Re: Huge drivespace Re: computing budgeting (fwd)

Date view Thread view Subject view Author view

From: Jeff Bone (
Date: Thu Sep 07 2000 - 07:28:42 PDT

Eugene Leitl wrote:

> Jeff Bone writes:
> > That's a great question. Discontinuities like that present huge
> > opportunities. IME, though, the "apps" usually lag significantly because
> > people just can't figure out how to do anything but more-of-the-same -w-
> > discontinuous tech, often for years after it appears. A VC buddy of mine asked
> I have not noticed that the Net had a lot of bandwidth slack at any
> time in the past.

That's true, but never before have we seen the kind of discontinuity that
fiber-to-the-home is about to give us. DSL has only been widely available for, say, a
couple of years now; in a 2-4 year period, we're going to increase effective
consumer-priced bandwidth by 3x as much as it increased from the period circa
1980-1998. That's huge. :-)

> By the time you will have a GBps to your house, you will complain that

This makes sense from a historical perspective, but you really need to ask yourself:
what are we going to do *next year* with Gb/s? There just isn't enough content, nor
automation, to saturate that. Most of its utilization is going to be passive,
carrying 9 million more pay-per-view channels.

> If I had a GBps to my house right now, I could afford to host magnet
> content at home without fear of being /.'d, I would run a large number
> of community services, irc, freenet, remailers and IP onion routers,
> broadcast several multimedia streams,

So you're going to run a commercial-grade datacenter at home? Linux + Apache + a
single PII can saturate a T1. The reason most people don't run magnet content at home
isn't because they don't have fatter pipe, it's because they don't have the server

> participate in real VR,


> watch TV
> and run distributed thingies, like ALife simulations and participate
> in large-scale molecular simulations,

Again, you don't need Gb for that.

> and running distributed spiking
> codes, pooling mine and my friend's machines. Believe me, if I have
> enough CPU I can fill up even a TBps pipe easily.

I dunno if you can fill a Gb/s easily, much less a TBps. (Who mentioned those? ;-)

> Btw, by the time I can have 1 GBps at least to my 100 next neighbours,

Later this year, consumer priced, in (oddly the low-rent parts of) Austin and

> and they're willing to let me have their spare cycles for free, and
> this holds globally (so that I can net ~1 million contributors,
> provided the project is interesting enough) the number of GFlops (and
> especially MMX type multimedia OPS) in your average desktop will be
> extremely formidable.

> Foreseeable future is how much? 2-3 years?

Think about the players: we're talking media / cable folks. It'll take longer. ;-)
I'd say end-of-the-decade starts to look more interesting, even though the pipes will
become available end of this year or next. Not to mention that all this new-fangled
shit wrecks havoc with their business models, assumptions, etc. The friction in
getting change to happen is most often business tension, not technical. (Look at
Napster, etc.) (Thought experiment: assume your a cableco. You give every
subscriber, essentially, a Tivo, stationed upstream in your offices. What does that
do to you from a business perspective? It creates tons of biz problems.)

> With email, you're bottlenecked by your eyeballs. You can't need more
> than a few GBytes, unless you only read the subject: lines ;)

I've been caught! ;-)


Date view Thread view Subject view Author view

This archive was generated by hypermail 2b29 : Thu Sep 07 2000 - 07:38:50 PDT