On 04/05/06, Timothée Anglade <timothee.anglade@...1261...> wrote:
Okay guys, seems I was really off the target on that idea.
Of course you are not off target.
If you want a top quality drawing application for the Mac OS, note that Coreldraw discontinued theirs a few years ago, and Freehand is set to disappear.
Canvas 8 is available for free, exempli gratia: http://www.paperworlds.com/forums/lofiversion/index.php/t853.html (and a few other places - enquire at your local Macintosh User Group)
The Creature House Expression beta still works: http://wiki.inkscape.org/wiki/index.php/Expression
It is perfectly reasonable to fill the gap left by Freehand with an existing open source application such as Inkscape, to start a new project such as Chlor http://sourceforge.net/projects/chlor/ , xvg http://sourceforge.net/projects/xvg , or PettySVG http://toxicsoftware.com/blog/index.php/weblog/entry/pettysvg/ , or to Open Source a previously closed product such as XaraLX http://www.xaraxtreme.org/developers/documentation/xara_lx_mac_build_instruc... .
There are four or five people who commonly post here on Mac topics, and as I suspect that there are around ten lurkers for every poster, we probably have 25 to 50 Mac devs and testers, and this could be the entire world supply of available talent for such an endeavour. I would guess that a Native Mac Port is seen as important, but not as critical.
You may have already picked up on this suggestion, but if you feel that the Wiki is missing a particular page then you can make yourself a user name and add that page. If it is a good idea, then people will add to it, perhaps in ways that you had not considered.
Speaking personally, I think that there are several things 'missing' from the roadmap, but I am glad to see that Animation is now on at 0.51 and Bug hunt 0.47 (which still leaves 'Rework Layers UI' and 'Blends', for example, as omissions). I hardly dare look at the RFE tracker, I suspect that there are hundreds of valid requests, even after removing duplicates, and despite the natural worry I have of Mongolian Horde http://www.everything2.com/index.pl?node_id=29845 programming I expect that Inkscape, being still a young project, needs to acquire many more devs whose main interest is adding the one feature that interests them, it it would not surprise me if, over the next few years of its life, Inkscape grows to support several hundred devs - something which ought to require some thought and perhaps planning before the size doubles for next time.
Note that your precise comment about the Mac's being used by creatives but not having sound open source applications is made at every mention of Pixen http://www.opensword.org/Pixen/ and Seashore http://sourceforge.net/projects/seashore , I would guess that there are plenty of applications for "creatives" perhaps some boutique or niche, but good enough to fit in for the most part, and stand out where appropriate. As I say, check with your local Macintosh User Group.
Unless otherwise notified, I see no bar to adding milestones to the Map, so 'Groundwork for Native Mac UI' would fit with 'Drawing Feature Enhancements' at 0.48, but I suspect that you will need to add a 0.56 for 'Commit Native Mac Port'.
When you speak of X11, this is very much a three edged sword. X11 support on MacOS X/Darwin is really very good if you want X11; Developments in GTK+ may solve some of your concerns; and being able to use SVG as a platform independent file format is sufficiently refreshing as to counteract the obvious infelicity.
Now, I appreciate that a substantial portion of the Mac community do not want to use X11, and may not agree that this is a good bargain for getting SVG support, but neither point of view trumps the other. I take my hat off to the people at Cupertino who have given us this choice. Further, speak to the people at your local MUG, you will find free as in beer editors (I think that Canvas is best of breed), but people are still getting excellent results from Appleworks and even MacDraw Pro.
You need to look at the archives of the list to see ther extent to which people have discussed the architecture of Inkscape. In practice the 'Engine' and UI are not totally bonded together. I fear that the 'Engine' does use services from GTK+, but this should be resolvable through re-factoring. (I suggest that you allow part of the calendar time of 4 milestones to do this).
You will probably find that most of the frequent committers have given this no thought, the frequent committers will either be scratching their own itch or have other concerns. Maybe it is not that interesting. Now, I have no insight into how the SoC projects are assessed, but if one scale is to weight against ideas which ought to have their implementation under the aegis or auspices of some other organisation then you might find that adding a Cocoa port to Inkscape is affected by that. Certainly I would wonder why Google should should be involved if nobody has stepped up to make a start.
I wouldn't agree that existence of many thousands of lines of GUI code in the current Inkscape projects is good or bad for a Cocoa port. We would write a Cocoa interface (using interface builder for the most part) and plug that into the 'Engine' using an MVC architecture; much as doing a brain transplant means preserving the pryamidal tracts. Do you think that the existence of hundreds of thousands LOC for XUL in Firefox and Mozilla went (or could go) any way towards dissuading the authors of Camino?
Also many of the users of Inkscape are on Windows despite most of the devs using linux: All that is needed is one person with a clear goal and the determination to keep the Cocoa port on track.
You might want to get involved with some of the other projects I have mentioned, perhaps work on GTK+ (I am having difficulty compiling the CUPS backend at the moment), add the documentation that you have adumbrated and perhaps bung the whole of the Inkscape code base into XCode (XCode has support for Test Driven Development, and I am sure that any work you could do on this for the Inkscape project in general would stand you in very good stead when you ask for help or support) and get it to compile. (Had you done that last year, you might have had to start over after the switch to Subversion).
I general, I would say go. I would guess that it is inescapable that you will have to do some planning, but from the outside it is as simple as:
* Work out what you are starting from.
* Work out where you are going to.
* Write down some ideas
* Start developing
I am not sure whether you should work on your own. Assuming that you have the necessary XCode and Interface Builder skills then that sort of purity is helpful. For aught I (or you) know, there could be people trying out these ideas as we speak, and it just needs someone to co-ordinate whatever endeavours are already under way; something that Subversion should make easier.
Ben