[rescue] Personal progressions (was: what to do with a dec alpha 255)

[Carl R. Friend]

    On Thu, 31 May 2007, Sheldon T. Hall wrote:

> My resume claims that I've used 25 programming languages on 9 operating
> systems.  If I were looking for a job, I'd probably update that.

    I've lost count of how many  OSes I've used both professionally
and in hobbyist/hacker mode.  The short list, more or less in
chronological order goes DG RDOS, TOPS-10, RT-11, RSTS/E, RSX-11M+,
VAX/VMS (in several revisions from 3 to 5), CLIX, AIX, ULTRIX, MVS/XA
(briefly), Windows NT 3.5 and 4 (3.5 was god-awful and was pretty much
bug-for-bug compatible with VMS 3), HP-UX, IRIX, SunOS 4, Solaris
(2.5.1 to 10) on SPARC, Symbolics Genera, DG AOS, OS-8, LAP6/DIAL.
I can absolutely guarantee I'm missing a few in there.  The language
list looks similarly eclectic, and I've forgotten more than I use now
(but, like riding a bicycle or shooting, it comes back pretty quick).

    Digging about in history in "hacker mode" is great fun because it
shows up, in pointed detail, that "we've been here before, we've made
the same mistakes before, and we didn't learn from them the first time
'round.  Systems like Data General's AOS from the late 1970s are a
pure joy to use, if somewhat verbose compared to UNIX (and clones),
but not so much as VMS (which once it got to V5 was actually very
stable, useful, and friendly; the same holds true for Mica, aka
Windows NT, and once it got to V5 ("Windows 2000") or so it, too,
save for the rubbish piled atop it, was pretty good as well).

    And some things boggle the mind -- like LAP6/DIAL which is a full-
blown IDE (Integrated Development Environment) that ran on LINC-8s
and PDP-12s beginning in the 1960s in as little as 4k of memory and
used block-addressable tape for mass-storage.  With that one, you get
an interactive on-screen editor that you move the cursor around using
potentiometers on the front-panel; the effect can be remarkably mouse-
like.

    Then there's humour -- like when Intel claimed it's technique of
"Extended Memory Access" (or whatever they called it) on the 32-bit
architectures to address more than 32 bits' worth of address space
was "all new" (and probably tried to patent it) when, in point of
fact, it was a re-warming of the *original* notion of paging where
the memory-management hardware was there to increase the amount of
physical memory a machine could have even though the logical address
space didn't change (see the pdp10s starting with the BBN pager for
the KA, the entire -11 line, DG Novas and Eclipses, and lots of other
16-bit machinery in the '60s and '70s).

    Even that old chestnut P-Code has come back once again, only this
time we call it "Java".  And, like P-Code, we still don't have quite
enough cycles available to make it perform terribly well in a real-
word setting.  The "thin client"?  See "di(s|c)kless workstation".
The list goes on and on.....

+------------------------------------------------+---------------------+
| Carl Richard Friend (UNIX Sysadmin)            | West Boylston       |
| Minicomputer Collector / Enthusiast            | Massachusetts, USA  |
| mailto:crfriend at rcn.com                        +---------------------+
| http://users.rcn.com/crfriend/museum           | ICBM: 42:22N 71:47W |
+------------------------------------------------+---------------------+