[FoRK] DEC in the past; was Re: What do you think of my
Joseph S. Barrera III
joe at barrera.org
Mon Jul 29 20:47:20 PDT 2013
Not sure how to reply without getting nitpicky. So I'll save most of it
for the next time we meet in person.
Most of what I admire in Cutler's work is much more at the kernel level
(as opposed to the shell level). E.g building the (NT) OS out of a large
collection of dynamically loaded libraries instead of a single binary
blob a la Unix. And doing most interrupt processing in DPCs (a la ASTs
in VMS) instead of entirely at interrupt level again a la classical Unix.
I would argue that (relatively) fast fork() was a seductive trap that
took years to get away from. Modern server architectures use process
pools or thread pools instead of relying on cheap process creation
(which in general is very difficult in any OS).
ps in VMS was SHOW PROC /ALL. You could abbreviate all the commands so
in practice it wasn't that verbose.
I suspect you mostly lacked VMS-knowlegeable mentors who could show you
how to do the basic stuff. When I first shifted from VMS to Unix I was
completely frustrated until I found someone who could translate the
basic tools for me.
On 7/27/2013 9:53 AM, Stephen D. Williams wrote:
> On 7/26/13 7:10 AM, Joseph S. Barrera III wrote:
>> On 7/26/2013 1:35 AM, Stephen D. Williams wrote:
>> > He [Dave] was apparently good at OS algorithms, but most of the
>> most obvious OS architecture choices were wrong for both VMS and
>> Which architectural choices do you have a problem with?
> Format typed files, at the OS level.
> No obvious (that I can remember) equivalent to /dev.
> No obvious-to-a-user structure of the operation system.
> Scripting language that didn't really support a subshell capability,
> so scripts ran in the current shell with all kinds of limitations that
> I hit right away.
> Overly verbose arguments to most programs, with lack of flexibility
> and power as I remember it.
> Nothing like piping commands at a shell.
> There was something I didn't like about volumes or filesystems, but
> that's too foggy now.
> Starting processes was extremely slow. Since I didn't know how to do
> the equivalent to ps, I'm not sure how loaded the system was, but my
> impression was that it was idle much of the time. Yet still extremely
> slow to start processes.
> I also installed and ran MicroVaxes at the time, but still found it a
> poor experience compared to early Unix that I was using at the same time.
>> BTW keep in mind that VMS supported tens of users with only 1 (one)
>> MB of RAM. In retrospect that is just fucking amazing.
>> - Joe
> FoRK mailing list
Captain Pierce was a strong man
Strong as any man alive
Stuck in his craw that they made him retire
At the age of 65
More information about the FoRK