Entries published in April 2007
A tale of two redesigns
Newsvine launched a redesign, or possibly a “realign”, today, adding a number of useful features while staying true to what has, overall, been an attractive and successful layout. Criticisms in the discussion thread are met with haiku and promises to work on it. Result: pats on the back all round.
Fark launched a total redesign today, and the number of design sins not committed is perhaps the easiest thing to enumerate:
- It doesn’t have any glossy Web-2.0 reflections.
And, well, that’s about it. Criticisms in the discussion thread are met with a ...
Django template utilities
In the course of working on a couple personal projects (which will be released eventually), I found myself factoring out a lot of common code and making it more and more generic, until one day I realized I had a whole separate application just waiting to be documented and bundled.
So here it is:
template_utils, a collection of generic and useful tags, filters and other tools related to Django templates. Full documentation is on the project wiki at Google Code, and distributed with the download.
Even though I’m using most of the ...
Quote of the day
Jeff is on a mission to find view functions and template tags in our software which don’t have useful docstrings (and hence don’t have useful auto-generated documentation in the Django admin). The result (copied from IRC):
<jcroft> congratulations to ubernostrum: weblogs wins the award for the most well-documented Ellington app
<jcroft> which is especially impressive, given that it’s also the one with a fuck lot more views than any other
Jeff diplomatically declined to publicly shame the author of the worst-documented Ellington application…
The types will save us!
So, that went over well. My goal was simply to point out that strong and static typing are not the same thing (and, correspondingly, that weak and dynamic typing are not the same thing), and somehow the comment thread got derailed into people trotting out exotic type systems and theoretical work to try to prove… well, I’m not sure exactly what they want to prove. But one big thing that seems to be causing trouble is this statement:
a common myth of some type systems is “if it passes type checks ...