[Enough already] Re: Proposal3: Make $_ a valid identifier character

JArchibald at aol.com JArchibald at aol.com
Thu Jun 1 07:22:48 UTC 2000


=> 6/1/00 12:36:59 AM EDT, ok at atlas.otago.ac.nz [personal communication] =>
<< The great ideas in Smalltalk are semantic.  They'll survive the new 
additional Squeak syntax (whatever it happens to look like).  And they'll 
survive underscores. >>

I agree completely. I tend to think of a Smalltalk program as a semantic 
construct associated with the parse tree for that program. The nature of this 
semantic construct is unfortunately ungrounded, somewhat waving the the air. 
The byte-code representation (the Blue Book description) is treading water 
very fast to be considered the standard for Smalltalk definition; I consider 
that it is rapidly on its way to becoming obsolete. It certain is obsolete in 
_my_ mental construct associated with the parse tree. At the moment, though, 
I don't see an alternate to the Blue Book description on the horizon. 
Unfortunately.

Nonetheless, you are correct, Smalltalk will survive underscores. But as a 
practical matter, my E-mail was attempting to state, in a somewhat rough and 
tumble manner, that underscores are not going to be introduced in this round. 
IMnsHO.

However, if the great ideas in Smalltalk are semantic (to which I agree), 
then why is there so much heat (and little light) being presented on this 
issue? I haven't counted them (I just did, I counted 44); why are there so 
many E-mail postings on this issue? And why are these E-mail postings so 
angst ridden? And why are these E-mail postings failing to converge (and 
cease and desist)?

<<  <<Do you think 
WE_DONT_WRITE_ENGLISH_THIS_WAY_SO_WHY_SHOULD_WE_WRITE_PROGRAMS_THIS_WAY is an 
improvement?>>

This is an invalid argument having the same form as "A.  I think salt is 
better than sugar on pies. B.  Oh yeah?  Then you must really love salt by 
itself." >>

I don't think it is an invalid argument. I don't think of it as an argument 
at all; it was a demonstration, as was your (correctly formed Smalltalk 
global variable) identifier preceeding it. User interface considerations 
associated with programming languages depend upon a common view shared by the 
users of the language. This common view (of the user interface 
considerations) comes from some rather subjective evaluations and sometimes 
practical admissions. There is some magic in the lexical fundamentals and 
syntax (as well as the semantics) of a computer language (certainly in 
Smalltalk) that cannot be analyzed entirely from a strictly formal 
perspective.

Smalltalk appears to be the simplest language around (in terms of its 
semantic definition) that has massive computation
power. FORTH, LISP in its original habitat, and Fortran (owhy) are unarguably 
simpler, but LISP bigots aside, Smalltalk has more descriptive power than is 
normally available in such a simple langauge. I want to know what they were 
smoking in Palo Alto in the 70's (and is there still any around) :-). Squeak 
seems to be what we're smoking in the 21st century, and it appears to have a 
fine aroma and bouquet. And besides, the price is right.

Cheers,
Jerry.
____________________________

Jerry L. Archibald
systemObjectivesIncorporated
____________________________

[owhy (for those still Dana Scott challenged) :- 'or what have you']





More information about the Squeak-dev mailing list