9/28/97 11:14 PM
Subject: MAX Digest - 27 Sep 1997 to 28 Sep
1997To: Recipients of MAX digests 

There are 2 messages totalling 59 lines in this issue.

Topics of the day:

  1. David Jackson's SOUNDBEAM
  2. RAM, ROM, naming, and such


Date:    Sun, 28 Sep 1997 17:50:21 -0500
From:    Gordon Adams 
Subject: David Jackson's SOUNDBEAM

Sorry for a somewhat tangential post, but could anyone with experience with
David Jackson's "Soundbeam" please contact me privately?

Gordon Adams.


Date:    Sun, 28 Sep 1997 22:30:18 -0500
From:    Fred Collopy 
Subject: RAM, ROM, naming, and such

Peter wrote:

>Except that lots of things are random access. Both my hard disk and my
>diskettes are random access media (as opposed to tape, fr'instance, which
>is sequential access).

The concept of random access functions in relation to time. Random access
is used to mean that the amount of time to access any item is such that
(for all practical purposes) you can organize its allocation randomly
without paying a time penalty when you access it. Hard disks are not such
beasts. Seek and latency make the amount of time to transfer a particular
piece of information a function of where that information is located on the
disk relative to the disk head's location. Hence, it might not be wise to
allocate randomly (and indeed a lot of thought is given to how segments
should be aligned, where the directory should be located, where data should
be stored relatively to the applications that use them, etc.).

>RAM is often used in opposition to ROM ("How much
>RAM does it have? How much ROM?") although the two acronyms have no
>direct relationship to one another, and they certainly aren't antonyms.

ROM should have been named RO-RAM. It is RAM! It happens not to be
writable. People got lazy (and professors don't like to correct students)
so we let them distinguish RAM and ROM by this writability criterion. It
even appears in most textbooks now. It is wrong. Some ROM is RAM (the stuff
in the chips inside your machine), some ROM is not (the CD-ROM). It is all
easy when you keep straight that the random accessibiity is about time, the
time it takes to get at stuff. If the time can vary in a significant way
(and in our world picosecs are significant), it's not randomly accessible
(and thousands of comp sci students are busy doing exercises to optimize
the locations of the information around various timing objectives).

I enjoy when this list ventures to address associated themes (such as this
one and our earlier discussion of randomness). I learn  about both the
subject in question, and how knowledge diffusion is prone to mutations
(even among the most informed and well-versed of the information culture;
it is little wonder that such vehicles as Time or Business Week seem to
almost never get these things straight). Thanks.



End of MAX Digest - 27 Sep 1997 to 28 Sep 1997