Conference turris::languages

373.0. "ICOS/TTA" by FIGS::PRAETORIUS (one score, six and ten) Mon Jan 03 1994 16:57

     Just a coupla years after the VAX came to market and the iapX432
was announced (it's hard to say that it ever really came to market)
and the pendulum had swung about as far in the CISC direction as it was
gonna swing, the Berkeley RISC project started (the IBM 801 predated
this, but I don't think results were published until after the Berkeley
RISC paper appeared).

     Now that RISC is all the rage (and the semantic density of execut-
able code is at an alltime low:-) (is there no enthusiasm for moderation?))
I think we're beginning to see the beginning of the swing back in the
other direction.  I recently pulled a paper off the net from the Univers-
ity of Amsterdam on Intelligent COde Scheduling and Transport Triggered
Architectures which, IMHO, is a seminal paper in the beginning of the
swing away from RISC (Alpha's gonna last 25 years right?).

     It also discusses the virtues of functional programming languages
for modern architectures (and mentions in passing the value of profile
feedback in the compilation process).

     Since I can't offer it for DAP access from where I am on the net,
you'll find it in Figs::Erlang note 3.0.