[prev in list] [next in list] [prev in thread] [next in thread] 

List:       imap
Subject:    
From:       Mark Crispin <mrc () cac ! washington ! edu>
Date:       2003-04-24 21:27:37
[Download RAW message or body]

On Thu, 24 Apr 2003, Larry Osterman wrote:
> And I don't think we want to tie the implementation of the IMAP protocol
> to such underlying OS mechanisms.  Without disparaging any current OSs,
> simply consider mrc's original TOPS-20 implementation.  It is completely
> possible to implement an RFC 2060 compliant IMAP server on a TOPS-20
> system, however I do not believe that it is possible to implement this
> kind of asynchronous mechanism on TOPS-20 (without significant kernel
> support), since the operating system doesn't support either asynchronous
> I/O OR threads.

Actually, both asynchronous I/O and threading are quite easy to do in
TOPS-20, especially if you program in PDP-10 assembly language as god
intended...  :-)

Instead of process trees, processes, and threads as on UNIX, TOPS-20 had
jobs and forks.  A job was somewhat between a process and a process tree,
and a fork was somewhat between a process and a thread.  Interprocess
communication (both between jobs and forks) was much more powerful than on
UNIX; and you could share (or not share) address space in whole or part,
readonly, copy-on-write, or readwrite.  On the flip side, CD was global
for the entire job, and a fork couldn't be disowned by the job the way a
process tree can disown a process on UNIX.

This moment of history was brought to you by...

-- Mark --

http://staff.washington.edu/mrc
Science does not emerge from voting, party politics, or public debate.
-- 
-----------------------------------------------------------------
 For information about this mailing list, and its archives, see: 
 http://www.washington.edu/imap/imap-list.html
-----------------------------------------------------------------

[prev in list] [next in list] [prev in thread] [next in thread]