Re: Lines in Startup.Int

[cld@xxxxxxxx (Carl Distefano)]

Reply to note from Wolfgang Bechstein  Thu, 21
Sep 2000 07:27:54 +0900

-> But I didn't know what Carl Distefano reports:
->
-> > The comment symbol is ;*; (semi-colon, star, semi-colon)
->
-> To me this symbolizes once again how utterly irrational
-> software (and even software as good as XyWrite) can be. Is
-> there any reason why the comment symbol for the startup file
-> should *not* be a simple semicolon as in the keyboard file or
-> the printer file, thus allowing users to deduce the fact for
-> themselves?

Actually, there was a very good reason to adopt an oddball
combination of characters, rather than a simple semi-colon (or any
other single character): to avoid feedback loops. In programming,
once you assign a meaning to a symbol, it often becomes unavailable
for other uses. This is due to the inherent dumbness of the
interpreter, which is very good at manipulating 1's and 0's but
hopeless when it comes to discerning context -- the dunce in the
machine, as it were. Since XPL is, above all, a text-manipulation
language, the semi-colon would have been a non-starter because it
occurs so often in the data to be manipulated. Programs, or rather
the computers attempting to execute them, would become "confused".

For example, it would become difficult or impossible to assign a
string like "The sky is blue; the grass is green" to a Save/Get. If
you wrote , the
interpreter, not being able to distinguish between a comment symbol
and a "real" semi-colon, would skip over essential code, and the
program would fail. It's far less likely that you'll ever need to
operate on a string like "The sky is blue;*;the grass is green".
It's a textual non-entity -- which is precisely the point.

This isn't an issue with keyboard and printer files, which are line-
oriented. There the parsing is easy: the comment symbol is either
the first character in a line, or it's not. XPL, however, doesn't
have "lines"; it's free-form. There are no contextual cues (such as
the first position in a line) that distinguish between "use" and
"mention". Hence the need for a unique and rarely-occuring symbol.

--
Carl Distefano
cld@xxxxxxxx
http://users.datarealm.com/xywwweb/