Re: Canonical argument for (and against) layered protocol

Joseph M. Reagle Jr. (reagle@rpcp.mit.edu)
Wed, 01 Jul 1998 18:00:58 -0400


At 04:31 AM 7/1/98 -0700, Rohit Khare wrote:
>Is there a better, more canonical paper on layering as an abstract
>technique to itself?

If you find one, let me know. Interesting stuff. Reading the paper, my mind
couldn't help but think, "this is what DMAs were created for versus old I/O
with lots of interrupts." And David does talk about the "front processor"
option, which I think we've moved towards somewhat with lots of peripheral
management conducted by independent processors.

Regardless, one of those books I'd like to see published is, "RFC Classics:
Timeless RFCs on Computer and Network Design." Wouldn't necessarily limit it
to network or RFCs, but there's a lot of good stuff out there that could be
compiled together in a nice book with some editorializing, context, and
where are we now type analysis.

_______________________
Regards, http://web.mit.edu/reagle/www/home.html
Joseph Reagle E0 D5 B2 05 B6 12 DA 65 BE 4D E3 C1 6A 66 25 4E
reagle@mit.edu independent research account