My biggest problem with this abstract is: haven't we already done this four
or five times already? AFS/Andrew is the most complete and real
implementation I know of. But there's also Sprite and Amoeba and V. I know
OSF picked up AFS but I don't know what else of Andrew they picked up or
replaced. And of course there's the poor man's approach: NFS (Sun or
generic), with something on the side to keep everyone's remote mount tree
kinda sorta the same.
I think the coolest aspect of this project is the name. "So, Bob, what
would cause an OS to write garbage to its own boot sector?" (We are shown,
presumably through Bob's eyes, a sequence of rapid-cut, hand-held shots of
reference counts prematurely dropping to zero, structures freed, NULL
pointers being chased.) Bob: "This system will corrupt itself again. Under
heavy load, it builds itself up to a state of excitement, but the only form
of release it is able to perform is defecation."
- Joe
Joseph S. Barrera III (joebar@microsoft.com)
http://www.research.microsoft.com/research/barc/joebar
Phone, Redmond: (206) 936-3837; San Francisco: (415) 778-8227
Pager (100 char max): 1338993@roam.pagemart.net or (800) 864-8444
-----Original Message-----
From: Rohit Khare [SMTP:khare@w3.org]
Sent: Tuesday, January 21, 1997 4:48 PM
To: FoRK
Subject: Millenium
around
Joe2's web page...)
Bitchin' plan. Sounds like a load of fun, and the kind of industrial R&D
worth funding. How to measure progress, well, you got me...
http://www.research.microsoft.com/research/os/richdr/mgoals.html
Abstract
We believe it is time to reexamine the operating system's role in
computing. Operating systems exist to create an environment in which
compelling applications come to life. They do that by providing
abstractions built on the services provided by hardware. We argue that
advances in hardware and networking technology enable a new kind of
operating system to support tomorrow's applications. Such an operating
system would raise the level of abstraction for developers and users, so
that individual computers, file systems, and networks become unimportant
to
most computations in the same way that processor registers, disk sectors,
and physical pages are today.
--- Rohit Khare -- World Wide Web Consortium -- Technical Staff w: 617/253-5884 -- f: 617/258-5999 -- h: 617/491-5030 NE43-344, MIT LCS, 545 Tech Square, Cambridge, MA 02139