Hmm. Qt’s kinetic is pretty cool. Of course, Amigas could easily do something very similar back in the 80′s (with BOBs), but it’s a big step forward for more “modern” systems
This is going to be a slightly cheesed off post. Better skip it if you’re not thick-skinned. Back to happier django posts soon… if I can forgive this. So I’m deploying my first real django app to a live server, with modwsgi. ModWSGI itself isn’t great, as it insists on a single python version per [...]
Continue reading about Django Development: Runserver Ideal vs. Deployment Reality
So I upgraded my desktop to 6GB recently (from 2GB), mainly for virtual machines, but also just because I hate virtual memory. That necessitated switching to full-time use of an AMD64 edition. Up to then, I’d been jumping back and forth trying different combinations of 64-bit OS and 64-/32-bit apps, given issues with drivers, software [...]
Aaron Toponce posted some interesting “libraries of congress” analogies about the size of the IPv6 address space. I loved how he said that “18,446,744,073,709,551,616…may not look large” compared to the (less than) 2^32 addresses available in IPv4 — he seems to think about future compatibility in the same way I do Anyway, I did find [...]
It seems that Python 3.1 alpha 1 has been released already. A quick scan of the NEWS file reveals some nice performance improvements in there: IO stuff reimplemented in C new garbage collector faster Unicode handling computed gotos (which I’ve never heard of, but apparently offers 20% speedup on supported compilers) Things like this keep [...]
I seem to be in a love/hate relationship with Git. On the one hand, it’s… well, horrible, usability wise, with all it’s unexpected behaviors and weird terminology for well known concepts. On the other hand, it does nice things like this when changes are committed: … rename noodil.py => kit_library/hello.py (71%) mode change 100755 => [...]
Hmm. Chrome-center has an article up performance of VIA Chrome chips on windows and Linux for Nexuiz, with pretty charts showing that Nexuiz runs faster on Linux. I think this is way off. For one thing, the Xorg and video driver guys are busy re-writing the entire Linux video stack because it’s entirely out of [...]
Jeffrey Way posted a screencast about fast coding using editors with text expansion. The basic premise is that repeating long boilerplate is time-consuming, and that it’s more efficient to use an editor with text expansion. There’s a little bit of merit in this under certain circumstances, and Jeffrey’s example was possibly contrived. However, I find it quite a wrong-headed approach to development under most circumstances. As such, I’m going to bemoan it a bit, as it’s a pet hate of mine at the moment. In the process, I’ll also attempt to be constructive, and sketch out an altogether nicer alternative.
Continue reading about Avoiding mundane repetition: Single-Sourcing, vs. Text Expansion
Initially, I was skeptical about the design, but it’s actually a pretty nice little tool. When someone wants to borrow a browser, being able to just hit guest, let them play on their own desktop, log them out, and return to what you were doing, is very handy.
So I upgraded to ext4 with jaunty on my desktop lately. I needed the 2.6.28 kernel to get my tv card (hvr-3000) working, and since ext4dev had been retagged as ext4, it seemed a good time. All in all, ext4 has been nice so far, but over about three weeks of use, I’ve gotten “no [...]
