Archives / Search ›

Tonight has been amazing. Got so much work done. Listened to an interview of Fred Eaglesmith on FolkScene immediately after the infamous Terry Gross/Gene Simmons one. Couldn't be more different.

Found more bugs in ICeCoffEE, fixed, refactored considerably, releasing soon I hope. Desperately need a test suite.

After getting one too many questions from someone using an outdated
version of ICeCoffEE, I realized it probably wasn't such a good idea
to leave links to old versions on my software page. So I
got rid of them, and added a few custom error
pages that I hope will be helpful for anyone following a broken link.

asdf asdf asdf. asdfasdfasdf asdfasdfasdf
asdf asdf asdf.

Live it.

Macros fixed and ALT text working. The problem was the weird
dichotomy in UserTalk's named parameter syntax. You declare a default
value by “name=value”, but you call it with “name: value”. Python
uses “name=value” for both. In some ways Frontier's method makes
sense, you're doing something different in each case, assigning to the
variable in the first and using the variable name as an identifier in
the second, but this never fails to catch me. It was worse when I was
programming in Python and UserTalk both on a daily basis.

Speaking of Emacs, it seems Radio is none too happy if you create two
files with the same name but different extensions. This has caught me
twice in the last day, once last night when I named a file “.old” and
wondered why the template didn't update, and once just then when Radio
preferred the file ending with a “~” (another type of Emacs backup
file; the ## one is from automatic saving). Blargh, should report
bug. I hope it gets some more attention than my previous text file
bug (which would have been astonishingly trivial to fix, like 30
seconds, but… oh well.)

Turns out Bill
Bumgarner
has written about this too. Thanks for the email, Bill.

‹ Newer Posts  •  Older Posts ›