Re: Enlightenment via avoiding the T-word

[Ronald Bourret <rpbourret@rpbourret.com>]

Tim Bray wrote:
> Since
> nobody is arguing that it's a bad idea to have context-sensitive
> content models,

Actually, I have, at least half-heartedly. Part of my objection is the
expense of figuring out what content model applies, but most of it is
that I would like a label to tell me what something *is*. (This is
different from what something *applies to*, which usually is
context-sensitive.) On the other hand, the cases put forward both by the
OO community (Matthew, me) and the well-formed community (Evan) are good
use cases.

My gut reaction to this sort of fence sitting is to not support it and
add it later if there is demand. (Along these same lines, the best
hiring advice I ever heard was that if you only think a candidate can do
the job but are not really sure, don't hire them -- you'll almost
invariably regret it.)

-- Ron