On 11/3/07, Barry Loo wrote:

GIMP just got some negative press at linux.com. What are y'alls opinions on it?

http://www.linux.com/feature/120635

This is quite unusual for Nathan to miss the point *that* much.

Alexandre

Barry Loo wrote:

GIMP just got some negative press at linux.com. What are y'alls opinions on it?

http://www.linux.com/feature/120635

As far as GIMP is concerned,

http://www.linux.com/?module=comments&func=display&cid=1169329

and

http://www.linux.com/?module=comments&func=display&cid=1169345

are the appropriate answers. And I do suspect that the icon problem isn't as serious as the author wants it to be, either.

Nathan has obviously been bribed by a closed source company to disrupt the open source community.

Michael

Date: Sat, 3 Nov 2007 14:47:18 -0400 From: "Barry Loo"

GIMP just got some negative press at linux.com. What are y'alls opinions on it?

http://www.linux.com/feature/120635

I've had my share of issues with the GIMP UI (and I've been vocal about it at times), but I think this piece is completely off the mark. It's really no different from a group of people going off into a small room to focus on something rather than being constantly interrupted by everyone passing by. That was a particularly unfortunate out of context quote of Peter Sikking's.

The only one of the three that I've met is Ellen Reitmayr (at the April 2006 printing summit in Atlanta), and my impression was most certainly not that she isn't interested in listening to other people. She and a couple of other of the UI folks spent a couple of hours brainstorming on Gutenprint UI issues.

Contrast the GIMP UI redesign with the GIMP project as a whole, which invites and receives patches, bug reports, and ideas from scores of outsiders.

The focus was on the UI redesign, not GIMP. In fact the quote above compliments GIMP (but at the UI redesign's expense).

I suspect the true "problem" isn't one about the process, but one about the perceived results. When you think about it, people rarely criticize a project Just for the process. People criticize MS Windows for being closed-source and thus full of bugs and functionality problems, but nobody criticizes Apple, which is More closed-source in many aspects, because it actually does a good job.

Basically, even though people do not realize it, their reaction may actually be something of the following: - GUI team: We can handle the work just fine right now, so we don't need extra people in our team.
- Others: (Yeah? Then why's the interface still so bad?)

I say this because after an amount of introspection, I realize that I might have been guilty of this.

I short, I suspect that people are unconsciously blaming GIMP's Current GUI shortcomings on the GUI team, even though it's not their fault because:
- they actually haven't had the opportunity to show what they're fully capable of
- and many GUI problems are actually due to internal architecture limitations (layer groups, brush folders etc)

This unconscious connection between the current GUI and the GUI team's possible accomplishments may give the impression that the GUI team is under-qualified or incapable of handling the job by themselves, which is why outsiders snipe at their reluctance to let others join their team in a more permanent way.

At the base of the issue, there may be a transparency problem. Basically, there is no easy way of tracking the works of the GUI team right now. There are only three ways of seeing what exactly what they're up to:
- reading the ideas they've come up with thanks to the GUI blog submissions (which are only made up of a few lines, aren't that easy to find back, and are relatively rare) - go to gui.gimp.org , click "User evaluation notes," then click "Notes" for individual scenarios, then be confronted with a wall of text on technical evaluations that most users don't want to bother reading.
- go to gui.gimp.org, go to "UI specifications," and find... a total of 1 entry, for 2.4.

Given this lack of easy, visible and regular updates, people "conclude" that little work is being done.

The easiest solution, as far as I can tell, is actually to apply the same principle that the GIMP site's "Feature" page and that the GIMP UI brainstorm blog apply themselves: using pictures to speak a thousand words.

The summary would basically roughly serve the same purpose as a visible 2.6 milestone (which should also have mock-ups) would from a PR point-of-view: show people what's in planning so that people won't think of GIMP as a dead project that isn't moving anywhere. Inkscape has a screenshots section for future features, and that does wonders for showing people the future evolution of the program.

Are there any plans for a "Future feature" page on the GIMP website? If there is, it could be made up of two sections: - Future features (with mock-ups based on the 2.6 milestone) - Future GUI improvements (a whole section dedicated to GUI improvement! This is sure to score points among critics of the GIMP interface)

The future GUI improvement section could contain screenshots of a few key UI improvements in planning (they don't need to include everything), with eventually an explanation of dependencies such as GEGL. As long as people know that they Are being planned, they'll be relatively happy and not get the impression that GIMP doesn't care about UI improvements. If the features aren't planned for any time soon but Are on the long-term plans, an explanation is enough to let users know that the GUI team isn't GUI-stupid but simply isn't capable of implementing the changes Yet.

Then add a call for help on implementing prior dependencies, and maybe you'll even get more developers.

Said section could end with "Got more ideas? Submit to the GUI-brainstorm!" And that's it. PR problem solved. Lack manpower? Get someone else to do the mock-ups for you. Given the number of submissions to GUI-brainstorm, it should be easy enough to find someone.

Add the occasional news update on GUI rework progress, and that's even better! "GIMP is finally taking the GUI problem seriously!" - would claim people who haven't noticed the work already under way.

___

On Sun, 04 Nov 2007 02:26:30 +0100, Tim Jedlicka wrote:

Contrast the GIMP UI redesign with the GIMP project as a whole, which invites and receives patches, bug reports, and ideas from scores of outsiders.

The focus was on the UI redesign, not GIMP. In fact the quote above compliments GIMP (but at the UI redesign's expense).

Hi,

I dont think Nathan's analysis is that far "off the mark". The interaction with the UI team is very one way and definately gives the impression of "please go away , we're busy."

The so called brainstorm blog is more like a super market's suggestion box than a discussion.

There is a definate attempt to keep "outsiders" at arms length and the blog really just looks like it's there so any critisism of not being open can be countered by saying "look we have the blog for ideas".

Whether this approach in understandable or desirable or ultimately more or less productive remains to be seen but I dont see much justification for calling it open or feigning indignation if someone suggests it's not.

Like with a door, there are different degrees of open, it's not a binary condition.

This case in point seems like being allowed to peer through a crack in the door and whisper through the keyhole rather than "come in and see if you can contribute".

How different projects operate is all part this great social experiment we call open source. The source is open in a well defined, legally meaningful way. The openess of the projects themselves seems more of a human psychology experiment. I'm sure the white mice are finding this facinating.

/gg

catking.net> writes:

I dont think Nathan's analysis is that far "off the mark". The interaction with the UI team is very one way and definately gives the impression of "please go away , we're busy."

The so called brainstorm blog is more like a super market's suggestion box than a discussion.

There is a definate attempt to keep "outsiders" at arms length and the blog really just looks like it's there so any critisism of not being open can be countered by saying "look we have the blog for ideas".

Well, I agree with the gist of your message... but one thing needs to be said: Designing a good UI doesn't require the same amount of people that implementing it in code does. I work in a small software house and our software is much more complex than the Gimp. We have exactly 2.5 people [1] who do UI design. But we have dozens of developers.

As for the UI blog, I think it's misleading: it is obvious, from the analysis done by the UI team, that they don't really care about the proposed solutions. This is just a way to find out what *problems* the users are trying to solve. The UI team will come with its own solutions in the end. Like you, I'd like to see them before they are implemented... I myself sometime have UI ideas which I post here, but I never know if the UI team did not think of them already.

[1]I'm the 0.5 - I'm mainly a graphic artist but occasionally help with the UI

Well, I agree with the gist of your message... but one thing needs to be said:
Designing a good UI doesn't require the same amount of people that implementing
it in code does.

This is why I suspect it to be a transparency problem and not really a process problem. People actually Won't criticize a process if they think it is doing a good job. In the case of the GUI team, we don't know if it's doing a good job. In fact, we don't see a job being done at all.

This, of course, is irrational behavior: just because you don't see it, it doesn't mean good work isn't being done. But as long as outsiders don't see what the GUI team is truly capable of -via a few terrific mock-ups or similar- they will simply assume the worst: that the GUI team isn't capable of handling the job on their own, And refuse outside help on top of that.

Solve the transparency problem, and the criticism will go away. Of course, you can just ignore it all and let the results speak for themselves, but then expect to put up with a lot of negative press. Also, it might be a missed opportunity for getting more developers to work on the architectural dependencies needed for implementing some of the GUI changes.

___

On 11/5/07, Valerie VK wrote:

This is why I suspect it to be a transparency problem and not really a process problem. People actually Won't criticize a process if they think it is doing a good job. In the case of the GUI team, we don't know if it's doing a good job. In fact, we don't see a job being done at all.

Who are these "we"?

Clearly I am not one of them, because I benefit from new rectangle tools that come out of a spec created by the UI design team, because I can read things like
http://gimp-brainstorm.blogspot.com/2007/10/team-review-contribution-2650.html etc.

Solve the transparency problem, and the criticism will go away.

You say what to do, but you don't say how.

Alexandre

2007/11/5, Alexandre Prokoudine :

Solve the transparency problem, and the criticism will go away.

You say what to do, but you don't say how.

allowing comments (with moderation... like most blogs) on http://gimp-brainstorm.blogspot.com/ will be a good start. the wiki of the GIMP UI redesign team is closed, so the blog above should be the public face* of this redesign process.

*that's what I was suggesting a new mailing list; because both the blog and wiki are virtually closed to "outsiders"...

As an example of why a GIMP.UI mailing list (or changes in the blog) is necessary

Send your image to us , put the word 'GIMP' in

the title of your email (to avoid spam, emails without GIMP in the title or without an image attachment will not be opened).

So, if I have a comment on this entry: http://gimp-brainstorm.blogspot.com/2007/11/add-or-remove-tool.html

that doesn't include an image manipulation of the images of the post my only option (to avoid being "filtered as spam")

is reply in this mailing list:

This will be a welcome change if implemented, in fact, it is one of the first steps for my idea.
But more changes are necessary. The toolbox should be separated from the second menubar (and merge both menubars) and from the options of each tool. In this way (and if each user has a simplified and personalized toolbox), the toolbox can be a 1xN grid that goes to the border (yey!).

Esteban Barahona wrote:

allowing comments (with moderation... like most blogs) on http://gimp-brainstorm.blogspot.com/ will be a good start.

The idea is that comments are done by images - if you do like something, you can add more suggestions based on it. If you don't like something, you have to create something better.

IMO the best way to cut down noise. Have a look at the comments in the linux.com article, many of them are just BS.

HTH, Michael

On 11/5/07, Esteban Barahona wrote:

You say what to do, but you don't say how.

allowing comments (with moderation... like most blogs) on http://gimp-brainstorm.blogspot.com/ will be a good start.

That won't work. Brainstorm means no discussion. Otherwise it's not a brainstorm.

See, I do understand your best intentions to help and I have my own ideas how to improve GIMP as well, but things that could work to a smal project cannot work to huge projects like GIMP. There are dozens, if not hundreds of us. All together we will make a lot of noise and distract UI designers from actual work. Now that we finally have a group of people working at UI this is the last thing I would want to see happening.

However I believe that the situation could be improved by writing a friendlier text at brainstorm page and providing a friendly explanation how and why UI team works at the main wiki page. Contributions are welcomed by GIMP team and people have to know that for sure.

Alexandre

This is why I suspect it to be a transparency problem and not really a process problem. People actually Won't criticize a process if they think it is doing a good job. In the case of the GUI team, we don't know if it's doing a good job. In fact, we don't see a job being done at all.

Who are these "we"?

Clearly I am not one of them, because I benefit from new rectangle tools that come out of a spec created by the UI design team, because I can read things like

http://gimp-brainstorm.blogspot.com/2007/10/team-review-contribution-2650.html

etc.

The "we" are the 80% users who expect "UI improvement" to be a lot more than just a new design for the rectangular tools, and won't bother reading every single post on the gimp-brainstorm blog, much more so since the only link is in the pile of "ways to contribute" list which does Not include "Check here for future UI improvement plans."

Solve the transparency problem, and the criticism will go away.

You say what to do, but you don't say how.

Yes I did, in perhaps such an extensive manner that you didn't bother reading it from my previous response to this topic. Here it is again since you've missed it:

The easiest solution, as far as I can tell, is actually to apply the same principle that the GIMP site's "Feature" page and that the GIMP UI brainstorm blog apply themselves: using pictures to speak a thousand words.

The summary would basically roughly serve the same purpose as a visible 2.6 milestone (which should also have mock-ups) would from a PR point-of-view: show people what's in planning so that people won't think of GIMP as a dead project that isn't moving anywhere. Inkscape has a screenshots section for future features, and that does wonders for showing people the future evolution of the program.

Are there any plans for a "Future feature" page on the GIMP website? If there is, it could be made up of two sections: - Future features (with mock-ups based on the 2.6 milestone) - Future GUI improvements (a whole section dedicated to GUI improvement! This is sure to score points among critics of the GIMP interface)

The future GUI improvement section could contain screenshots of a few key UI improvements in planning (they don't need to include everything), with eventually an explanation of dependencies such as GEGL. As long as people know that they Are being planned, they'll be relatively happy and not get the impression that GIMP doesn't care about UI improvements. If the features aren't planned for any time soon but Are on the long-term plans, an explanation is enough to let users know that the GUI team isn't GUI-stupid but simply isn't capable of implementing the changes Yet.

Then add a call for help on implementing prior dependencies, and maybe you'll even get more developers.

Said section could end with "Got more ideas? Submit to the GUI-brainstorm!" And that's it. PR problem solved. Lack manpower? Get someone else to do the mock-ups for you. Given the number of submissions to GUI-brainstorm, it should be easy enough to find someone.

Add the occasional news update on GUI rework progress, and that's even better! "GIMP is finally taking the GUI problem seriously!" - would claim people who haven't noticed the work already under way.

___

Hi,

On Mon, 2007-11-05 at 22:34 -0800, Valerie VK wrote:

Are there any plans for a "Future feature" page on the GIMP website?

You obviously completely missed that we are currently discussing the roadmap for 2.6 on this mailing-list. The goal is to get this task list published by the end of this week. So could we please stop talking about some completely irrelevant article and instead deal with the actual problem? Thank you.

Sven

Alexandre Prokoudine gmail.com> writes:

Solve the transparency problem, and the criticism will go away.

You say what to do, but you don't say how.

Very simple: Peter has a blog, right? and very occasionaly, he posts something relevant to the Gimp UI, such as the post about the print dialog. To add more transparency to the UI effort, all he needs to do is to post some more. And not just on relatively small issues such as the print dialog. The big questions are what most people are anxious to know about (If you look at the brainstorm, these are the questions that most people are trying to address):

what will the future toolbox look like? What windowing paradigm will Gimp use? SDI? MDI? Tabbed documents? Where will the menus be?
will dialogs dock to the edges of the canvas?

Even if non of these is as yet agreed upon, he could post his (and the rest of the team's) current lines of thinking and options under consideration.

Sven wrote:

So could we please stop talking about some completely irrelevant article and instead deal with the actual problem? Thank you.

since I am responsible for this 'department', I do want to say something, but I'll keep it short.

I do take the hart of the matter, which is behind that article and some of the comments in this thread, serious.

there will be more innovation on the communication side from the UI team. thinking about expanding the UI team too (warning: interaction professionals only).

the brainstorm itself is a method to open up the UI process and the team reviews are there to show what we get from them (apart from documenting it for ourselves). both were received enthusiastically. takes 2-3 hours btw, to discuss 25 brainstorm contributions. that should be communicated too.

--ps

founder + principal interaction architect man + machine interface works

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

I wrote:

I do take the hart of the matter [...] serious.

really, I did not mean "an adult male deer", I meant 'the heart of the matter...'

--ps

founder + principal interaction architect man + machine interface works

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

Michael Grosberg wrote:

Very simple: Peter has a blog, right? and very occasionaly, he posts something
relevant to the Gimp UI, such as the post about the print dialog.

uhm, the print dialog stuff is for openPrinting. That project plots the future of printing for all linux desktop systems.

To add
more transparency to the UI effort, all he needs to do is to post some more.

and that, is exactly my plan...

--ps

founder + principal interaction architect man + machine interface works

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

On Tue, 06 Nov 2007 22:47:12 +0100, peter sikking wrote:

I wrote:

I do take the hart of the matter [...] serious.

really, I did not mean "an adult male deer", I meant 'the heart of the matter...'

you probably also meant "seriously". I'm surprised an intelligent dutchman would make that careless mistake , it's usually only uneducated native speakers that use adverbs in place of adjectives.

It must be late. ;)

Seriously, I'm glad you found the discussion useful. I'm sure a bit more communication will be appreciated.

regards, gg