dean@pplace.COM (Dean Ritz) (09/24/88)
Hello Comp.lang.smalltalk,
We at ParcPlace Systems enjoy reading all the recent net activity
regarding the Smalltalk-80 system. We'd like to take this opportunity
to correct two incorrect statements about the ParcPlace Smalltalk-80
system which have appeared both recently and in the past.
1) While it is true that most Smalltalk-80 systems "use byte-code
interpretation," the Smalltalk-80 systems produced by ParcPlace Systems
compile to and execute native code. They are built upon innovative
compiler technology, called "dynamic translation." In brief, we use a two
stage compiler which compiles from source text to byte-code, and then from
byte-code to native code (which is saved in a cache and executed as needed).
This technology enables Smalltalk-80 to have the compactness and portability
of byte-codes, and significantly better performance than interpretion technology
permits. For technical background, see the paper by Deutsch and Schiffman in
the proceedings of the 1984 POPL conference.
2) The second incorrect statement is that "ParcPlace Systems shows little interest
in [linking to assembly/C/C++ code]". Our current releases on the Apple
Macintosh, Sun-3, Sun-4, Sun386i, HP-9000, Compaq DeskPro 386, IBM PS/2-70,
and Apollo workstations supports user defined primitives. These
enable the integration of routines written in common languages like C.
Best Regards to all,
Dean Ritz
ParcPlace Systems
E-mail: Info@ParcPlace.com
Phone: 415.859.1000jr@amanue.UUCP (Jim Rosenberg) (09/25/88)
In article <364@parcplace.pplace.COM> dean@pplace () writes: >We at ParcPlace Systems enjoy reading all the recent net activity >regarding the Smalltalk-80 system. We'd like to take this opportunity >to correct two incorrect statements about the ParcPlace Smalltalk-80 >system which have appeared both recently and in the past. Has ParPlace ever gotten its act together and announced a *PRICE* for run-time systems so that "real" applications can be written in Smalltalk-80? It's not exactly viable commercially to write an application in a language where the end-user has to pay $1K just to license the langauge!! I've mentioned this to ParcPlace folk a couple of times. Each time they say yes, yes, we're going to have "reasonably" priced end-user run-time licenses -- but then they can't tell me what it will really cost. Digitalk wants $500/year for an unlimited run-time license, BTW. -- Jim Rosenberg CIS: 71515,124 decvax!idis! \ WELL: jer allegra! ---- pitt!amanue!jr BIX: jrosenberg uunet!cmcl2!cadre! /