already doing so.

I know at least one important contributor who expressed an interest in doing paid development.

Starting paid development means changing a lot in how the project works. I understand that the way things are going now seem easier to everyone, because right now noone really takes a formal/written responsibility for anything and would rather be happy to see this state of affairs to continue.

Yes, part of the 2.8 issue is because we tried to do to many things at once. But I don't really think that branch-based development is going to cure everything.

We practically lost Enselic who was supposed to do all the heavy-lifting in the final migration to GEGL and implement some of the much desired hi-end features. What's been happening for last several years is a loss of credibility for the project. We can only cure that with that final switch. And only temporarily. Right now it looks like it's still bloody years away.

If we can't organise the project in a way that developers come themselves, then saying a strict "no" to paid development while allowing GSoC (which _is_ paid development) seems silly to me.

That said, we haven't really done everything to attract new developers, and I suggest that we start with tying loose ends:

- better intro docs - more hackfests

Alexandre Prokoudine http://libregraphicsworld.org

gimp-developer-list mailing list
gimp-developer-list@gnome.org
http://mail.gnome.org/mailman/listinfo/gimp-developer-list

On Sat, Mar 10, 2012 at 12:25 AM, Alexandre Prokoudine wrote:

No amount of money will help if there is no one available with the skills, level of knowledge of GIMP's source code, and the time to work on the project. I think the people with enough knowledge of GIMP's internals and the who have the time to work on the project are already doing so.

I know at least one important contributor who expressed an interest in doing paid development.

After rereading my inbox: two important contributors.

Alexandre Prokoudine http://libregraphicsworld.org

On 03/09/2012 09:01 PM, Kevin Cozens wrote:

Someone wrote:

So what kind of money do you think would it take to catch up? Or to at least make a significant step towards catching up?

No amount of money will help if there is no one available with the skills, level of knowledge of GIMP's source code, and the time to work on the project.

Money can buy you time: Depending on where you live, a fulltime paid developer could invest 35 to 45 hours a week into the project without having to fear that there is no money at the end of the month to pay the rent. Compared to a developer being able to invest some of his/her spare time only, this can be a noticeable difference on productivity per month.

Choosing a skilled and knowledgeable developer and guide/motivate/control him/her to the benefit of the project goals at hand is another challenge altogether. This process sometimes is referred to as "herding cats".

Just saying ,,,

Regards

Stefan Peter

Hi

here are some thoughts from me to the topic and postings before:

1. Solve the real problems: Full acknowledgement with Alexia and Alexandre to first solve the real problems that caused the delay. I like the proposal for the further Git branching and the proposal to iteratively check-in in small steps instead of big features. In my opinion you're on the right way. Before every new feature version a decision should be made, whether the feature will be part of the next GIMP's version at all, only with its working parts or in complete. If it's decided to include the feature in the new version, only the tested and working parts should be accessible from the UI. This will save GIMP from complaints about missing or buggy functionality and let developers force their development. Users and people involved here will be more pleased.

2. Yes to employed persons, further support possibilities: IMHO it's good for the product GIMP to have at least one employed person. If we want a high end product, as I can read at http://gui.gimp.org/index.php/GIMP_UI_Redesign#product_vision, this would be simply necessary. This can be a programmer, project manager, product manager, coordinator etc. The people involved here for a long time might know best what is missing. For the payment I see various possibilities: - donations (which already exists). I think Sourceforge provides something suitable, too.
- regular sponsoring by a company
- a company employing this person/these persons - social payment through Pledgie, Flattr etc. - a GIMP E-Book like 'Digikam Recipes' which proceeds are used to support GIMP production
- producers of commercial GIMP products, like book authors and publishers. Why not invest and return something back after benefiting from a free product?
- an own foundation, which can care for fundraising and organization. Look at the Eclipse project – they are very successful with this. Eclipse project releases a service pack every four months a new version and every year. This foundation could care for GIMP only or you unite with other free graphics projects for this. You could discuss this on the next LGM.

Perhaps the Free Software Foundation (www.fsf.org) or the Software Freedom Conservancy (sfconservancy.org) can support us financially or as mentors, too?

3. Next steps
Why do we not just start here to list what is specifically necessary, what do we already have und what is still needed? Then everybody here can look out for concrete solutions. Here's my proposal, partially summarizing points from the former postings in this thread:

Needed: - motivated and good developers with the skills, level of knowledge of GIMP's source code, and the time to work on the project. Which else skills over C (and Python or Scheme) knowledge do they need? - somebody with the big picture of GIMP's architecture - a project manager with influence, but not bossing people around - somebody who can care for the 'paper stuff' for the employees - overview of easy to solve tasks
- funding to pay all employees, but not only developers - an attractive project organisation, which entices contributors - … (add your thoughts here)

What we already have: - experienced developers, doing their job for free and in spare time - developers willing to contribute permanently - people (able and) willing to spend money - UI experts
- former GSoC students and hopefully new students on the next GSoC - artists and their artwork
- experienced users here and in various forums - a stable GIMP 2.6 and lots of useful plugins - people reporting bugs
- users of the official GIMP and curious users for new features - ideas, ideas, ideas
- … (add your findings here)

So, now it's on you to make more out of this list.

grafxuser

On Sun, 2012-03-11 at 19:52 +0100, grafxuser wrote:

Needed:
- motivated and good developers with the skills, level of knowledge of GIMP's source code, and the time to work on the project. Which else skills over C (and Python or Scheme) knowledge do they need?

Some idea about the product vision, and about image processing. One approach there might be to pay existing gimp people (Peter, Mitch, Margin, Alexia, yvind, Sir Nigel Gresley) to do (between them) a week of intensive immersion to a group of fresh-faced raw recruits :-)

Liam

Liam Quin - XML Activity Lead, W3C, http://www.w3.org/People/Quin/
Pictures from old books: http://fromoldbooks.org/
The barefoot typographer
Ankh on irc.gnome.org freenode#xml irc.sorcery.net#gaycafe

_______________________________________________
gimp-developer-list mailing list
gimp-developer-list@gnome.org
http://mail.gnome.org/mailman/listinfo/gimp-developer-list