Lambda the Ultimate

Implementing Distributed Systems Using Linear Naming
started 3/26/2004; 8:32:41 AM - last post 3/31/2004; 3:48:12 AM

Luke Gorrie - Implementing Distributed Systems Using Linear Naming

3/26/2004; 8:32:41 AM (reads: 9686, responses: 3)

Implementing Distributed Systems Using Linear Naming

This (beautifully written) thesis describes a novel way to represent programs for automatic distributed execution. There is a research group carrying on the work. I believe at least one member of the group is an LtU reader.

How is the work going? Are there any implementations available? Where does it fit in with other research on automated parallel execution, and what is the current state of that whole area anyway?

Posted to Parallel/Distributed by Luke Gorrie on 3/26/04; 8:43:45 AM

Olivier Lefevre - Re: Implementing Distributed Systems Using Linear Naming

3/30/2004; 2:15:04 PM (reads: 125, responses: 0)

Woa. Something truly out of left field: not one of the authors' names rings a bell with me. May I know the exact context in which you stumbled on this?

Luke Gorrie - Re: Implementing Distributed Systems Using Linear Naming

3/30/2004; 4:38:19 PM (reads: 124, responses: 1)

This was recommended to me by Darius Bacon a few years ago. I think the context was that I was asking if people wrote doctoral theses that are enjoyable to read, in the hope of actually being able to read and understand one. This is really nicely written. I'm not sure how Darius got onto it.

Alan Bawden (the author) is one of the "golden era" MIT AI-lab hackers. I don't know very much about his other work, but he sometimes chimes in on the lightweight-languages mailing list, and has some entertaining contributions in a Unix Hater's Handbook.

A really splendid thing he's done is the Incompatible Timesharing System website. That contains tarballs of the file systems of the last two ITS machines, and is a real treasure trove. Among other things it contains the TECO sources to ITS Emacs, the Maclisp Manual ("Moonual"), and (I believe) the archive of Common Lisp's definition and standardisation effort (which was conducted over email).

Let us know if you find some treasures in there!

P.S., sorry about the slip of linking to a big postscript file without mentioning the author's name!

Darius Bacon - Re: Implementing Distributed Systems Using Linear Naming

3/31/2004; 3:48:12 AM (reads: 95, responses: 0)

Probably I'd seen some good posts from Alan Bawden on comp.lang.scheme and became curious about what else he might have done. Henry Baker's paper on linear naming might have been another vector... it's been quite a while.