|
<![CDATA[
To: info@cinox.demon.co.uk
Hi Tim,
Sorry to be so long replying. Bit of a late one last night.
>Thanx for your replies to my message. Glad you and Frank liked it. I agree it could
>be two orders of magnitude but was being a bit reserved in case you thought I was
>mad!
This gave me a laugh. You're not mad, it's probably three.
>I cannot find any references either to software PADS, but I don't think it is that
>important to the neural nature of your project. PADS were more an idea of freeing up
>and drastically cheapening software development by extending accessibility and
making
>it all leggo like and therefore cheap and transferable.
Your mention of leggo is very interesting. Each structure (neuron axon, store, etc)
in the assembler version contains full information about itself (except RAM location
(handled by the LUT/XLUT)) and its i/o links to axons and stores.
This means that all the structures for a system can be jumbled about anywhere in
RAM, it
doesn't matter where each is physically on the machine. What I am saying is that the
whole lot is effectively a self-assembling leggo kit.
The concept of pluggability also relates to webfun, websub and webneuron libraries.
There are
some pictures about these on the pictures webpage. At present I think webfuns and
websubs
will overcomplicate, so maybe we should just stick to a simple network with
pluggable
webneurons only.
As far as I can tell, Frank is also investigating the concept of having extremely simple
"pluggable" webneurons right down to the operator level, e.g. the ">" operator would
itself be a webneuron! Thus we eliminate programming script altogether and the whole
network is self-similar from top to bottom. I will try and include all our latest
Emails on the website so you can have a look through. You may have already had a
look
at all the earlier ones via the "Email thread" at
http://brain.eu.org/629.htm
>The important bit of these papers that has direct relevance to web neurons is in the
>wired article towards the end where it says.
>"I asked Miller if the Internet was accomplishing his dreams for hypertext. "What the
>Web is doing is easy," Miller answered. He pointed out that the Web still lacks
>nearly every one of the advanced features he and his colleagues were trying to
>realise. There is no transclusion.
Can't find "transclusion" in the dictionary?
I believe all these advanced features can happen if HTML was a complete
programming language
rather than just a publishing language (http://brain.eu.org/564.htm - original posting).
>There is no way to create links inside other
>writers' documents.
Excellent stuff. Other writers documents are just webneurons so the links can be
created
by our (not built yet) webneuron editor. This could easily be done remotely. Security
has to be considered. I am very interested to pursue your "killer app" Webmail, but
take it a step further. I believe we can make a CGI Email webpage editor. I want to do
this also so others could edit the Website remotely, and put in links as Miller suggests.
This will save me time! You could have your own section on the site and administer it
yourself. We could alternatively do it using HTML forms.
>There is no way to follow all the references to a specific
>document.
This is because back links are seldom put in. The "specific document" should have a
list
of all its callers. If you notice on the Website, every link has a reverse link so you
can find your way round. This is compulsory in webneuron networks which are
designed to
be mappable starting from any point.
>Most importantly, the World Wide Web is no friend to logic. Rather, it
>permits infinite redundancy and encourages maximum confusion. With Xanadu - that
is,
>with transclusion and freedom to link - users would have had a consistent, easily
>navigable forum for universal debate."
Yes, I think there is a similarity here, but they don't ssem to have spotted
how easy it is to achieve. Really, almost the whole answer lies in one page i.e.
"Doing it in 32 bit assembler" or for the HTML version the "Original posting" page.
>"Still, Miller conceded that the Web's existence means that it must be accepted as
>the basis for a better form of hypertext. "We've got to use all of our technical
>insights to migrate the Web to a higher plane," he said.
The Web may really want deleting and start again though. The webneuron concept
goes right
down to how you build microprocessors and right up to the brain itself.
Yet I know we need a quick entry point. We cannot do it all at once.
>Miller is pondering how to
>allow readers to add links to other writers' Web pages without copying the original
>documents."
Use the CGI Email "quickie" method. Each server must be set up for it though. We
can
start on the webneuron site.
>In relation to your idea of using the Virtual Glastonbury website I will ask the
>sponsor what he thinks about the idea. Although we may have to run that bit off a
>different server or move the whole site. It is using a virtual server at the moment.
The webneuron site is also virtual. My ISP is not being 100% co-operative. I think
they want more money.
>In relation to the Nerdathon, personally, I am on for it.
Great news. I am updating the site today.
>Where do you guys hang out BTW?
Frank is in Holland, but I am in Northampton. I could certainly meet you up for
a beer maybe halfway somewhere.
Excellent. Going to your other mails now.
]]>
|
John's Emails
