[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
- To: franz!ucbarpa!B.GP.CS.CMU.EDU!Scott.Fahlman
- Subject: Re: DEFAULT-CASE
- From: franz!frisky!jkf@ucbarpa.Berkeley.EDU (John Foderaro)
- Date: Fri, 17 Mar 89 11:36:14 PST
- Cc: franz!sail.stanford.edu!cl-cleanup
- In-reply-to: Your message of Fri, 17 Mar 89 11:19:37 EST. <8903171622.AA04031@ucbarpa.Berkeley.EDU>
I completely agree that all 'official' Common Lisp symbol names should
be use one case, and should there ever be a library of contributed programs
only those program that work in a case-insensitive mode should be
accepted. When Common Lisp starts up it should be case-insensitive
just like it is now.
>> I think that this is a recipe for confusion
>> and major portability problems. As soon as there is code around that
>> assumes case sensitivity, anyone who wants to use that code in conjunction
>> with code of his own has to go along with case sensitivity.
This is incorrect. I've had ten years of experience with using
code written in both case-sensitive and insensitive-modes in the same Lisp
(first it was debugging Macsyma in Franz Lisp, and now it is working
with programs like PCL in Common Lisp). The only difficulties
I had were due to the default case of symbol names (and that is what
my proposal is directed to fix). Keep in mind that if you
decide to live solely in the case-insensitive world you still have
access to all symbols, you may just have to type extra backslashes or
vertical bars sometimes.
>> We'll either have two incompatible software words within Common Lisp,
>> or it will drag us all through a major incompatible change we don't want.
I honestly don't know what you mean by this. Perhaps you could describe some
of your experiences or just some scenarios that you imagine might happen.
Keep in mind that there is already considerable freedom in
how people write programs. What if we started getting flooded with
great programs written in French, or Spanish or Italian which
many of us couldn't understand. Should we mandate that for portability
all programs shall be written in English (or perhaps Esperanto)?
What if someone developed a loop macro that half the users loved and half
despised? Should we not include that macro in Common Lisp because people
might start using it and then write code that half the users can't read
without feeling ill?
Remember, the only thing that will change if this and the
READ-CASE-SENSITIVITY proposal pass is that the default case for symbol
names will be lower. Absolutely nothing else about your world or
the way you program need change.
If you want, we can privately correspond on these issues and then
send a summary of what we've concluded to this mailing list.
- john foderaro