Category archives: Snaking
Discussions of programming, particularly of programming Python
Eating, napping and graphics (Good finish to the day I suppose...)
Written by
on
in
Snaking.
Cooked up some curry this evening. Wasn't that spectacular, mostly because it burned a little in the wok. Rosey is still sick, so she went to bed before trying it.
After eating I was just so exhausted I took a nap. Got up around 1am and decided to fix a bug that's been annoying me ...!-->!-->
Wow, where did *that* day go? (Gotta love meetings...)
Written by
on
in
Snaking.
Good day, all-in-all. Much of it spent on documentation. A bit on making the demo code a little less CPU-greedy (though not quite there yet), and the meeting with Steven.
Also flipped bits on 10 or so "billable" items... almost a month's work... I didn't count, but it's probably 200 or more hours. If we ...!-->!-->
Mmm... sales (Well, sales documentation...)
Written by
on
in
Snaking.
Wound up taking almost 3/4 of an hour explaining a new (heuristic) feature (that triggered a little too readily last night) to Bryan in email (and coming up with a way to improve it as I did), so not as much of the "what's in the box" document as I'd planned, but I did get ...
OpenGLContext grows more Interpolators (Morphing and moving and colouring and fading! Oh my!)
Written by
on
in
Snaking.
Spent an hour or two on OpenGLContext last night. In particular, generalised the OrientationInterpolator code into a generic Interpolator and sub-classes for position, colour, scalar, orientation, and coordinate interpolators. Only VRML97 interpolator node still missing is NormalInterpolator, which is both very rare and a bit of a pain to create (because it requires a quaternion ...
Day dissolves (Fragmented work-day doesn't produce much...)
Written by
on
in
Snaking.
I know I did work today, but given that length of time spent, it doesn't seem like much. Wound up having interruptions (D'Arcy's call, Tim's visit and the management meeting) that kind of chopped the day into little fragments (I don't work well when there's someone in the room, I'm just driven to talk to ...
A nap and then OpenGLContext (TimeSensor and OrientationInterpolator implemented...)
Written by
on
in
Snaking.
Had to sleep for 3 hours or so to give my hands a chance to rest, but then I sat down to work on OpenGLContext. Decided that the most important issue was the lack of easily accessible animated content, so set about fixing that.
The TimeSensor node is basically just a wrapper around the Timer. ...!-->!-->
Making the emailing work (Limiting and balancing and hierarchically-ordering... oh my!)
Written by
on
in
Snaking.
Today was all about tweaking email and trying to get it to the point where alerts are not flooding the users, but are also not delayed so much for consolidation that they're pointlessly late (getting a message that says, in effect, "your CMTS crashed 30 minutes ago" isn't much help).
Along the way I burned ...!-->!-->
Nope, more Cinemoning (No wonder everyone else requires $100,000 servers...)
Written by
on
in
Snaking.
Bryan remarked on slow changes to the online views after the offline event yesterday. It's an artefact of our (delicately balanced) scanning method, which tries to make sure that we're not overloading the processor or network while still trying to quickly report changes to the status of groups in the network.
Problem is, it doesn't ...!-->!-->
Time to restart the Mikey on Linux project? (Have to get off Microsoft eventually...)
Written by
on
in
Snaking,
Vindaloo.
As all my loyal readers will no doubt remember, I hit a rather annoying snag way back when that forced me to switch back from Linux to Win2K. I fixed that a while ago, but since then I've spent all of my time on Linux trying to get FreeVO installed and configured to work with ...
Quiet day of Cinemoning (Working on demos)
Written by
on
in
Snaking.
Improving demo-ing capabilities and handling CMTS failures and resets was today's focus. Nothing spectacular, just hooking up code in MockCMTS such that it can be accessed when the application is running in demo mode and then providing a UI for it.
The demo ui stuff showed a number of problems in how the MockCMTS works ...!-->!-->