Microsoft open sources GWBASIC

John Ames commodorejohn at gmail.com
Wed May 27 14:40:12 CDT 2020


Liam Proven wrote:
> I don't know. There is a huge amount of tradition and culture in
> computing now, and as a result, few people seem to have informed,
> relatively unbiased opinions. There hasn't been much real diversity in
> decades.
>
> 25 or 30y ago, people discussed the merits of Smalltalk or Prolog or
> Forth; now most people have never seen or heard of them, and it's just
> which curly-bracket language you favour, or does your preferred
> language run in a VM or is it compiled to a native binary.
Agreed. While I'm much more favorably disposed towards C than you are,
the increasing homogeneity of almost all modern languages is
discouraging and, I think, detrimental to the field as a whole. Forth
and Smalltalk alike were eye-openers when I discovered them (and
Smalltalk in particular was a breath of fresh air, after I'd spent
years failing to ever really grok OOP with the likes of C++ and Java,)
because both presented genuinely *different* and beautifully
consistent ways to think about structuring and specifying a computer
program. These days, though, outside of deliberately jokey
ultra-esoteric languages, it's pretty much just a bunch of
domain-specific Java/Javascript knockoffs from horizon to horizon.

> I am just surprised that this (to me) rather inelegant design survived
> and got to market, given what you've said about the same company's
> ruthless drive for cost-cutting removed one PCB trace even though it
> killed floppy-disk performance, or wouldn't use an extra ROM chip
> because it was too expensive.
>
> It seems inconsistent.
It's marketing - consistency there is a non-consideration, if not
actively striven against. The whole saga with CP/M on CBM was a
boondoggle - the CP/M cart existed because business customers wanted a
CP/M add-in to run their spreadsheets and their whatnot, but it didn't
end up being a good fit for reasons already stated (slow CPU, slow
disk, 40-column only.) The 128 improved on those points, but not
nearly enough to become competitive with the advancements CP/M
machines had made in that time, and in the process wasted precious
man-hours and drove up the cost and complexity of the unit - and all
the while CP/M had been losing ground to MS-DOS in the business market
for years! But marketing promised it, so it had to happen... :/


More information about the cctalk mailing list