[net.lang.mod2] IS Modula-2 compiler -- let me know too!

bsa@ncoast.UUCP (Brandon Allbery) (09/12/85)

Expires:

I too am interested in the IS Modula-2 compiler and would like to hear any
experiences with it.  I've been waiting for something decent to come out
(I can't bear Pascal even with ``fixes'' -- the best I've ever seen was
the souped-up Touretzsky(?sp) compiler at Case which looked like the Pascal-
based early Ada model; it was barely tolerable.  C is ok but I would prefer
something that will catch me if I make a multiple-file no-no. . .

Innformation on support, how readable the manuals are, how complete the
manuals are, what machines are supported (I have an ITT XTRA with 256K RAM)
and so on would be appreciated.

--bsa
-- 
/****************************************************************************\
* Brandon S Allbery, 6504 Chestnut, Independence, OH 44131  +01 216 524 1416 *
* (phone and address subject to change in ca. 1-2 months when I get an apt.) *
* 74106,1032 CIS   BALLBERY MCIMAIL  TELEX 6501617070  ncoast!bsa@Case.CSNET *
\****************************************************************************/

krausej@stolaf.UUCP (James J. Krause) (09/17/85)

> I too am interested in the IS Modula-2 compiler....
> 
> Innformation on support, how readable the manuals are, how complete the
> manuals are, what machines are supported (I have an ITT XTRA with 256K RAM)
> and so on would be appreciated.


  I received a response to my query regarding the Interface Technologies' 
  Modula-2 compiler.  Unfortunately, it was not a positive one.  I decided to 
  post it to the net with the hopes that it might save someone the 
  disappointment of buying the product.  Here is the response I received
  (I am leaving the author anonymous):
       
       I bought a copy a few weeks ago and find it to be quiet 
       buggy.  The editor doesn't know modula all that well and 
       won't let you type in certain perfectly legal constructions.  
       The incremental compile feature forces you to save and re-
       load your program after editing it.  If you don't, it won't 
       notice the changes you've made and will re-generate the same 
       errors as before, pointing at the line(s) you just fixed!  
       The library modules are non-'standard' (i.e. don't match 
       Wirth's books).  The 8087 math library is totally non-
       functional (machine hangs on any reference and must be 
       powered down to recover).  The path through the menu driven 
       interface is bewildering sometimes.  The manual is totally 
       hopeless...I got 10 pages of TYPEWRITTEN changes which they 
       expected me to WRITE into the manual!!!  Even after all 
       this, the first example in the book is still wrong!  I've 
       stopped using the product and consider it an $80.88 lesson 
       in life.

       I have the Modula Corp (Provo, UT) product and like it a
       lot...it works, and is an 'honest' implementation.  Although
       the current version just generates M-code which runs under an
       interpreter, a native code compiler for both the PC and the
       Mac are in the works.

  If anyone has something good to say about the compiler, I'm all ears.
  Also, can anyone confirm the rumor that Borland is working on 'Turbo
  Modula-2'?  (I hope so...)

                                James Krause
                        ...!ihnp4!stolaf!krausej