Muli Ben-Yehuda's journal

Day 10 was rather uneventful. Didn’t work out (gasp!) but did go out for a nice long walk in the evening with Orna.

Day 11 is today. As has become my custom lately, the first thing I did in the morning we get weighted. I’m already 1.1kgs down for this week, which is pretty damn nice, especially as I don’t feel particularly hungry. I think the intense exercise is paying off in increased metabolism, and that’s just swell.

Got to the gym in the morning, did a nice cross training session, which left me feeling great. That’s a pretty good sign that I need to increase the dossage, and I’ll be adding short runs after the cross training and before the weights starting from next Sunday.

All in all, I’m feeling good, and starting to feel some slack at the waste line of my pants. Yay!

“The Lisp Machine is a personal computer. Personal computing means that
the processor and main memory are not time-division multiplexed,
instead each person gets his own. The personal computation system
consists of a pool of processors, each with its own main memory, and
its own disk for snapping. When a user logs in, he is assigned a
processor, and he has exclusive use of it for the duration of the
session.”

“Each logged in user of the Lisp Machine system has a processor, a
memory, a keyboard, a display, and a means of getting to the shared
resources. Terminals, of course, are placed in offices and various
rooms; ideally there would be one in every office. The processors,
however, are all kept off in a machine room. Since they may need
special environmental conditions, and often make noise and take up
space, they are not welcome office companions. The number of
processors is unrelated to the number of terminals, and may be smaller
depending on economic circumstance.”

“The memory is typically 64K of core or semiconductor memory, and is
expandable to about 1 million words. The full virtual address space is
stored an a 16 million word disk and paged into core (or
semiconductor) memory as required. A given virtual address is always
located at the same place an the disk. The access time of the core
memory is about 1 microsecond, and of the disk about 25
milliseconds. Additionally, there is an internal 1K buffer used for
holding the top of the stack (the PDL buffer) with a 200ns access time
(see [CONS] for more detail).”

“The complete LISP machine, including processor, memory, disk,
terminal, and connection to the shared file system, is packaged in a
single 19″ logic cabinet, except for the disk which is
freestanding. The complete machine would be likely to cost about
$80,000 if commercially produced. Since this is a complete,
fully-capable system (for one user at a time) it can substantially
lower the cost of entry by new organizations into serious Artificial
Intelligence work.”

Oh, just go read the thing already 😉

Been reading a bunch of interesting papers and notes lately.

• 
  Thoughts on Social Processes and Proofs of Theorems and Programs, by
  De Millo et al., discussed on this blog previously.

• 
  The GNU 64-bit PL8 compiler: Toward an open standard environment for
  firmware development, W. Gellerich et al. This is a very
  interesting paper, describing the PL.8 (“80% of PL/I”) which is used
  for zSeries firmware development, and developing a gcc front end for it. (via Lambda the Ultimate).

• Lisp
  Machine Progress Report, by the Lisp Machine Group. Some choice
  quotes from this venerable piece of history coming right up. (via Lambda the Ultimate).

• Complexity
  (Notes on Works in Progress), by Jonathan S. Shapiro. Could’ve
  been called “why C is not my favorite language” (via listpmeister.com).

• 
  Compiler Optimization-Space Exploration, by Spyridon
  Triantafyllis et al. Interesting, but not earth
  shattering. Compilers choose optimizations based on heuristics, and
  heuristics are sometimes wrong. Trying all optimizations is time
  consuming. Let’s find a way to minimize the search space by only
  considering interesting heuristics, trying them, and choosing the
  best.

Date:   Tue, 18 May 2004 21:00:07 +0100
From: Al Viro
User-Agent: Mutt/1.4.1i
To: Mark Gross 
Cc: Christoph Hellwig, Tim Bird, linux kernel
Subject: Re: ANNOUNCE: CE Linux Forum - Specification V1.0 draft

On Tue, May 18, 2004 at 12:32:48PM -0700, Mark Gross wrote:
> These are some of the types of problems engineers at REAL software shops have
> to solve to be able to ship REAL product for REAL money.  If you haven't HAD
> to produce code like this yourself at some point in your carrier then you've
> lived a sheltered life.
>
> Its disingenuous for you to get on your ivory tower to point and laugh.

        Well, you see, after spending years cleaning up the excrements
of self-styled "REAL engineers" it's either get on the tower to point and
laugh or get on the tower to point and shoot.