Cart
[Home] [By Thread] [By Date] [Recent Entries]
Tim Bray <tbray@t...> writes:
> At 12:12 PM 4/24/00 -0400, Simon St.Laurent wrote:
>
> >It doesn't mention XML at all, but these are in many ways the models I'd
> >like to see for information exchange using XML, rather than the data
> >center/hub-spoke kinds of models that companies seem to be fixated on.
>
> All this is totally orthogonal to whether you use XML or not. The real
> point is this: what does it mean when every desktop is a web server as
> well as a web client? And [a problem that becomes obvious after a brief
> look at Gnutella] what degree of intermediation is necessary to make this
> scale? -T.
Further to Tim's point, Napster is a special case because there's an
extremely high degree of redundancy: even a moderately popular song
will show up on dozens or hundreds of computers at once. Since most
of the time many people are chasing relatively few (say, a few
thousand) songs, it doesn't much matter who you get the song from.
Note that there are still central servers to help you *find* the song
you want.
For general information exchange, this model would not work so well.
What if SAX were available only when I happened to have my notebook
plugged into my cable modem?
All the best,
David
--
David Megginson david@m...
http://www.megginson.com/
***************************************************************************
This is xml-dev, the mailing list for XML developers.
To unsubscribe, mailto:majordomo@x...&BODY=unsubscribe%20xml-dev
List archives are available at http://xml.org/archives/xml-dev/
***************************************************************************
|
