Re: p2p working group/standards
> 1) What is needed (or wanted) by most p2p developers thatLike Lucas said, service discovery (or directory?) is a big one that
> would actually be worth standardizing?
various P2P systems could benefit by sharing. Another is a global
public key "identity" directory (sort of like a big LDAP); with
appropriate use of PK could guarantee anonymity where desired and
could also guarantee identity if desired. Both of these things BTW
are useful for a service-based web like .NET, Sun's ONE, and whatever
Oracle calls theirs are talking about.
> 2) Are there any "plumbing" or mundane tasks, that every appThe rest is gonna be hard. We are talking about consumers here.
> could benefit from?
> 3) are there any other things aspects in or surrounding p2p that
> should be standardized, interoperable or compatible?
Consumers happily go out and buy a Sony Digital Camcorder without
caring that the video capture board they used for their Panasonic
camcorder won't work, since the two companies implemented
differently. Businesses respond to FUD about standards ("You mean,
if I use XML without waiting for ebXML then my business might
spontaneously combust? I'm going to go pressure Microsoft to support
ebXML right now!!") Consumers might grumble, but they just want a
cool product. That's one conclusion I reached about Groove - it's a
very consumer-friendly interface. It's simple. And for developers,
they have a cool API that can be programmed from any programming
language, so why would someone care about the underlying protocols?
The only people that will want interop are the P2P vendors who don't
have large communities yet, so it will be the small vendors with
little power pushing the big vendors to be gracious (and when I
Groove's eye candy will be a very powerful force to beat I think).
So any standards would rely on the magnanimous nature of the dominant
players. This is just the nature of the consumer market; users don't
buy standards, they buy cool features. Standards aren't bad for the
consumer market, in fact it is a noble goal -- but I notice that at
least four of the really hot P2P platforms today (Groove, OpenCola,
Napster, XDegrees) all invented totally different approaches to
solving their problems, including different approaches to directory
and discovery -- which nearly everyone would agree should be
standard. So judging from that, I'm not hopeful at all. (I am sure
that the p2pwg will write some specs, and so will other groups, but
it will be an exercise in futility -- specs are not killer apps, and
the killer apps are already out of control).
- Justin Chapweske said:
> ... switch to SHA-1.Probably going to SHA-1 isn't too big of a problem. I'll bring it up with
> The biggest group that I havn't yet talked to about this is the Gnutella
> guys, but I'm sure they'd be into it as well. Any Gnutellians on the
those that I know. Interestingly, there are ways to add file hashes within
the existing protocol specifications - it should even be backwards