Morendil comments on LessWrong anti-kibitzer (hides comment authors and vote counts) - Less Wrong

59 Post author: Marcello 09 March 2009 07:18PM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (52)

You are viewing a single comment's thread. Show more comments above.

Comment author: Morendil 02 July 2010 08:48:38AM 2 points [-]

Yep, I've been seeing just the same thing. It's the defect I've corrected, and it will go away as soon as Tricycle push it to production. That sometimes takes a few day or more.

The specific defect in the code might make an interesting anecdote in that "programming as a rationalist skill" post I've been planning for ages. My thesis is that every single "bug" in software is the result of a map-territory mismatch.

In this particular case, I'd been assuming that once a Web page was loaded, its CSS stylesheets could be accessed by a Javascript program through an array whose ordering wouldn't change. This assumption turned out to be wrong: the "AddThis" widget at the bottom of a post inserts a stylesheet right at the top of the array, shifting the main stylesheet one place forward, so that it is on that stylesheet that the AK script stomps.