Re: MP, tools & algs

Get better scientists.
better intellectual tools than computers.
A computer is a physical tool to leverage intellectual tools. I agree
Sometimes, it's not enough. Engelbart learned that.
Something about a brick....

In article <GN-dna7oPbNh4aXZ4p2dnA@xxxxxxxxxxx>,
David DiNucci <dave@xxxxxxxxxx> wrote:
I think Engelbart would be on my side on this one--e.g. from the same
section of his 1962(!) paper where he described brick writing, "It seems
reasonable to consider the development of automated external symbol
manipulation means as a next stage in the evolution of our intellectual
power."

I've seen his brick photo.

The nice discussion that I've had with him was to consider other better
tools such as Sherlock Holmes analog using a hand lense. He has just
now seen that that might be the better way to go. But last we spoke
he got some money from the NSF. A small chance I may see him tomorrow
nite at a monthly dinner.

Regardless, it's always good to read about his work. And, born in
Portland, worked at Ames at Moffett, looking for tools to augment
intellect...but maybe the similarities end there.

Yes, Doug last came by when a German film crew came to film where he
used to work in the 40x80 but was then closed. I mean no one except a
an even normally small number, then smaller had any keys. It was just
reopen by the USAF and USA.

I recommend Markoff's latest book on that era.

Still not enough. Just think what Woz would have done minus Jobs.

So was Engelbart Woz or Jobs? Some techies can also sell.

Doug is likely neither. Doug is in a class by himself in that era
when punch cards still reigned supreme. What's so amazing about Doug
is that he still has a huge group of followers from that era and to this
day. They fall in 2 camps (subgroups): the technical guys who worked
with him at the Augmentation Research Center like Bill English and the
rest whom I occasionally see even in grocery stores like Trader Joes,
and then there were/are the less technical groupies the Stew Brands,
the Ted Nelsons (and gusss Kevin Kelly was one [at Kevin's marginal
LongNow talk last month]). I am not certain why the Catmulls and Smiths
can't sell a Pixar as well as a Jobs.

We are discussing diagrams (schematics
"Then a miracle happens" areas.

(I guess this is OT, but I'm not on whatever mailing list you keep
mentioning, and c.l.visual seems out of commission.)

Oh, its a hackers thing.

That's just data flow, how about control flow--re my parenthesized
comment above? People like to see control dependences as well. It's
why structured control constructs utilize indentation, to add another
(partial) dimension, to allow the control points of a structured
construct to be visually related to each other (by being juxtaposed at
the same level of indentation) as well as being related to the
neighboring statements (by being juxtaposed, period). Entities which
are juxtaposed in a program trace are juxtaposed in the code (using one
definition or the other). I consider this similarity of code and trace
the essence of structured programming.

I think the problem is your word: See. That's why I think Parnas and
Habermann variously considered global variables harmful. The problem,
like many, became scale. Single words of memory weren't a big deal.
Arrays, array handling and inconsistent array state likely also contributed
to data flow's death.

Indentation was an afc thread which involved Python.


Once you get into parallel programming, and the dependences (both data
and control) go all over, and the program trace becomes a partial
ordering/DAG rather than a sequence, indentation isn't sufficient.
Still, if you represent data and control dependences in the program by
graphical connections (e.g. lines) and consider operations as being
proximal in both the program and the trace whenever they're
connected, you get the same essence of structured programming. (I'm
sure I discuss this in some paper, too.)

You will still have problems with exception handling.
Anyways, I'm not working on that anymore.

Look at Usenet. It survives.

At least so far (though Big-8 is apparently neutered for now), probably
on purely political merits alone (shared ownership, lack of censorship,
etc.). If one could create a blog with those properties, it'd probably
win out over Usenet due to other factors.

Maybe a Wiki more than a blog.
Depends. Hard to say. I know the security guys who prefer NNTP.
Listservs are still in use.

The prophet has a problem.

Smart/living messengers have learned to carry some good news with the bad.

Who is your Judas? [I just saw the news before the weekend.]

"what's an algorithm?" question.
"parallel algorithm" becomes at best undefined and at worst an oxymoron.
....
Think "data parallel" 8^). Danny did, and look where it got him.

So where did it get him? I don't know if I'd consider the CM1/2
especially successful, though it was unique and an important experiment.
Not that I've gotten further.

In the short run, it wasn't successful. But I like his idea of a 10,000
year clock out in the middle of Nevada (really Eastern), independent of
computers. Danny had to pay Gordon Bell for his mid-1990s bet.

He did spawn a few low level ideas as a reaction to the CM, just not
visible in public in detail.

Yes, data parallelism does evade the oxymoron, by leaving algorithms
sequential and making the operations parallel. (Often requires
considering loops as operations with their bodies as arguments, but
that's OK.) That's almost cheating, though, not to mention specialized,
pretty restrictive, and not particularly efficient, which is probably
why that machine evaporated. Much better to make the
"algorithm-like-thingy" parallel and the operations sequential--or at
least decomposable to deterministic atomic pieces.

I hear you.

I suspect that interpreters helped language development more than
compilers writers are willing to accept. But this is a guess on my part.
I mean that I realize that APL was interpreted, but I don't Iverson's
notation and semantics well enough. The place to watch might be
something like Matlab combined with spreadsheets.

[Moore's Law]
It's not double every 18 months, it's quadruple every 3 years.
Double is only used by managers and journalists.

I didn't understand your "it's not double" statement.

I think "doubling" is deceptive. It's like talking about bytes on word
oriented architectures like Crays, DEC-10s/20s, Univacs, etc.

Look, the improvement in mask resolution is a 2-D issue: Newton and
inverse square laws (it's slice and dice in this case).

--
.