What is P2P?
- I must be an idiot savant. (Hold the flames please.)I just wrote this paragraph on Scripting News:"When we're finished with the corner-turn, Radio will be an easy to use writing tool that works on or off the Internet, with a strong bias to being on the Internet, and has an integrated Web server. (And is highly customizable.) Then I think Radio will be where Manila was approx one year ago. Ready for feature-refinement, a simple-enough base for growth, and thankfully few of the scaling issues that come with operating thousands of free sites. There's a very pragmatic reason for P2P my friends. It's nice to use the CPUs on the users' machines to render content. This goes to the question that Doc asked this morning. Doc man, you should interview Clay and ask him about the dark matter at the edges of the universe (oops I meant Internet). First we give the power to the people (the PC revolution) then we centralize it (the Internet) then we give the power to the people (P2P) but this time we give them networking too. A conceptual hula hoop. Round and round and round."So, ipso facto, P2P stands for Power 2 People.DavePS: Doc asked why software doesn't follow Moore's Law. I argued that it does. However ignorance is a strong opposing force.
- --- In email@example.com, "Dave Winer" <dave@u...> wrote:
>Right On. Couldn't have said it better myself.
> So, ipso facto, P2P stands for Power 2 People.
(Check out http://www.workingclasshero.com/discography/powertot.html)
> ...few of the scaling issues that come with operating thousandsAs persuasive a case for P2P as I've ever heard. It's good to see real
> of free sites. There's a very pragmatic reason for P2P my
> friends. It's nice to use the CPUs on the users' machines to
> render content.
usecases emerging. I can't tell you how many VCs I pitched who said "That's
nice, but what do you actually do with P2P?"
- Yup, it's something an engineer would understand but most VCs don't have
much of an engineering background. The net benefit to users is performance.
A funny thing. Over the last year we've optimized our core software a lot,
but I wasn't running it very much on my desktop so it always seemed to run
at about the same rate. In the last few weeks I've been running my own site
on my desktop computer and am blown away by the relative performance
compared to running it on the server, where I share the resource with
thousands of users. On my desktop I share it with no one, it's mine, and
At the same time Doc is asking the question why Moore's Law applies to
hardware and not software. It probably feels that way because many of us are
using a clogged network and overworked servers to do things we used to do on
our PCs. As we swing back to the PC, expect users to be saying "This thing
is fast!" That's something anyone who has upgraded their PC in the last few
years can appreciate.
----- Original Message -----
From: "Lucas Gonze" <lucas@...>
To: "Decentralization@..." <firstname.lastname@example.org>
Sent: Tuesday, November 28, 2000 2:01 PM
Subject: RE: [decentralization] What is P2P?
> > ...few of the scaling issues that come with operating thousands
> > of free sites. There's a very pragmatic reason for P2P my
> > friends. It's nice to use the CPUs on the users' machines to
> > render content.
> As persuasive a case for P2P as I've ever heard. It's good to see real
> usecases emerging. I can't tell you how many VCs I pitched who said
> nice, but what do you actually do with P2P?"
> - Lucas
> To unsubscribe from this group, send an email to:
- Lucas Gonze: Tuesday, November 28, 2000
> As persuasive a case for P2P as I've ever heard. It's good to seeThe best use of distributed processing I can think of is to crunch
> real usecases emerging. I can't tell you how many VCs I pitched who
> said "That's nice, but what do you actually do with P2P?"
> - Lucas
massive XML structures. That would let you put all your data in
XML files, and all your automation or processing needs in other
files, XSLT transformations, for example, with Xpointers and Xlinks
and indexes everyway from sunday.
What if there was a vast, infinite storage space, just a whitespace
and you could put unlimited content out there. What if everybody
on the network was required to use 1024 bit encryption because it
was built into the clients on the network and therefore, even dumbshits
encrypted 100% of their stuff. That's the way Windows and Linux should
have been, but were not, designed. They were designed for top-down
The client for this network? a pure w3c XML parser and XSLT
transformation engine. Is there any kind of content or processing
that can't be done or rendered, etc. by an XML processor? And, when
you go to read your stuff or process with it, you draw upon all the
CPUs that are sitting idle nearby...
That would be a computing environment worth having because it would
take back control of security from sysadmins and affix it firmly in
the leaf nodes. heh heh! Giving access control over the data
to the individual, permanently and irreversably.
The data storage could be same as today, either peer to peer or on
some ISP or ASP or even a relational database. Why not? Finally,
a good use for an RDBMS. As soon as the peer network has a general
ledger built into it, it would start to be adopted. Right now, there
is noplace to put your quarter to pay for your resources. Isn't
that ridiculous? Surely, this won't continue much longer.
- -----Original Message-----
From: Todd Boyle [mailto:tboyle@...]
Sent: Monday, December 11, 2000 5:41 PM
Subject: [decentralization] XML parser in a forest of files
"Is there any kind of content or processing
that can't be done or rendered, etc. by an XML processor?"
Can't? No. Can't reasonably/efficiently/feasibly/cost-effectively?
Absolutely yes, lots of them -- e.g., in defense, industrial
automation, telecommunications, ...
E. Douglas Jensen (traveling)
Home voice 508-653-5653, home fax 508-653-3342
Cell phone: voice 508-728-0809; text http://www.bam.com/send2.htm
Pager 888-916-9802 and http://www.skytel.com/paging (PIN 8889169802)
Office voice 781-271-2514, office Fax 781-271-4686