Re: FoRK This Action!! (P2P's Dark Side)

From: Stephen D. Williams (
Date: Sat Mar 03 2001 - 07:06:57 PST

Jeff Bone wrote:
> Adam Beberg speweth:
> > In other words, capitalism and P2P do not mix
> > because P2P is a socialist construct, just like open source.
> >
> BTW, the above, classic Beberg-style, is one of the most ridiculous leaps of
> analysis I've ever heard. Here're two counters.
> Assume a fully P2P file-sharing facility. No privileged "servers." No way for
> anyone to make money, right? Wrong. I've got 50GB of files I'd like to store,
> but only 5GB of storage. You --- an individual, or whatever --- happen to have
> 500GB of storage. You're willing to let me use some of your space for some
> $amount / time period. Voila! Capitalism, facilitated by p2p technology.
> Similarly with open source. I've got this free software I'll give you, say a
> display server. What's that, it doesn't support your video card? You can fix
> that yourself, or I'll be happy to do it for you with quality guarantees and
> support for $amount. Voila! Capitalism. Facilitated by p2p technology.
> Folks, p2p is NOT a market, it's NOT a business model, it's a *freakin'*
> communication pattern, just like push was. It's not even a technology in its
> own right: it's merely an enabler. It's just a generalization of the
> client-server pattern; where in client-server only one side can initiate
> requests and the other respond to them, in p2p any component can do either.
> I'm so sick of hearing about p2p I could puke. Folks! No! Let's not turn p2p
> into a bad word!

Note: There are many different motive models at work in Free
Software/Open Source and even though I mention a couple below, there are
others I believe exist also. For instance, 'Guerilla Marketing' is a
perfectly valid reason to participate even if you don't buy the more
philosophical reasons.

I agree with all you said above completely. Equating Open Source/Free
Software and P2P to socialism is very shallow thinking. In the former
case, it's really a new kind of capitalism. I view it as the modus
operandi of scientists applied to software, at least in the greater
sense. Scientists build directly on the work of others, both recent and
past, and then publish and share their results and achievements so other
can build on them and for recognition to keep themselves funded. This
is what makes them useful. Building on and sharing source is exactly
like science publishing.

Another interpretation is that Free Software is a perfect implementation
of 'Pay it Forward', a principal I recognized long ago as an early
teenager when people helped me and I couldn't help them back at the

In other words, traditional capitalism can be seen as:

need/desire->hoarding->contribute for pay(work)->hoarding->pay for
access->get access

while Free Software is:

Both the following on different occasions:
need/desire->given solution (usually)->owe for solution->contribute when
possible to repay

need/desire->no solution available->
use existing building blocks plus effort to create new solution->
give back to group to get support, extensions, etc.

The main reason is probably more pragmatic:

need/desire->no solution available->
Not able or not worth building a complete solution from scratch,
Commercial building blocks are too expensive, restrictive, uncertain
survival, not hackable, poorly designed or implemented,
Use Free Software/Open Source because it is better in some way (even
just having source can make a mediocre package better than a commercial
Having agreed to participate in Open Source/GNU as payment for use of
building blocks, give back results to be built on some more.

This pattern is driven by actual need for software that does something
rather than the desire to produce a product that you can get people to
spend money for. I should point out that this is a very compatible
model for service business which is where we've been purportedly been
heading anyway. AOL for instance does quite well by not selling any of
their own software. I contributed to free software, at least in minor
ways, when I was working there.

I will point out that P2P does have a model that lends itself to less
centralization and control and is therefore somewhat beneficial to new

> Get over it, Beberg.
> jb
> PS - while we're at it, HTTP needs *fewer* methods, not more; it's universal
> RPC; you can do anything you want with any synchronous request-response
> protocol; and WebDAV sucks. So there. ;-)

Synchronous request-response is suboptimal for many existing and
emerging uses. In the long run, conversational async message based
protocols are the way to go. SOAP acknowledges this in the spec but
doesn't really achieve it with it's only binding of HTTP. I much prefer
Blocks/BXXP ("beep") or a future version of Jabber's protocol.


Stephen D. Williams         Insta, Inc./Jabber.Com, Inc./CCI
43392 Wayside Cir,Ashburn,VA 20147-4622 703-724-0118W 703-995-0407Fax 

This archive was generated by hypermail 2b29 : Fri Apr 27 2001 - 23:13:22 PDT