Re: RE: A basic question on encoding Latin characters

From: Rick McGowan (rmcgowan@apple.com)
Date: Tue Sep 28 1999 - 20:12:13 EDT


Ken said...

> Pattern matching such as that described above for Expect is another
> disguised instance of ASCII-think in computers
> ...
> Is that an argument to keep encoding [...]
> without change? I don't think so.

And Robert Brady replied:

> Yes, there is no adequate solution. This is rather a shame.

The only "adequate solution" in this case is time. In the face of
requirements for new languages to be expressed, "ASCII-think" protocols will
become increasingly inadequate, and eventually disappear. In the past,
likewise, one might have lamented any number of technologies that, for one
reason or another, didn't scale well or didn't perform well enough in the
face of new challenges. So they are superseded by different technologies.
For interactive terminal protocols, one obvious solution is to design with an
unambiguous command/response termination, like line-feed. In other words,
don't design protocols like "login: " and expect them to scale appropriately.
 This is precisely the same problem that people with walkie-talkies used to
fix by using phrases like "over" and "over-and-out".

        Rick



This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:53 EDT