[comp.sys.apple] Porting XLISP; ][ Languages

secrist@msdsws.DEC.COM ("Richard C. Secrist") (03/11/89)

The latest incarnations of XLISP are too big to simply compile under C
and run on a ][-class machine within 64K.  It would be best to use an
earlier version of XLISP less than v2.0.  I know V1.4 fit under CP/M 2.2,
which has similar memory constraints to a ][-class machine, so that might
be the best place to start.  I've built V1.4 on a VAX under VMS no sweat,
using the DEC C, but never had the patience to MANX upon it for DOS 3.3.

As to the dearth of ][ languages, there are actually quite a few hiding
in the corners and/or inactively marketed by perhaps bankrupt companies --
although perhaps there aren't so many for ProDOS per se.  If you add a
cheap CP/M-80 card to an Apple though there is essentially no language
you can't get, even APL.  In fact if you have one of the Applicard 6 MHz
Z-80s you might run faster than an otherwise unenhanced ][ anyway, although
maybe not a 'gs.

On the ][ native there is APP-L-LISP for DOS 3.3, FORTHs for every O/S,
PASCALs for every O/S, several LOGOs, Fortran-77 (Subset G) for UCSD,
Modula-2 for UCSD, COMAL, C for DOS or ProDOS, and a fistfull of compiled
BASICs.  If you cound CP/M then you can add all of the previous languages
plus APL, Fortran-IV, PL/I, COBOL [perish the thought], a blue million
obscure languages, and more...

Some of the languages are pretty neat, some of them are less than robust,
but you can program in more languages on a ][ than most people have time
for.  There aren't sexy interactive debuggers and windows for the most
part, and I'm not recommending any of it for amateurs, but if you want to
chase icons around all day with a mouse and hunt for things in pull down
menus you belong on a different system anyway...

rcs

lwv@n8emr.UUCP (Larry W. Virden) (03/11/89)

In article <8903102123.AA16292@decwrl.dec.com> secrist@msdsws.DEC.COM ("Richard C. Secrist") writes:
-->
-->The latest incarnations of XLISP are too big to simply compile under C
-->and run on a ][-class machine within 64K.  It would be best to use an
-->earlier version of XLISP less than v2.0.  I know V1.4 fit under CP/M 2.2,
-->which has similar memory constraints to a ][-class machine, so that might
-->be the best place to start.  I've built V1.4 on a VAX under VMS no sweat,
-->using the DEC C, but never had the patience to MANX upon it for DOS 3.3.

That sounds like a good start.  But realize that with clever programming,
one need not limit one's self to 64k.  For instance, one could potentially
use bank switching on a IIe/c/c+ since they quite often have 128-1000k of
memory.  And for the IIgs, one has an 8 meg limit... my Vax 11/785 only
has 8 meg of real memory....
-->
-->As to the dearth of ][ languages, there are actually quite a few hiding
-->in the corners and/or inactively marketed by perhaps bankrupt companies --
-->although perhaps there aren't so many for ProDOS per se.  If you add a
-->cheap CP/M-80 card to an Apple though there is essentially no language
-->you can't get, even APL.  In fact if you have one of the Applicard 6 MHz
-->Z-80s you might run faster than an otherwise unenhanced ][ anyway, although
-->maybe not a 'gs.
That is true - and of course if I buy a PC Transporter then I can have access
to all of the MSDOS languages.  But if I wanted to spend that much money,
I would go ahead and pay $1400 to buy a 16mhz, 80 meg drive 80386 and run
Unix.  What I want to do is run MINIX on a NATIVE IIgs or II series computer.
Surely a 65c02 or 65816 is not THAT much more limited than an 8088?

-->
-->On the ][ native there is APP-L-LISP for DOS 3.3, FORTHs for every O/S,
-->PASCALs for every O/S, several LOGOs, Fortran-77 (Subset G) for UCSD,
-->Modula-2 for UCSD, COMAL, C for DOS or ProDOS, and a fistfull of compiled
-->BASICs.  If you cound CP/M then you can add all of the previous languages
-->plus APL, Fortran-IV, PL/I, COBOL [perish the thought], a blue million
-->obscure languages, and more...
Okay, I forgot about Logo.  Note that there is only Pecan Pascal for Prodos;
Apple no longer is supporting Apple Pascal and Kyan is extinct.  Fortran
and Modula 2 also are only available from Pecan.  What is comal and is it
being actively supported?  ProMAL, another language being sold for the Apple
II series, is no longer being supported.  I only know of Beagle Basic is one
of the few compiled basics available for the whole Apple line - if Zbasic
is still around they are the other.  The problem is that why many of these
languages WERE being supported, most of these, along with most all other software
is no longer being supported on an Apple platform.

Folks try to rationalize the Apple II death away by saying 'well, as long
as I am using the machine, it isnt dead.'  Okay, by your definition this is
true.  But by most of our definition, if there are no more magazines (or
only 1-2) , no more software publishers (or only 1-2 small garage operations)
, etc then the computer is dead.  Witness the Vic 20, TI 99/4, Exidy Sourcerer,
etc.  While many of these computers (PET 2001 even) may still run - you
can get very little softwar efor them.  For that matter, how much software can
you get for a Mac 128k or even 512k?


-->
-->Some of the languages are pretty neat, some of them are less than robust,
-->but you can program in more languages on a ][ than most people have time
-->for.  There aren't sexy interactive debuggers and windows for the most
-->part, and I'm not recommending any of it for amateurs, but if you want to
-->chase icons around all day with a mouse and hunt for things in pull down
-->menus you belong on a different system anyway...
-->
-->rcs


-- 
Larry W. Virden	 674 Falls Place, Reynoldsburg, OH 43068 (614) 864-8817
75046,606 (CIS) ; LVirden (ALPE) ; osu-cis!n8emr!lwv (UUCP) 
osu-cis!n8emr!lwv@TUT.CIS.OHIO-STATE.EDU (INTERNET)
The world's not inherited from our parents, but borrowed from our children.

c08_d042@jhunix.HCF.JHU.EDU (Stdnt 42) (03/13/89)

Yeah there might be tons of languages for the Apple II but PRACTICALLY
ALL OF THEM ARE INTERPRETERS!!!!  Which makes them totally useless except
for limited learning purposes.  I have yet to see a decent prodos-based
Pascal or C COMPILER that produces STANDALONE MACHINE CODE!! and don't
say Manx Aztec C because that thing prints text slow as all hell.  What
we need are compilers that handle memory management decently, allow
automatic overlay activation if the program is run on a II+ with only 64k,
and a decent meshing with the o/s.  This might already exist, I dunno though,
I haven't really seen it.  Hyper C Prodos looks promising but they are
bankrupt I believe.

Who knows maybe someday....

chris coleman      c08_d042@jhunix