More information about the Underscore mailing list

[_] BRIC / BCTN / Broadband

Tom Gidden gid at litebase.com
Wed Oct 24 17:32:33 BST 2001

> Most of you are probably aware of these bristol projects. more info:
> http://www.bctn.net To continue to be useful, we need to have more
> tangeable services and products.
>
> Now I've done a fair amount of reading and thinking about this but I'm yet
> to see any applications which aren't just souped up narrowband apps.
> Nothing wrong with that but are we missing something? Anyone care to take
> to the pulpit and give us some visionary statements?

Yep.. okay, I've got two.  Probably a bit too bluesky-ie for your
purposes, but....  Now listen for half of Underscore groaning
about this -- a lot of you have heard these two before!  Oh, and
expect a slur of some kind from a certain Bath resident =)  However, I'm
always willing to jump into the pulpit.

I don't want to give too much away in this forum right now, but here's a
hint or two.

Both ideas are my take on the 'work smarter, not harder' concept.  I
wholeheartedly believe that the web as currently implemented is incredibly
ill-suited and inefficient for the job.

1. Litebase.   I haven't yet seen a language, database or content
   management mechanism that really impresses me.  IMHO, they all have the
   wrong idea.  As a result, web development is at least a level of
   magnitude more complex than it needs to be.   Litebase is my vision of
   the ideal web development infrastructure.  It goes further, detailing a
   mechanism that can be used for all types of information application
   development.  It starts as a new language, unlike any other I know of.

   I've been working on it on-and-off for about three years, and am
   convinced it is, at the very least, feasible and useful.


2. Osmo.  I've been working on this even more on-and-off since 1996,
   based on some experimental work I did in 1991.  It's
   my all-encompassing replacement for the World Wide Web, and could
   develop further into a network operating system.  If taken to it's logical
   conclusion, it could replace operating systems and applications as we
   know them, and reduce IT costs drastically.  One of the core concepts
   is to get rid of all the redundant narrowband but noisy communication
   to free up the net for broadband content, including pre-emptive
   intelligent caching.

   I have a very good idea of how it would be implemented, but it's far
   too big for me to develop on my own.  The first step would be to build
   a prototype, based on a heterogenous network.

   The thing is, hard drive is much cheaper than asphalt and JCB diggers,
   so we should be using the widespread narrow infrastructure (local loop,
   etc) more effectively by utilising the cheaper, wider but shorter
   backbones in an effective way.

   If done correctly, as long as the material was popular, a very high
   traffic website could be published from a modem, with almost instant
   global update and widespread automatic caching.

   This is a lot more than a web cache system.  It's more than Freenet
   too.  The key to it is the restructuring of content to make it suitable
   for distribution.

   The web is all about bludgeoning redundant inefficient dumb content
   through a weak infrastructure with a one-to-one model to simulate a
   interactive broadcast medium.   Broadband's similar, but just widening
   the pipes and forcing even less efficient content through it.   The
   network is as weak as it's weakest point if this model is used.

   Once you've modelled content in a much smarter way, you can actually
   start spreading it around to where it needs to be before it needs to be
   there.  If you've got that, you've got a lower latency, which means you
   can start putting together proper applications rather than grotty web
   apps.  You also free up the existing pipes for rich broadband content
   at very low cost.

Sorry.. I'm rambling now.  It's late in the day =)

Anyone who wants to hear more, please contact me.  I'm happy to discuss
these ideas further on a one-to-one basis.


Tom
-- 
Tom Gidden
Litebase Solutions Ltd.
e: gid at litebase.com
t: 07976 690 578