Documenting our decisions

[David A. Moon <Moon@SCRC-STONY-BROOK.ARPA>]

I have not seen the Lucid manual either, but I know from my own
experience that a language reference manual for users and a formal,
unambiguous language specification are two very different things.  In
other words, starting from the Lucid manual may not save a significant
amount of effort.

While TeX and Scribe are both widely available text formatters, and
there may be others, I would like to argue against using any text
formatter at all for the language specification.  In this project we
need to concentrate on content, not on style of presentation.  Wrestling
with a text formatter would simply be a distraction from our real
business.

But maybe it's better to back off from such implementation
considerations and first decide what it is we're trying to do.  Do we
really have the resources within these committees to write a formal,
unambiguous language specification?  Or should we be starting by
developing what amounts to an appendix to the Steele manual?