eugene@wilbur.nas.nasa.gov (Eugene N. Miya) (06/05/90)
I go to the great Pacific NW to get some vacation/kick back part time at
a conference etc. And this is what I get.
I endorse Dave's comment during the open session about the need for more
standardized reading lists in architecture. I do not think comp.arch is
the place to do it. It's too unmoderated and architecture means
eveything from hardware to software Unix architecture. [Confused?
let me tell you of my scale of softness {opposite the geologist's scale
of hardness 8) }].
Yes, I there is too much confusion largely developed over time between
terminology in architecture. But I am not a computer architect.
My infrequent postings of references are in the course of maintaining
my CAN bibliography on parallelism, and this includes non-architectural
software, parallel algorithms, distributed computing, etc.
Burton Smith noted this at Blake Island. In time, I've learned
the need for "Top-10" and "Top-100" for
starters. The SNR in comp.arch is too low, so with Steve Stevenson's
help I now use comp.parallel. That is a good group.
I think the IEEE and ACM
architecture communities could benefit from something like a good up to
date reading list/terminology/biblio.
I've talked to Doug Degroot, Duncan Lawrie, and others about this to some
degree. Note: in nearly all cases, I think most academics underestimate
some of the work involved. $1000 a year isn't enough. You have to get
the architecture community involved. You have to get architects who
read and used to read the net to annotate and comment good and bad
points. A librarian can't do that. I'm more of a software type.
But you can only benefit. That's science.
P.S. I've kept Kung's systolic array biblio and the RISC biblio as well
for standby. You just have to build on these.
Personally, I started off never knowing what a file was. There existed
DATASETS and DATASET NAMES. AND YOU APPROPRIATELY USED YOUR
// EXEC PGM=IEBGENER cards, and EBCDIC was THE CHARACTER SET and
everyone else was doing things in STUPID ways (you know ASCII).
But the ARPAnet saved me. Then I started working on Univac 1108s.
Boy, did I warp my head on those machines.
Every young programmer byte-oriented programmer should get a little
exposure (dosimeter required) to that world, then came Crays.....
Terminology and references are nice, but real hands on (see below), that's what
counts. Schools aren't providing this. Wish they did. That is the
kind of exposure students need and the early ARPAnet tried to provide.
That kid of survey is part of what a young architect needs to know.
In closing I think it would be a great idea if comp.arch came up with a
"standard" architecture reading list ["What Sieworek, Bell, and Newell
isn't enough? Should at least be on reserve in a library, however dated].
Review it every so often, publish it in CAN for those without net
access. Etc. etc.
I posted this at the urging of a few ISCA attendees, so if you have
comments, send email. To quote George Michael: "N-urchins from Berkeley
are welcome to forward this to Dave." 8) I think this is a chance for
c.a. (or perhaps some up and coming school) to prove me wrong and
do something useful for the field. . . .
But,.....naw.....
--e. nobuo miya, NASA Ames Research Center, eugene@orville.nas.nasa.gov
{uunet,mailrus,other gateways}!ames!eugene
Hacker "ethic"
from Hackers by Steven Levy
Page 40-45
Access to computers -- and anything which might teach you something about
the way the world works -- should be unlimited and total. Always yield
to the Hands-On Imperative!
All information should be free.
Mistrust authority -- Promote decentralization.
Hackers should be judged by their hacking, not bogus criteria such as
degrees, age, race, or position.
You can create art and beauty on a computer.
Computers can change your life for the better.