Search the web
Sign In
New User? Sign Up
decentralization · Implications of the end-to-end principle
? Already a member? Sign in to Yahoo!

Yahoo! Groups Tips

Did you know...
Want your group to be featured on the Yahoo! Groups website? Add a group photo to Flickr.

Best of Y! Groups

   Check them out and nominate your group.
Having problems with message search? Fill out this form to ensure your group is one of the first to be migrated to the new message search system.

Messages

  Messages Help
Advanced
p2p working group/standards   Message List  
Reply | Forward Message #1343 of 7047 |
Re: p2p working group/standards

> I've never been able to figure out what the heck Intel is thinking.
> They make chips, not software, formats, standards or working

Well, chips are just software that has been compiled into silicon,
and strangely, when you comile your code into silicon, you don't have
the same crowd yelling at you to GPL it. If you slow down innovation
by making the code release cycle depend on chip fabs, and justify the
expense by saying "well, you can touch it can't you?", then somehow
people think you're not ripping them off. I mean, we software guys
rip everyone off because we "live under the false assumption that
software is manufacturing", but to artificially add manufacturing to
the process by burning code in the chips is just fine... (Real
efficient, too). Intel is getting into switches and routers now,
which are the classic software-that-became-hardware. There is
another nice thing about compiling your software into silicon, if you
sell to large corporate interests -- hardware is a physical thing
(you can touch it, can't you?) so you can depreciate it, and
corporate budgets for physical assets are usually much more liberal.

As for their involvement in P2P, I think I can guess.. Remember the
outrage about five years ago when Intel put the unique identifier on
every chip with instruction codes that could read it? At that time,
they claimed it was being put there to make Internet transactions
more secure. Everyone else thought it was "big brother", and the
whole thing was scuppered. But believe it or not, the whole thing
*did* have some merits, and I was sorely disappointed to see it go.
I guess we're OK without it, but ... enter y2k+1, and P2P is the
rage. Intel sees potential for security problems, and realizes that
the only way to prevent man-in-the-middle attacks is with cooperation
from the CPU (same as their claim in the last controversy, which was
equally valid). So ka-bam! The chip_id scheme is resurrected, just
in a different skin. I bet they figured out how to make a
chip "trusted" this time without the same old privacy risks, but
that's just speculation.. In any case, Intel is very much interested
in digital media - video, audio, etc. as the next phase of PC
evolution, and they know that their boxes will need to have the
ability to protect content-owners from piracy, or else Sony will just
ship a playstation/stereo/PC all in one box with the digital media
protection built-in. They are pushing standards for digital media
protection (and remember that P2P ala Napster is a *perfect*
distribution mechanism for digital media, with a few changes -- I
mean, you get edge-caching based on hot items built-in for free...).
They are trying to be careful to not go too far, as the CPRM case
shows. (see: http://www.theregister.co.uk/content/2/15718.html) So
I think is not so much about P2P as it is about the PC of the future,
and if Intel doesn't take leadership and figure out how to make the
privacy/digital-rights thing work (that's pretty much what they've
limited their code on sourceforge to so far), then they will be
eclipsed by the "real" hardware vendors like Sony/Matsushita, who
have utterly terrible track records on standards since their entire
devices are proprietary.




Tue Feb 13, 2001 5:56 am

allenjs@...
Send Email Send Email

Forward
Message #1343 of 7047 |
Expand Messages Author Sort by Date

... I know of BXXP, but I'm not sure if that's what you mean. BXXP's design was built around a sort of database to web app framework if I remember correctly ...
Dan Moniz
dnm@...
Send Email
Feb 13, 2001
7:29 am

... Well, chips are just software that has been compiled into silicon, and strangely, when you comile your code into silicon, you don't have the same crowd...
allenjs@...
Send Email
Feb 13, 2001
5:56 am

... The most useful thing to standardise is the terminology. This list has gone some of the way of developing a useful vocabulary (clouds, etc), and some ...
Nick Lothian
nl@...
Send Email
Feb 13, 2001
5:10 am

... question. OK, you're right of course. My own believe is that IETF and W3C are already working on this stuff. If I want to work with a standard body for...
Dave Winer
dave@...
Send Email
Feb 13, 2001
5:29 am

... Like Lucas said, service discovery (or directory?) is a big one that various P2P systems could benefit by sharing. Another is a global public key...
allenjs@...
Send Email
Feb 13, 2001
6:15 am

... A useful thing would be a practical NAT standard for NAT (box or software) access/control from the peer/client applications. To make these capabilities...
egroup@...
Send Email
Feb 13, 2001
8:36 am

I think to some extent you are correct. I'm less in touch with IETF, but I do follow the W3C pretty closely and am currently on the XML protocol working...
Jeffrey Kay
jkay@...
Send Email
Feb 13, 2001
2:59 pm

... We are right about empowerment, and they are right about subversion, because ___1___ Empowerment = Control ... While there has been and will continue to...
Clay Shirky
clay@...
Send Email
Feb 13, 2001
7:56 pm

... There are two reasonable approaches I know of for resolving the XML performance problem: 1. WBXML tokenized compression Anybody know of a library to do...
Tony Kimball
alk@...
Send Email
Feb 13, 2001
5:40 pm

... The lookup service itself has to be decentralized. There's no point in locking in to a centralized directory. So: ability to find a decentralized service...
Lucas Gonze
lucas@...
Send Email
Feb 13, 2001
5:54 pm

... But Lucas, there's no point in _not_ locking into a centralized directory, if that's helpful. Decentralization is better as a tool than a goal, and as...
Clay Shirky
clay@...
Send Email
Feb 13, 2001
7:22 pm

Are you seriously suggesting that decentralized lookups are only for static resources, Clay? Think of a CPU sharing app that sends out a message saying "I need...
Lucas Gonze
lucas@...
Send Email
Feb 13, 2001
7:45 pm

... No, what I'm suggesting is much much vaguer than that. <g/> What I'm really suggesting is that the impulse to decentralize every bit of an application is...
Clay Shirky
clay@...
Send Email
Feb 13, 2001
8:04 pm

... I definately agree with you on that principle. I'll be talking about the numerical inefficiency of decentralized designs (well, by implication) at ...
Lucas Gonze
lgonze@...
Send Email
Feb 13, 2001
8:48 pm

... I don't get the impression that that's what he was suggesting at all. It appears that the discussion is becoming centralized peer discovery vs. completely...
Chris Cummer
cjclists@...
Send Email
Feb 13, 2001
8:04 pm

... I don't mean to make a blanket statement against centralized directories, but rather to say that AFAIK there are no decentralized service directories. -...
Lucas Gonze
lgonze@...
Send Email
Feb 13, 2001
8:58 pm

... Jini....too bad its not open source though. E-speak may also have some interesting stuff in this regard and it is open source. P2P infrastructure must be...
Justin Chapweske
orasis@...
Send Email
Feb 13, 2001
10:35 pm

... Yep. -- Damien Stolarz Chief Technical Officer Static Online, Inc. dstolarz@... (818) 968-7626 http://www.static.com...
Damien Stolarz
damien@...
Send Email
Feb 13, 2001
11:21 pm

... some ... I would say that P2P infrastructure must be open spec, from which an open source implementation can flow. I don't begrudge those who want to build...
wesf@...
Send Email
Feb 20, 2001
7:29 am

... I'm referring mostly to avoiding having P2P systems rely on closed libraries such as .NET or Jini. Dave has made his case time and again about how his...
Justin Chapweske
orasis@...
Send Email
Feb 20, 2001
8:05 am

In article <Pine.NEB.4.31.0102131549300.14391-100000@...>, Lucas Gonze <lgonze@...> writes ... There are shades of grey here. Arguably DNS...
Julian Bond
julian@...
Send Email
Feb 14, 2001
10:51 am

... For any generic service I believe this devolves to gnutella-style broadcast searches. To make this more efficient some sort of hierarchy would need to be...
Justin Chapweske
orasis@...
Send Email
Feb 13, 2001
10:32 pm

... A bunch of us at U.C.Berkeley and ACIRI (AT&T Center for Internet Research) have recently been working on designing a highly scalable, distributed indexing...
Sylvia Ratnasamy
sylvia@...
Send Email
Feb 14, 2001
3:31 am

... I just had a chance to read through this, and it looks like an interesting addition to the pool of options. Some comments: You mention the fact that...
allenjs@...
Send Email
Feb 15, 2001
11:09 am

Thanks for the comments. ... I totally agree.. i think our point was not that Napster does not scale but rather that beyond a point it is expensive to scale a...
Sylvia Ratnasamy
sylvia@...
Send Email
Feb 15, 2001
10:46 pm

Hello to everyone asking about the working group! The de-facto standard list for *technical* p2p discussions is of course: ...
Damien Stolarz
dstolarz@...
Send Email
Feb 16, 2001
3:27 am

... The following night help: http://www.research.att.com/sw/tools/xmill/ XMill groups XML text strings with respect to their meaning and exploits similarities...
Jacobsen
jacobsen@...
Send Email
Feb 13, 2001
6:05 pm

Hello, Clay -- ... Excellent formula. I think that your point indicating that some amount of control is lost with empowerment and vice versa is correct, but...
Jeffrey Kay
jkay@...
Send Email
Feb 13, 2001
8:51 pm

In article <45F51952AE8AD41180B3009027E5AF6E1AAFA8@ENGENIA1>, Jeffrey Kay <jkay@...> writes ... Maybe there's another lesson to be had from Napster....
Julian Bond
julian@...
Send Email
Feb 14, 2001
1:18 pm

I think P2P infrastructure, if there ever is such a thing, should be open and documented with lots of choice. No one religion, no one right way to do it, no...
Dave Winer
dave@...
Send Email
Feb 14, 2001
12:00 am
 First  |  |  Last 
Advanced

Copyright © 2009 Yahoo! Inc. All rights reserved.
Privacy Policy - Terms of Service - Guidelines - Help