begin  quoting James G. Sack (jim) as of Mon, May 07, 2007 at 05:36:34PM -0700:
> Yesterday's slashdot had a link to A Van Jacobson talk:
>  "A New Way to look at Networking"
>  http://video.google.com/videoplay?docid=-6972678839686672840&hl=en

Any transcripts?

> Warning: starts to play w/o asking [I wish it didn't do that!].
> 
> There are probably many quoteworthy remarks, but I liked this one:
> 
>   "Integrity and trust are properties of the data, not of the way that
> you obtain it."

Um, not so much. 

See the discussion about the color of bits (on -kooler?) not too long ago.
 
> The general gist was that the internet (eg, TCP/IP) gave a new viewpoint
> of a "side effect" in telephony networking (namely, conversations), and
> now it's time to consider a new viewpoint of a new side effect --
> related to the dissemination of data.
 
Conversations were a side effect of telephony neworking? I thought
they were the *reason*.

> In each case, the old world was stuck within a viewpoint that did not
> address the new problem. TCP/IP made explicit the concept of endpoint
> addresses, and discovered a new  definition of "network". Now we need to
> make explicit the identification (&cetera) of data.

TCP/IP gives circuit-switched semantics to a packet-switched network.

I thought endpoint addresses predated TCP/IP.
 
> Another way of saying it is that a new solution to the old problem is a
> waste; what's needed is recognition that the problem has changed, and
> investment in new viewpoints.
> 
> ..well something like that.

Sounds like marketing hype.

-- 
Beware announcements of paradigm shifts -- or calls for such.
Stewart Stremler


-- 
[email protected]
http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-list

Reply via email to