Anyhow, playing on a personal project (calendar and project management system for GTK on BSD) I realized one of my bad habits might have applicability to a wider audience. During the development stage, a required file in my world is called "developer.notes". The point is pretty straight forward, it's a repository of my thoughts, rants, comments, etc about the code I'm working on. I never allow it to leave the project (i.e. no one who hasn't worked on the software will see my notes). It captures almost everything, since it's convenient, it's a great way to think with my keyboard and I know it won't ship as part of the final package. It is, however, exceptionally useful as a reminder, developer documentation, whatever.
Some sample paragraph openings (I see no need to share the entire note set):
Thoughts for evolution...apparently MySQL can dump as XML, so...
Look! I documented the data structures.
Next up, start building even more happy little dialog boxen.
This is horrifying, I'm not even half way to alpha and...
You get the idea, I hope. If you find this a useful idea, please use it and share. If you've already been doing this, well done. Otherwise, have a great week-end.
After the monthly operations review, I've actually started serious thinking about what the required elements of a support model would need to be for those products that have an Open Source software base. From a corporate perspective, this is actually harder than it seems, but an interesting problem. Anyone who's managed a large, geographically diverse support group probably has some idea of what I'm trying to do.
An interesting piece of fallout from my efforts to deal with supporting Open Source software in proprietary hardware scattered all over the world is an examination of the nature of supportability and communications system resiliency. If there's anything worth discussing here, I'll put together an article and post it. If not, then I probably won't. We'll see how I feel and what happens.
On the distributed information store issue, only a small amount accomplished. It seems to me, that if an individual server is aware of all servers in it's "area" and knows of one server in each other possible are, I can solve part of the problem. Each server would behave as a "DNS Root server" within the area it knows, and as a "caching only DNS" that does recursive queries for the remaining areas. Sort of :). There are a large number of issues I have to come to terms with, adding new servers to an area, tracking new nodes on the information store, reducing the network load for searches, and a host of related things. Fortunately, this starts as a closed, one-server only system. The distributed bit happens when it becomes an open-source project. There's also some thought needed for types and levels of extraneous functionality.
Other than that, if we can get the air-condioner going around here, should be a good week.
I've had some thoughts on this and I'm starting to get some feel for where this will need to go. There are, fortunately, some simplifying assumptions I can make, because the raw problem is probably not solvable.
Now I gotta run, the network guy is here to move my subnet. To be continued...
Anyhow, on the hacking front, been busy in too many areas. You ever notice I don't talk to much about what I'm actually hacking on? That would mostly be a result of the fact that I don't intend to release most of what I hack on. Some is for work (internal only), some is just for play and a lot is just to scary to contemplate putting out in the world.
Work is interesting these days, but for all the wrong reasons. Apparently someone noticed that managing a group in Ottawa from California probably needed adjustment so we're promoting someone from within the group to be the new 'local' manager. Might even find out who it is today. Four of us are at least marginally interested in the position, although I suspect any of us could do the job. The only real drawback is the hit to technical skills. If you don't use it, you lose it. As a manager, the time spent hacking drops dramatically.
I have no idea if this means I'm going to be writing here regularly again for a while or not. We'll just have to wait and see what happens ;)
Nothing noteworthy in the hacking dept. A little work on Intraverse and CS, cleaning up the machines at home and work, that kind of mindless silliness.
FOAF updates: Trust rankings are now exported, making the data available to other users and websites. An external FOAF URI has been added, allowing users to link to an additional FOAF file.
Keep up with the latest Advogato features by reading the Advogato status blog.
If you're a C programmer with some spare time, take a look at the mod_virgule project page and help us with one of the tasks on the ToDo list!