Well, last day of Tokamak. It’s sad that I must say goodbye to all of these people that actually lived with me for one week. It was so intense that I almost didn’t realize that we were together for one week. But let’s talk about what we did during the sprint.
During the first days I talked with everybody I could about anchor layouts, animated layouts and Qt stuff. It was great as it’s hard to explain the stuff and get feedback from a lot of great hackers at once and as fast as we can get when we are in this kind of developer sprints. We had great discussions about Qt, regarding bugs, features, future and everything else we could talk about. I really can see that every time that we put all these great hackers in the same room, we save some months (maybe years?) of development. All of that to make you take less time to code
While everybody else was merging their brand new stuff on trunk, commiting bug fixes and new features I was fighting with cross compiling KDE to the “mysterious device” that Aaron brought to us. What I could achieve ? We have KDE 4.2.X working PERFECTLY on this device, with all the features and beauty that KDE could bring to you. It runs really really fast on this ARM device. We were all impressed!
And then came the second part of the experiment: run trunk on it. While it was ok to compile stuff on scratchbox using Maemo SDK, the lack of a proper SDK for the device gave us some headache: some bug on qmake prevented it to detect correct paths and made me find out each library that was not properly added to cmake’s links.txt file and correctly add it. By the end of kdebase compilation I already had almost all the names and did a script to fix that, but until the time I found out all the libraries it was compiling and stopping all the time (I don’t need to tell that it was really slow to compile and due to this problem, it was even slower).
After some days I finally had kdelibs + plasma-destkop and plasma-netbook compiled. However when we put all this stuff on the hardware we faced a problem with Svg files not being show. So althoughf we have plenty new stuff full of new features running on the device, it looks a little bit “ugly” without the proper background, etc.. we just have white background…white as snow. But come on…we just had a few days with the device…we can do MUCH more with it and imagine if we had a proper SDK ? We can make this hardware AWESOME with our software and I can only hope that we’ll have some more time with it in the next months…(we would need less than 1 month with a proper SDK to make it really really really ROCK!).
Ah, meanwhile it was compiling kdelibs I created the pastebin dataengine/service, migrated the pastebin applet to use that and also fixed some bugs on other applets…
Talking about snow, today was the first day I saw snow in my life It was a unique experience for me :). It’s really hard to describe with words how happy I was and how beautiful it was in the top of that mountain. Thanks Mario for this trip…it was one of those moments in our lifes that you never forget.
Tomorrow morning I’m heading to Brazil again. It was an awesome sprint. If someone asks me what comes next from my point of view, I would answer: running plasma-netbook on the “mysterious device” (please, I just need one week with a proper SDK ), finishing and optimizing anchor layout for Qt 4.6 (I really hope this will help people out there to create rich UIs with Qt) and of course: keep all the stuff we have discussed during Tokamak going on!
Thank you Mario, thank you Plasma developers, thank you Qt/INdT and thank you KDE What a wonderful personal and professional experience.