Robert Sasu wrote:

I would like to participate to GSoC 2012 and I found the Implement the combined selection tool project interesting.

I've read about the research made by Voralberg students in 2010, as suggested, and I am really interested on making a new and nice UI for some tools.

I suggest we do the same as for the 2011 student work:

the work was taken further by my design team in a 3-day design sprint:

in that short time and short spec, the main problems are solved, some very advanced design goals are met, the general design of the tool is done and as a solution it has gelled.

with that there are no more surprises what direction the technical design and implementation should take.

about the 2010 student work: yes, the integration of same-ish selection tools was exactly my brief. the difference between my class and the design sprint is that in the former there is lot of freedom, and in the sprint there are these advanced design goals. this is because GIMP is not a toy, it is serious software production. also it would be to overwhelming to ask the students to meet the advanced goals.

I do not expect things like gestures or pie menus to make it through the design sprint, simply because of higher standards.

Last, one bit of risk in this project is the future of the SIOX tool. I heard rumours about a replacement. there needs to be a decision by the GIMP team about that before the design sprint starts.

--ps

founder + principal interaction architect man + machine interface works

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

On Mon, Mar 19, 2012 at 2:22 PM, peter sikking wrote:

Last, one bit of risk in this project is the future of the SIOX tool. I heard rumours about a replacement. there needs to be a decision by the GIMP team about that before the design sprint starts.

Rumours?

Well, our current SIOX tools sucks as it doesn't make it possible to refine the selection. The GSoC project on that was successful, but the code doesn't compile and noone tried to work on that really.

An alternative tool that works really nicely was created by Jan Regg [1], but we couldn't merge it, because a) it was too late for 2.8 anyway, and b) we do not accept new tools that are based on the old core. So Jan implemented a GEGL operation for matting [2] which would be used by a new GEGL based selection tool, but not the actual tool.

The way I see it is that we have few issues there:

1. SIOX tool as such could be obsoleted by a new combined selection tool 2. We have concurrent matting operations, one from GSoC and one from Jan.

[1] http://youtu.be/kkQ1r5g49d4 [2] https://github.com/rggjan/gegl-global-matting

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 Tue, Mar 20, 2012 at 11:27 PM, Robert Sasu wrote:

Futhermore Alexandrine issued to get rid of SIOX tool and to choose between Jan's project and the GSoC project on SIOX tool.

1) Alexandrine is a girl's name. I do genuinely believe that I'm a boy. I'd even push my luck and go as far as "man" :) 2) I don't think I ever suggested to choose between Jan's project and the semi-successful GSoC SIOX project. None of them can be accepted in their present form.

How important is this project for this year's GSoC?

Well, we are aiming to replace as much of the legacy code with GEGL based code as possible. If we can make better tools while at that, I'd hazard a guess that this project would be important.

Are there tasks for me to do in this period, what should I learn, is there some research to be done ?

ps. I saw the list of the needed GEGL ops on the Ideas page. Last year I have implemented coloar-to-alpha and rotate-colors ops in GEGL.

I dare say you implemented far more operations :)

Alexandre Prokoudine http://libregraphicsworld.org

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

Robert,

I suggest we do the same as for the 2011 student work:

the work was taken further by my design team in a 3-day design sprint:

The 2010's Voralberg's research is about combining four selection tools (fuzzy select, select by color, intelligent scrissors and foreground selection). The suggestion from Peter Sikking was to do as the 2011 student's have done. This research is about iWarp tool. I've read also this research, and also studied the work of the design team.

No I suggested that the same procedure is followed, not the same project content.

That means: if the SoC work goes forward on the combined four selection tools, my team does a design sprint, taking the design of the combined four selection tools forward. this would be a solid basis to for you to work from.

--ps

founder + principal interaction architect man + machine interface works

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

Hi all, I like to implement the GIMP "combined selection tool" as my GSoC project. I'm a 3rd year student in department of computer science and engineering, University of Moratuwa, Sri Lanka. I'm good in C and familiar with GTK+. currently I'm studying GIMP source code. I like to fix some bugs as I will understand the source code properly. Can someone guide me to find some bugs to fix?

Thanks all.

Regards,

*Malaka Gallage Department of CSE
University of Moratuwa*
[image: Facebook] [image:
Twitter] [image:
LinkedIn]
[image:
Blogger]
Contact me: [image: Google Talk] mpgallage@gmail.com [image: Skype]malaka.gallage