It’s now almost one year since I started my job at BlueSystem to work full time on KDE software. It’s a perfect time to look back and do some retrospection as well as trying to look into the future.
Of course my focus of work was on KWin, but especially over the last months I worked all over the KDE workspaces. Overall we have achieved a great result. KWin is running in near production-ready quality after the port to Qt 5 and the Plasma workspaces are also in a very good shape already. I tend to mentially compare the experience with the state of Plasma 1 six years ago which was just a few weeks before the first release. We are in a better state and there is still lots of time till we will do our Plasma 2 release.
Without the possibility to work full time on KWin we would not yet have a ready port and I fear the quality would have to suffer. There were many issues which just needed lots of time to break through. Sometimes I had the feeling that I run all day long with the head against a wall just to end up in front of the next wall once I broke through. Luckily such problems are behind us and I’m also kind of proud to have solved those problems.
But not everything this year had been fun. There was one event which seriously affected my work in a negative way. It demotivated me, I needed to spend time on it, it made me spent time on useless issues instead of the important porting. I was attacked and insulted for my opinion – an experience I had never made before and I don’t want to make. Of course it affected me personally and I can see the impact of it. In retrospection it is clear that I nearly got burnt by it and that it affected the way how I worked in the free software ecosystem. I notice that I don’t have any interest in collaboration with other projects any more, that I feel attacked whenever someone disagrees with me on a technical level, I am answering questions in a short and blunt way with tendency to arrogance. I try to avoid discussions, stopped caring for bug reports and are lacking on code reviews. To a certain degree I tried to be on my small island where I can just hack and ignore everything else.
I am wondering how I would have taken the whole situation if I were just a spare time developer. Maybe it would not have affected me at all because the pure thought that I am supposed to add support in my spare time would have been even more ridiculous as it is. Maybe I would not have had the time to do an analysis for the Kubuntu team as I did which then would not have resulted in me getting attacked for it. On the other hand if the attacks and insults would have also happened if I were a spare time developer I don’t think I would still be a KDE developer. For a hobby I would not have endured that.
One of the conclusions is that if it would ever happen that someone submits a patch for Mir support in KWin, I would immediately step down as maintainer. After what happened I am no longer in a position to judge that imperial and objective. I would either accept the patch out of appeasement or reject it due to what has happened. In any way I would not be in a position where I can be sure to do the right decision for the KDE community. Even on a technical level I would not be sure whether I would be able to review it. Being nitpicky about small issues can be done subconsciously and also moving around code just to prove the point is easily done.
Please do not read this as a threat to block any patches as there’s more into it. I do not expect that the area of conflict will go away. I expect that the issue will come up again and again and I expect that those lobbying us will continue to do so. The next time will probably be the preparation for 14.10 where I’m quite sure that we (and probably also GNOME) will be attacked again for our decision. As I explained above this is affecting me and it’s something I do not want. The logical consequence is that I will have to reduce my involvement in KWin and the kde-workspaces and focus on other areas. After having worked for six years on KWin it might anyway be a good point to look for new challenges, especially given that I decided years ago to re focus once the Wayland port is done.
Of course Wayland will be the big item for 2014. It’s one of our internal aims for the next few months and given that I had focused mostly on porting of all kind of workspace applications I’m looking forward to work on it again. Given that the number of applications ported to Qt 5 is increasing each day it will also be a nice thing to work on with hopefully soon switching to a dog-foodable environment having real-world Wayland enabled applications and not just the Wayland demo apps.