Last night at ABE.pm, I was talking a little bit with The Gang about some of the things I came to believe while doing the same thing in multiple languages. In explaining some of the issues I have with Ruby, both the scope of variables and the resolution of methods, this example came to me:

There’s been a fresh outburst of acrimony about how CPAN testers just send unpleasant and useless email to people who don’t care and can’t benefit. Coincidentally, I got one of the most useful bug reports I’ve gotten in a long time from an automated tester.

Rx was a lot of fun to implement. It’s all about data in memory, not representation in files, so I got to design it that way, too. I spent a lot of time thinking about what the various kinds of in-memory structures were that I’d need to validate, and then I’d think about how to represent schemata to validate them.

Okay, the title might be a filthy lie, but it’s just a reference to my previous posting about the fact that I couldn’t find a single data validation system (read: schemata) for JSON-like data. I found plenty of schema languages for XML, one for YAML that was never going to be suitably cross-platform, and one for JSON (json-schema) that seemed over complicated and likely to become unmaintainable, and then some other things that don’t warrant much mention.

Today I spent a good while trying to figure out why I wasn’t seeing a runtime error from code that looked like this (grossly simplified):

…or at least the problem that came to light recently.

I don’t like to use the /ignore command. First of all, I very rarely choose to frequent an IRC channel with anyone I really can’t stomach. Also, as you ignore more people, conversation begins to become incomprehensible, because threads of conversation start and you can’t tell why or who all is involved.

The PHP quick reference lists all the built-in functions – all 5274 of them. It just links you to the manual page for that function. This HTML document is over four hundred kilobytes in size.

At work, we recently moved our internal project and defect tracking from Trac to LiquidPlanner. I’ve said it before, and I’ll probably say it again: LiquidPlanner is fantastic. It’s really helping get an eye on what’s going on.

The canonical source for my journal posts is at rjbs.manxome.org in my Rubric install. Rubric doesn’t do comments, though, and I am really, really not interested in adding that feature. That’s why I syndicate to use.perl.org.