<div dir="ltr"><div><div><div><div><div><div><br></div>I'm guessing most of you have already noticed my absence in the<br></div>last 6-7 months. This neatly corresponds to the birth of my son<br></div>Henning. Because I've had such a great time working on MythTV and<br>
</div>meeting you all I'd like to actually say goodbye rather than just<br>fading away. I've been working on MythTV for almost a decade, at<br></div>first it was as a break from schoolwork but later it developed into<br>
</div><div>much more. I've made friends, some of whom I've happily met in<br>real life, and many more that I'd like to. If I've ever promised you<br></div><div>a pint of beer, be sure that you can take me up on it :)<br>
<br></div><div>So why am I leaving rather than just taking a break? It's really about<br></div><div>a wandering interest. It's been a long time since I was interested in<br></div><div>TV as such, but MythTV kept me engaged with the challenges of<br>
</div><div>writing good C++ code and presenting a stable API in front of<br></div><div>unstable hardware. Even beyond that, I felt part of a community that<br>cares about the software they are producing. But code and community<br>
</div><div>can only hold one so long. I haven't yet figured out what my next<br></div><div>hobby programming project will be, but it will probably center around<br>children or traffic safety.<br><br></div><div>In the tradition of Eisenhower, I'll offer some advice based on my<br>
</div><div>own mistakes. First, big rewrites are always a bad idea. Write up the<br>larger plan on a wiki somewhere but implement the change in<br>master in easily reviewable bits; just because branching is easy<br>doesn't make it a good idea. Second, avoid recurring arguments<br>
</div><div>with newbies by pointing to the previous incarnation of an argument<br>rather than rehashing the argument; attention is a precious commodity<br>in an open source project. Third, do occasionally revisit well made<br>
decisions. For instance, this mailing list would probably be better<br>served by a forum now even though it was entirely the right decision<br>to make it a mailing list when the project was founded. Fourth, don't<br></div>
<div>take anyone's advice too seriously.<br></div><div><br></div><div>--<br></div><div>Daniel Thor Kristjansson<br></div></div>