Sign up now! · Forgot password?
RSS/Atom feed identi.ca Twitter

hesitant about compiling a list...

This discussion is connected to the gimp-developer-list.gnome.org mailing list which is provided by the GIMP developers and not related to gimpusers.com.

This is a read-only list on gimpusers.com so this discussion thread is read-only, too.

7 of 7 messages available
Toggle history

Please log in to manage your subscriptions.

hesitant about compiling a list... peter sikking 17 May 14:19
  hesitant about compiling a list... Nicolas Robidoux 17 May 14:30
  hesitant about compiling a list... gfxuser 17 May 14:55
  hesitant about compiling a list... Øyvind Kolås 19 May 12:32
  hesitant about compiling a list... Jon Nordby 19 May 13:59
   hesitant about compiling a list... Bruce 05 Jun 09:41
    hesitant about compiling a list... peter sikking 05 Jun 12:21
peter sikking
2012-05-17 14:19:30 UTC (over 2 years ago)

hesitant about compiling a list...

guys,

a couple of days ago Bruce appeared on irc and we had a chat. (Bruce is now subscribed to this list)

within a minute we were talking about whether GIMP has a list of known UI design issues. as far as I know we do not have one, it is certainly not part of gui.gimp.org.

(I am not talking about a bug list, with nuts and bolts issues, I am talking about a list of medium level to big-picture issues and to-dos. the kind of design tasks I pick for my teaching are medium size ones, see
)

Immediately I realised this list is missing and that there are real benefits to maintaining a list like that:

- a clear list of design tasks to pick one from for people who want to contribute or run projects in interaction design;

- GIMP as a project says clearly yes, we know we have problem with XYZ in the UI. this can make certain discussions a lot shorter, by pointing at the list. also I feel that _part_ of the GIMP has so far to go feedback that the 2.8 release gets is caused by us not communicating yes, we have problems.

- coordinating between open technical project work (GEGL migration, etc.) and interaction design project work.

- the big picture concerning UI becomes clearer because it is written down.

so far so good. but while talking to Bruce I realised that there are a couple of side effects to this list that make me hesitate to just go and get started with compiling it:

- just thinking of what I can contribute to this list, I know that this list is _not_ going to be short. also because all the issues that I can put on it are medium level to big-picture, none of them are going to be trivial to solve. thus my hesitation is what this is going to make GIMP look like, and if the definitions of open worked where meant to stretch this far.

- all of the issues that will end up on the list have been created by contributors to GIMP. some of these issues have been created 10 years ago, some of them last month. I wonder what it will feel like to GIMP contributors when something they just made, almost immediately and automatically, (at least, feels like that to them) ends up on the UI issue list.

so I would like to hear some opinion on this.

--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
Nicolas Robidoux
2012-05-17 14:30:42 UTC (over 2 years ago)

hesitant about compiling a list...

How about you start your list with issues that are at least 18 months old?

gfxuser
2012-05-17 14:55:27 UTC (over 2 years ago)

hesitant about compiling a list...

Hi Peter,

my opinion is 'the clearer the better'. Such a list could IMHO improve GIMPs product quality. Even if it will be a long list (and it will be growing, as user requirements and wishes always become more) this will show the necessity to priorize.
I'm not (or not yet) a GIMP developer, but speaking as a software developer I know of the benefits of priorization. Of course, this priorization must be done together with the developers. The GIMP team could then define mandatorily, what will become realized at all or in the next version - which could make it easier to define release plans. The product vision is the big picture, but it needs refinement and this list would fill this hole.
I think, contributors don't have much problems at all seeing their proposals/contributions at this list as long as they know it's part of the development process and they don't _end up_ on this list. If some contributors come with a ready-to-integrate idea (like Tito, a new brush engine etc.), such a list could/should be taken to examine, how these appreciated contributions can become a useful part of GIMP and keep track of that progress.

So, you have my +1.

Best regards,

grafxuser

Øyvind Kolås
2012-05-19 12:32:21 UTC (over 2 years ago)

hesitant about compiling a list...

On Thu, May 17, 2012 at 4:19 PM, peter sikking wrote:

within a minute we were talking about whether GIMP has a list of known UI design issues. as far as I know we do not have one, it is certainly not part of gui.gimp.org. - just thinking of what I can contribute to this list, I know that this list is _not_ going to be short. also because all the issues that I can put on it are medium level to big-picture, none of them are going to be trivial to solve. thus my hesitation is what this is going to make GIMP look like, and if the definitions of open worked where meant to stretch this far.

- all of the issues that will end up on the list have been created by contributors to GIMP. some of these issues have been created 10 years ago, some of them last month. I wonder what it will feel like to GIMP contributors when something they just made, almost immediately and automatically, (at least, feels like that to them) ends up on the UI issue list.

This is what doing work in the open is about, be open about what one is doing; known shortcomings as well as calling out for help. Maintaining such a list also can improve the perception users have; in that in reading such a list they get a feeling for what direction GIMP as a project wants to move in; working on such a list in the open is also something that can make developers feel larger ownership of the projects direction. You might even be surprised about what existing GIMP developers might volunteer as problems with things they themselves have made ;)

/

The future is already here. It's just not very evenly distributed
                        -- William Gibson
http://pippin.gimp.org/              http://ffii.org/
_______________________________________________
gimp-developer-list mailing list
gimp-developer-list@gnome.org
http://mail.gnome.org/mailman/listinfo/gimp-developer-list
Jon Nordby
2012-05-19 13:59:40 UTC (over 2 years ago)

hesitant about compiling a list...

On 17 May 2012 16:19, peter sikking wrote:

a couple of days ago Bruce appeared on irc and we had a chat. (Bruce is now subscribed to this list)

within a minute we were talking about whether GIMP has a list of known UI design issues. as far as I know we do not have one, it is certainly not part of gui.gimp.org.

- just thinking of what I can contribute to this list, I know that this list is _not_ going to be short. also because all the issues that I can put on it are ‘medium level to big-picture,’ none of them are going to be trivial to solve. thus my hesitation is what this is going to make GIMP look like, and if the definitions of open worked where meant to stretch this far.

- all of the issues that will end up on the list have been created by contributors to GIMP. some of these issues have been created 10 years ago, some of them last month. I wonder what it will feel like to GIMP contributors when something they just made, almost ‘immediately and automatically,’ (at least, feels like that to them) ends up on the UI issue list.

I support creating such a list. If we hope to solve these issues, and keeping the amount of similar issues we introduce down, they need to be visible for all to see. Being open is about transparency and accountability. And it is just as important, or more, that we act in such a way with respect to our problems.

Bruce
2012-06-05 09:41:41 UTC (over 2 years ago)

hesitant about compiling a list...

Hi guys, I'm the Bruce that Peter referred to in the first post in this thread.

I thought I'd share some feedback about this, but in more of a brainstorm sort of way where we can just share and discuss ideas for what such a list might look.

That way we can put the idea of creating it aside for the moment and better evaluate the idea and how it could be done in a way that is appropriate.

So far the core issues we're discussing seem to be:

(1) how open does it make sense to be? (2) can we do this in a way that is sensitive to the efforts of GIMP developers, and if so, how?

* * *

*How open does it make sense to be?*

On Fri, May 18, 2012 at 12:19 AM, peter sikking said:

- just thinking of what I can contribute to this list, I know

that this list is _not_ going to be short. also because all the issues that I can put on it are ‘medium level to big-picture,’ none of them are going to be trivial to solve. thus my hesitation is what this is going to make GIMP look like, and if the definitions of open worked where meant to stretch this far.

I think a good approach is to be open to the degree where it is constructive and feels good, not just for the sake of it.

E.g. If you have personal notes for how things can be improved, you don't have to share those notes--they're for your personal reference.

I have an idea for how we can have "public flags in the ground" in terms of what can be improved (which are helpful for reference purposes, and also for communicating "this is something we feel we could improve"), while still allowing people to keep private, personal notes and ideas that they can dip into and reference as it makes sense to (rather than just sharing them all publicly).

I'll make another email after this one and you'll see a better idea of what I mean.

As for the other topic of discussion…

*Can we do this in a way that is sensitive to the efforts of GIMP developers, and if so, how?*

On Fri, May 18, 2012 at 12:19 AM, peter sikking said:

- all of the issues that will end up on the list have been

created by contributors to GIMP. some of these issues have been created 10 years ago, some of them last month. I wonder what it will feel like to GIMP contributors when something they just made, almost ‘immediately and automatically,’ (at least, feels like that to them) ends up on the UI issue list.

Well, I think one way to approach it is how things are communicated. E.g. You can communicate things in a constructive and positive way (i.e. "here are some cool directions we would like to go in in regards to the various UI elements of Gimp"; sort of like the UI Brainstorm does), or you can communicate things in a remedial way with a focus on deficit (i.e. "this is broken and needs to be fixed").

I feel the positive approach is more constructive and also more productive (and it also feels different). We want to inspire the GIMP community and developers and help them be clearer about the direction the user interface and user experience of GIMP is going.

In that sense, a list of current issues (which, by focusing on "issues," focuses on deficit) may not be the best way to approach this. That was what we (Peter and I) talked about at first, but let's iterate on that.

Perhaps a better way to approach this is to have such a list list be more of a vision of what Gimp can be and a place where ideas for how things can be done can be documented for reference. E.g. There's a difference between a bullet point that says "X icon isn't very good" and "Several people have proposed ideas for how X icon can be improved; here are links to mock up images that people created." Again, one feels different to the other and has more of a collaborative, inclusive spirit that's good for the community.

*So, here's my idea:*

As you already do, you could allow people to submit ideas to the UI brainstorm in the form of mock-up images. This keeps things solution-oriented. The idea would be to have all UI suggestions and ideas go through the brainstorm in the form of images that propose solutions.

Then, you could create a page at http://gui.gimp.org called *"List of ideas and proposed changes for the GIMP user interface" ** *(or something like that) and have:

- a sub-section that describes (or links to) the overall vision of what you'd like to do with the Gimp UI in a general sense (which helps people to be on the same page)
- sub-sections that act as buckets for documenting submissions made to the Gimp Brainstorm so you have a nicely categorised overview of the things people would like to see in Gimp in terms of the UI, and also a pool of solutions developers can draw on.

The idea behind that page would be two fold:

1. developers can go to it to get inspired and also see what can be worked on (in a UI sense) and basically pluck items from there and start a thread over in the UI contribution process for more in-depth focus on hashing out a solution that can iterated on and eventually be added to Gimp [assuming the UI contribution process goes ahead] 2. people who are using Gimp (i.e. users) can look at the *"List of ideas and proposed changes for the GIMP user interface" *page and get an idea of ideas and solutions that have been proposed, and if they have an idea for how the Gimp UI can be improved that isn't listed on that page, they can submit an image via the UI Brainstorm.

This way things will always be constructive.

And you guys can still do your regular UI Brainstorm review--the idea is just to document what is submitted to the UI Brainstorm as a way of communicating what has already been considered, but in a categorised list format that is easier to browse through (for people looking to see what the state of things is and whether an issue has been addressed) and also easier to pluck ideas from (for UI developers looking to start a UI contribution process thread based on what they've seen on the *"List of ideas and proposed changes for the GIMP user interface"** page. * *
*
People could still discuss UI topics in the developer mailing list, but "communicating ideas through images" via the UI Brainstorm would be encouraged. Image submissions don't have to be perfect; just "sketches" of what could be that:

- communicate to the developers what the community would like to see in Gimp
- give the developers new ideas
- and help flesh out the *"List of ideas and proposed changes for the GIMP user interface"* page as a community and development resource.

** * **

Those are the best ideas I have so far in terms of how to make the process more constructive, productive, and hopefully a nicer experience for the developers.

I think it's still fine to document raw notes (like those found here: http://gui.gimp.org/index.php/Evaluation_Notes_-_Photo_Realistic#introduction ),
(A) that can be done elsewhere, and (B) notes, by definition, are unrefined reference resources that have not been processed. Notes may contain good ideas, observations, and solutions, but those things--the "gold"--have yet to be mined out, processed, and then displayed in a format that is more useful and appropriate for consumption by the Gimp community and developers. Curation (what and how things are said) matters.

Again, the idea behind the *"List of ideas and proposed changes for the GIMP user interface"* page isn't to spotlight and highlight what's wrong, but to help build bridges from where Gimp currently is to where it can go. Then, when someone wants to help construct one of those bridges, they can pluck out the notes on the *"List of ideas and proposed changes for the GIMP user interface"* page and create a new thread in the GIMP UI contribution process (if that goes ahead) and the actual development process (for that particular item) can begin.

It's still possible that a new release of Gimp might come out and then shortly after people post images to the UI Brainstorm with ideas for how a feature can be improved, but when those images are reviewed and added to the *"List of ideas and proposed changes for the GIMP user interface", *when developers reference that page they'll see constructive ideas and visions of what could be rather than "X icon isn't good," or "this feature needs to be better designed." [Which aren't necessarily bad statements—they are what they are—they're just not that helpful when it comes to what we're trying to do.]

On Sat, May 19, 2012 at 11:59 PM, Jon Nordby wrote:

On 17 May 2012 16:19, peter sikking wrote:

a couple of days ago Bruce appeared on irc and we had a chat. (Bruce is now subscribed to this list)

within a minute we were talking about whether GIMP has a list of known UI design issues. as far as I know we do not have one, it is certainly not part of gui.gimp.org.

- just thinking of what I can contribute to this list, I know that this list is _not_ going to be short. also because all the issues that I can put on it are ‘medium level to big-picture,’ none of them are going to be trivial to solve. thus my hesitation is what this is going to make GIMP look like, and if the definitions of open worked where meant to stretch this far.

- all of the issues that will end up on the list have been created by contributors to GIMP. some of these issues have been created 10 years ago, some of them last month. I wonder what it will feel like to GIMP contributors when something they just made, almost ‘immediately and automatically,’ (at least, feels like that to them) ends up on the UI issue list.

I support creating such a list. If we hope to solve these issues, and keeping the amount of similar issues we introduce down, they need to be visible for all to see. Being open is about transparency and accountability. And it is just as important, or more, that we act in such a way with respect to our problems.

-- Jon Nordby - www.jonnor.com
_______________________________________________ gimp-developer-list mailing list
gimp-developer-list@gnome.org
http://mail.gnome.org/mailman/listinfo/gimp-developer-list

peter sikking
2012-06-05 12:21:23 UTC (over 2 years ago)

hesitant about compiling a list...

Bruce wrote:

[I will snip here quite a bit to keep this post from ballooning]

Hi guys, I'm the Bruce that Peter referred to in the first post in this thread.

How open does it make sense to be?

I think a good approach is to be open to the degree where it is constructive and feels good, not just for the sake of it. I have an idea for how we can have "public flags in the ground" in terms of what can be improved (which are helpful for reference purposes, and also for communicating "this is something we feel we could improve"),

I have ben thinking for the last weeks that this list should not be called the issue list, but the to-do list. this matches how the project is run. the UI team is simply working its way through areas of GIMP, older ones and brand new ones. when we get there, we design it, and make it work.

Can we do this in a way that is sensitive to the efforts of GIMP developers, and if so, how?

Well, I think one way to approach it is how things are communicated. E.g. You can communicate things in a constructive and positive way (i.e. "here are some cool directions we would like to go in in regards to the various UI elements of Gimp"; sort of like the UI Brainstorm does), or you can communicate things in a remedial way with a focus on deficit (i.e. "this is broken and needs to be fixed").

well, design is not about cool, it is about identifying the problem and then solving it (the hard design part). talking about design in such terms is for me important; it is about respect for what interaction design is and the value it brings to software.

thus it needs to be said that this is broken. but lets introduce the convention that it can only written down that this is broken, when the next sentence discusses the (beginning of a) solution. this means there is always light at the end of the tunnel, that the UI team is able to put direction into the UI work, and that any contributor has the chance to start from this direction.

So, here's my idea:

As you already do, you could allow people to submit ideas to the UI brainstorm in the form of mock-up images. This keeps things solution-oriented. The idea would be to have all UI suggestions and ideas go through the brainstorm in the form of images that propose solutions.

there really is an ocean sized gap between the anything-goes world of UI brainstorming and the rigour of interaction design. anyone can participate in the brainstorming, that is the purpose of it. the only way to bridge the gap from brainstorming idea to interaction that is designed and can be implemented is to _design_ it.

the whole discussion about opening up this design process to more contributors is in the contribution processes thread.

Then, you could create a page at http://gui.gimp.org

gui.gimp.org is a repository, of interaction designs, usability projects and documentation of interaction design projects and processes. similar to a code repository, the integrity of gui.gimp.org must be maintained or else it is worth nothing.

this means there is no place for ideas on gui.gimp.org, just as there is no place for casual talk about code in a code repository.

the to-do list fits gui.gimp.org, it contains already quite a bit of contribution by interaction designers: the evaluation and analysis where the real issue is, and by showing the light at the end of the tunnel.

--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
https://mail.gnome.org/mailman/listinfo/gimp-developer-list