RE: [xml-dev] Linking global context

["Bullard, Claude L (Len)" <clbullar@ingr.com>]

Hi Didier:

 

I was up at 3AM today to do a TV show at 5AM.  IOW, here at about

4PM, not enough neurons are still linking to do a reply justice to

your topics.  If this wanders more than usual, I plead exhaustion.

 

Does the linking process transform the element, or does the act

of navigating the associations transform the navigator (the

intelligent entity, the wetware)?   Whereas both are possible,

and while post-modernists say one cannot truly separate

the observer and observed, I still think it critical to do that.

Otherwise, transcendence of the navigator doesn't happen

and I believe it does.  Call that the faith position that if we

care to use it, free will does exist but that it is as in many

transactions, of value only when traded for consensus.

 

A language need not contain linking capabilities to be linked

to.  That is why independent links (aka, out of line) and the

separation of link and locator are fundamental now and were

part of Hytime.   Yes, in the sense that a standard system

for resolving links enables global sharing, it need not be

the same linking syntax as long as it meets the same

architecture.  This is also fundamental and a reason for

architectural forms.  That said, it is a lot more cost effective

to share the syntax so one might say this is a useful thing

to do.    But we should not confuse the system with the

knowledge, the container with the contained although
I believe it to be true that the evolution of these are linked

through use.   We may want to examine the nature of

this coupling and determine if the effects real or

imaginary, and how they scale across populations.

 

Anecdotal evidence may suggest for example, that

participation in lists such as these is raising the

knowledge values for both individuals and in groups,

and that the feedback by these entities into the

design of the system creates a power pattern or

amplification of the evolution.  Some see that as

the root cause of the Internet Time phenomenon,

but I don't see this as particular to time, or that

the link there is direct.

 

Intelligent linking suggests to me the emergence

of selectors over links and types of links.   As noted

elsewhere, there is a "subtle" preference to link to

clusters or attractors, but that is a weak description

of selection.   We must beware of superstitious

linking, or as I said elsewhere, the consensus

of the uniformed that emerges from the network

effect and actually retards the advancement of

the technology or the intelligent entity.   Devolution

may be as likely as evolution so I resist Panglossian

descriptions of the power of the web although I

am attracted to them.  Again, awareness of desire

is the key to transcendent potentials.  In the works

of Frost, "the road not taken" can make all the

difference.

 

We have to be aware of the localization of intelligence

and competence in the forms of network subcultures.

Think of a model of a n-dimensional space of topics

of attention in which only some subset of the available

topics can be attended, and the rest ignored.  What

we attend; we know.  What we don't attend may not

exist as far as development potential is concerned unless

there are hidden couplers.

 

I essentially agree with your positions and also believe

linking capability to be key.   I view the Xlink debate as

a symptom of localization in which a few members understand

n-way linking, understand why links and locators are

separate concerns, and assert that as the topic is discussed

more here, others will see that this is a convenient and

powerful way to describe many existing artifacts using a

different model; that is, first one should be able to essentially

swap functions for links and vice versa.   One should be

able to view a web as a recursive model.   One should begin

to understand the spatio-temporal issues of link emergence

so that link persistence shares many qualities of real time

systems with late binding, memory properties such that

topics can be discovered and forgotten, and so on.

 

That's enough for today.   I need to go.   Happy Labor

Day holiday to all of you in the US.   Happy weekend

to the rest.  May you be smarter and faster this time next

week, and richer by whatever measures you have for wealth.

 

len

-----Original Message-----
From: Didier PH Martin [mailto:martind@netfolder.com]
Sent: Thursday, August 29, 2002 1:14 PM
To: Bullard, Claude L (Len)
Cc: xml-dev@lists.xml.org
Subject: RE: [xml-dev] Linking global context

Hi Len

 

With the web we discovered that the linking process transform each element by integrating them into a whole: the World Wide Web. The integration process is through links. Through links we can navigate through thoughts associations.

 

With XML we can create our own domain language. If that language do not include linking capabilities, the document stands alone without any opportunity to relate to or be linked to another document. If the domain language includes linking capabilities, it is then possible to create a whole greater than the parts. If each domain language has its own way to express links, then, a navigator will have to be aware of all the different idiosyncratic ways to link. If all these domain languages share the same linking syntax then the linking process can be modularized and re-used. If the same linking syntax is used, then we assist to a clustering mechanism and itself the linking syntax creates a new network effect governed by the "power distribution" pattern (in opposition to a normal distribution - in the mathematical sense a power distribution is based on attractors and is not normally distributed).

 

I think that having a consensus on certain layers would tremendously leverage human knowledge. For instance having the same syntax helped reduce the knowledge load by re-using the same rules (i.e. by using XML). So far so good. But staying at the syntax level is only a minor first step. As we already know, several languages share the same alphabet but it does help the people to understand each other since the languages are different. If we add a standard linking layer on top of that we put in place the basic tool used by intelligent systems: linking. A neuron is nothing useful per se but linked to others it becomes the source of thoughts and thoughts the source of actions.

 

I think that the recent debate about xlink and the XHTML workgroup conclusions lost sight about a more global perspective and a potential opportunity for mankind if a common linking specification reach social acceptance. Instead of talking of semantic web, we should focus on the basic fabrics of the web and the basic principle underlying intelligent complex systems: linking.

 

You know, at first I thought that XML was too verbose and that there are more efficient way to structure information. Then, I accepted more work (more thing to do and write) in order to gain from the new standard. I think that the argument about the xlink namespace notation is not really a show stopper since I myself learned XML and a more verbose way to structure information in order to gain something else. People would probably invest in xlink if they gain something more from it.

 

The linking debate is more important that we may envision on first sight. And I think that, as always, Byzantine fights leads to some myopic view of the world.

 

"A different point of view is worth a thousand point of IQ". I do not know who said that at first but I know that reality is not a single facetted world and that the more we can perceive it through different perspective, the more it reveals itself. There is a whole universe in a grain of sand. And when this grain of sand is linked to the other through gravity of other attraction forces it creates beaches or sand sculpture or.... Semi-conductors and who knows what else it could be.

 

Cheers

Didier PH Martin

 

-----Original Message-----
From: Bullard, Claude L (Len) [mailto:clbullar@ingr.com]
Sent: 29 août, 2002 10:57
To: 'Didier PH Martin'
Cc: 'xml-dev@lists.xml.org'
Subject: RE: [xml-dev] Linking global context

 

Looks like a good read.  The blurbs hype it as revolutionary, and it is not.

 

Any erudite Hindu can explain all of this to one.   But since the

dominant metaphor these days is the web, explaining things in

terms of webs and networks makes sense as a pop communique.

So do tantra and karma.   A lot of us have been using ecosystem

metaphors for some time, and that is a somewhat better metaphor in that

it takes up communities in a less flat metaphor:  hubs (attractors)

might not be organized as a set of connections, but in terms of

the sign systems (say message types) they share, so this

becomes the self-organizing principle and one has to think

about how and in what ways sign systems and power laws

are connected (it isn't hard).

 

Some questions:  how does any member unattached to a

network/community choose its community, and by what

means does it signal such intentions?  

 

What is the nature of habit, how are habits acquired,

and are we capable of transcending habit?  Is the trancendence

of habit the key to transcending the network itself?

 

Desire is the maker.  Desire is the destroyer.  How do we

become the masters of desire?   Networks are a very flat

way to look at a complex adaptive system.   Given an origin

event, how long does it take and at what frequency of

interaction for intelligence to emerge?   What events can

be posited that will amplify this emergence or curb it?

What events destroy intelligence (capable of manipulating

and influencing events) and how are they detected?

 

One can resort to pattern and link analysis, but they

are wholely unintuitive.  At the operational scales,

intuitive knowledge is vital for selecting actions.

 

len

From: Didier PH Martin [mailto:martind@netfolder.com]

Hi

 

In the context of linking, does anybody has read "Linked - the New Science of Networks" by Alberto-Laszlo Barabasi?

 

I would be interested - in the perspective of linking - to listen to your thoughts about these concepts.