[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Issue: READ-CASE-SENSITIVITY (Version 2)
- To: peck@Sun.COM
- Subject: Re: Issue: READ-CASE-SENSITIVITY (Version 2)
- From: Kent M Pitman <KMP@STONY-BROOK.SCRC.Symbolics.COM>
- Date: Thu, 23 Mar 89 23:49 EST
- Cc: KMP@STONY-BROOK.SCRC.Symbolics.COM, jeff%aiai.edinburgh.ac.uk@NSS.Cs.Ucl.AC.UK, CL-Cleanup@sail.stanford.edu
- In-reply-to: <8903240214.AA04765@denali.sun.com>
I shouldn't even be bothering to reply to a message like this at this late
date. I have better things to be doing. However, I'll let this be my one
for the day -- if only to lend a little support to Jeff because I think
the tone of ridicule in your message to be somewhat out of line. In spite
of this, I've tried to keep my tone constructive and to answer your questions
Date: Thu, 23 Mar 89 18:14:53 PST
>(2) the lack of need to add CHAR-INVERT-CASE (which I don't think is very
> useful outside of this context).
I guess i don't see how this is useful even in this context.
Is this a Symbolics'ism?
I didn't find this remark to be particularly professional. I wish we could
just avoid that kind of thing.
The answer happens to be "no", it is not something we use here.
It's something Jeff dreamed up, I guess.
I didn't oppose it because I'm not in the habit of out-of-hand opposing
things just because I personally don't see as having practical value. I
think the real acid test of willingness to cooperate on compatibility is
a willingness to tolerate noops and useless features because they turn
out to be useful to someone else.
My offhand guess is that it is in fact useful in some implementations.
Your example below which uses mixed case doesn't give it fair play.
Suppose there's an intermediate situation where you implement an embedded
language in which you can only use all-uppercase or all-lowercase names.
Suppose you want the all-lowercase names to be the ones that correspond to
Lisp names. I don't happen to want to do that, but it seemed plausible to
me that someone might -- and it might be what Jeff had in mind.
The cost of the feature he's asking for is very small, especially if you
consider the hair someone would have to go through to write that embedded
language portably if you didn't offer the feature.
If :preserve is an option, why would someone want :INVERT?
dOES SOMEONE THINK :invert IS EASIER TO TYPE THAN eSCAPES or vERTICAL-bARS?
dO YOU HAVE files WRITTEN WITH :invert?
How about throwing out :INVERT *and* CHAR-INVERT-CASE?
How about being civil and first asking Jeff politely why he wanted the feature.
Which of READTABLE-KEYWORDS or READTABLE-FUNCTIONS would you prefer then?
It doesn't affect my vote. I still prefer the former over the latter, and I
still don't seriously oppose the latter.
[given a sufficiently powerful Emacs that can escape the chars before
passing them to the Lisp reader, does any of this matter to X3J13?]
The issue is not text editors. Given a sufficiently powerful Emacs, you could
code in C and still pass your information off to Lisp. ``It's only software''
as they say. The issue is that the language must be defined as the interface
between the outside world and Lisp. The language is exactly what you can expect
to be held constant as you move from system to system, text editor to text editor.
Either the language handles case conversion or the text editor does.
Jeff is suggesting that he would like the text editor to do so. I don't happen
to want to do that, but I can't deny that he is making a fair request.
While we are busy trying to be KSR33 compatible, the rest of the world
may zoom on by.
I have never used a KSR33. I think I've seen one. I have no particular desire
to be compatible with one. I can't imagine why this remark is relevant here.
The Japanese won't be interested in much of this code.
Oops, sorry, that is not a cleanup issue.
In my mind, it is not our purpose to design a language suitable for the Japanese.
It is our purpose to design a language suitable for us, and to try to listen
to the Japanese (and anyone else) about problems what we do might cause. While
this feature might not be interesting to them, it's hard to see how it could
cause them any problem.
The Japanese will have their opportunity to speak, and I will pay close attention.
If you say what you personally want and why, sans ridicule, I will try to pay
close attention to that, too.