top bar
QuickTopic free message boards logo
Skip to Messages

TOPIC:

no hierarchy

^     All messages            8-23 of 23  1-7 >>
23
Steve Yost
04-03-2001
08:32 PM ET (US)
I'm on a mailing list where reporters can post questions asking for leads on articles they're doing. Here's one I thought might be of interest to this audience. Reply to the mentioned email address if you're interested and have something to contribute. You can refer to the profnet service as the source of the query.
----
**4. ROLE OF TAXONOMIES - KNOWLEDGE MANAGEMENT MAGAZINE. For an article about taxonomies, I'm looking for vendors of such solutions, users of them, analysts and professors who can talk about the role taxonomies play in the creation of knowledge management systems at businesses. Using taxonomies,
how does organizing knowledge effectively have an impact on areas such as building corporate portals, document management and collaborative environments? Need leads by Friday, April 6. >>> Marc Rapport <mrapport@sc.rr.com> Phone: 803-917-5918 Fax: 312-672-8254 [c::4/03:3231]
22
03-30-2001
02:00 AM ET (US)
21
Steve Yost
02-06-2001
09:50 PM ET (US)
Jack, more towards real document annotation is Quick Topic's new Document Review feature. Give it a try and let me know if it works for your purposes.

I've heard good things about At Home in the Universe -- might be a good candidate for my next book to read.
20
Jack Park
02-06-2001
07:14 PM ET (US)
What a concept! I recently downloaded a couple of open source Java Wiki implementations. I am seeing lots of value here. However, I am thinking that it might be useful to explore the possibility of making it into an annotation system rather than a page-edit system. Instead of opening up a page to edit, you, instead, click an "annotate me" button and get your own page (in a separate window) to edit. Just a thought.

BTW: I am presently reading Stuart Kauffman's At Home in the Universe. I think it to be appropriate to all of the thoughts discussed here and elsewhere along the same themes.
Also, BTW: I am clicking 'submit' again. QT just appears to have crashed while submitting this. It further turns out it's probably not QT, rather it's Verio, what appears to be an awfully popular, but unreliable ISP.
19
Jack Park
02-06-2001
07:11 PM ET (US)
What a concept! I recently downloaded a couple of open source Java Wiki implementations. I am seeing lots of value here. However, I am thinking that it might be useful to explore the possibility of making it into an annotation system rather than a page-edit system. Instead of opening up a page to edit, you, instead, click an "annotate me" button and get your own page (in a separate window) to edit. Just a thought.

BTW: I am presently reading Stuart Kauffman's At Home in the Universe. I think it to be appropriate to all of the thoughts discussed here and elsewhere along the same themes.
Also, BTW: I am clicking 'submit' again. QT just appears to have crashed while submitting this.
18
Steve Yost
02-05-2001
10:06 AM ET (US)
Glad to hear that, Bernard! We're using Wiki now within my company, and I even use it for my personal weblog (http://www.quicktopic.com/blurcircle).

A little offtopic, but relevant to some of your interests: I just read a fascinating book that might explain your radical change of mind (and radical changes in general). It's called Ubiquity, http://www.booksunlimited.co.uk/firstbook2...6194,385522,00.html. It's a popular introduction to self-organizing criticality. One major reference for the book is Per Bak's book http://public.logica.com/~stepneys/bib/nf/b/bak.htm ; at that site is an introduction to the idea.
17
Bernard Vatant
02-05-2001
05:32 AM ET (US)
I'd written last November some too much definitive things about was going on at Wiki.
Since then I somehow radically changed my mind about it, and get involved in this process where interesting things are emerging indeed.
Enter this non-hierarchical space for example at
http://www.voght.com/cgi-bin/pywiki?YellowPagesDiscussion
or http://www.voght.com/cgi-bin/pywiki?TopicMap
16
Bernard Vatant
12-22-2000
05:14 AM ET (US)
The reflexion seems to have been dormant here, but the ongoing debates among the Topic Maps "community" are a good real use case of collective intelligence building, and linked problems.
See http://www.infoloom.com/pipermail/topicmapmail/2000q4/thread.html
and http://www.egroups.com/group/xtm-wg
I wonder if we are not presently moving outside the convergence attractor -if any - or if a bifurcation is not taking place between various tendencies, ISO, W3C, Xlink ...
Jack has been very silent about all that.
Some comment, anybody ?
15
Steve Yost
11-12-2000
04:00 PM ET (US)
FYI: in the Nov 6 blog entry at http://www.peterme.com/, he mentions the "fantabulous 'Synonyms and Taxonomies' one-day seminar" (http://argus-acia.com/acia_event/seminar_roadshow.html). This sounds tremendously applicable: the idea of building a thesaurus as the way to map from one discipline to another.
14
Bernard Vatant
11-08-2000
05:17 AM ET (US)
Jack : Don't get me wrong. Ontologists sure are useful.
What I mean by the joke, which made you jump in, is that in that domain like others, we are to be vigilant about too much over-specialization, and not let them tell us all the what, why, and how about our own concepts. No more.
Use them. Don't abuse.

The initial formula is, I think :

"War is too serious a thing to be let to soldiers".
I don't remember the author. There are many variations on it.
13
Jack Park
11-07-2000
10:35 PM ET (US)
<Bernard>
Ontology is too serious a thing to let the ontologists
                       manage it.
</Bernard>
Gads. Talk about sound bytes!

It is interesting that most folks consider knowledge a social construct: that which we believe to be true is that which we agree (with nearly everybody else) to be true. And then, the philosophers jump in and analyze heck out of all that, and some of them branch off and become ontologists. And, as you say, they try to manage it. Well, somebody's got to. Say what?

I think that, in order to construct a collaboratory at which humans gather to somehow amplify their own intelligence, we are forced to, at least, pay attention to what the ontologists are saying. I have learnt a lot from them. Of course, I don't have to do anything they say. In my case, I am attempting to perform a grand synthesis, one that I hope will result in an engine (software system) that, itself, will have the capability to evolve as the knowledge structures it manages evolve. Those knowledge structures will only evolve if people interact with them, and that implies some compelling reason to do so.

Thus, I choose hard problems: environment, energy, water, cancer, and lots more. My hunch is that these constitute compelling reasons to be involved; of course, the elections in the US going on today appear to serve as a counter argument to my ideas.

Like you, Bernard, I'd prefer to see what I can do with what already exists. So, I am casting about in the open source community for things I can adapt to serve purposes I envision as needed.

Do things need to be hierarchical? Perhaps not, but that's the way a lot of stuff is organized. For now, I envision something along the lines of the SGML 'Grove' structure, a hypergraph, each node of which can be anything, including another hypergraph, a tree, whatever. The trick here is to adapt all that SGML stuff to XML. In terms of HyTime, I just discovered a family of XML dialects, one of which is called CellML, and one of which is called FieldML -- a markup language for time varying fields. I am thinking that we are not too far off from having enough tools that we can begin implementing some interesting trial projects.
Cheers, Jack
12
Bernard Vatant
11-07-2000
06:32 PM ET (US)
<Steve> Bernard, let me know where you'd like this to go from here </Steve>

I don't know really myself where I'd like all that to go. My main concern is a social one. How will we be able to share knowledge and information without being crushed by unique thought ? How can we be collaborative and keep our diversity ? To make it short, taking again the Alliance's purpose, how can we manage to be *plural* and *united* ?
That's why I care less about new structures or tools emerging for the fun of it than making existing ones meet and understand. As a teacher and now as a consultant, my work has always consisted of porting and explaining concepts. Two things I've learnt in that job are:
1. It takes time and redundancy for concepts to spread from their first emerging to their wide and full understanding. Either in a single mind or in collective intelligence.
2. Understanding is basically going from complexity to simplicity. Simplicity is the hardest thing to achieve, since it needs letting down almost everything. For example Web users and designers at large have not yet understood the simplicity of hyperlinks. We see hypersophisticated technical tools needing plug-in and applets and so on performing basically no more than a simple bunch of well built links.

<Steve>
Within small-scale disciplines, information is loosely organized as a network of continuously changing and appearing/disappearing links and content ... ontologists/librarians are motivated to allow tools (and ultimately people's understanding) to span disciplines by defining a mapping between the keywords and structures of one discipline and another, and maybe to define super-structures that span disciplines.
</Steve>

That's the way the things go and it's a real problem. Ontology is too serious a thing to let the ontologists manage it. Why do not information and knowledge managers at bottom level ( societies, research centers ...) try to discover by themselves what is the range of validity or their concepts and make them coherent with their neighbors', out of a bottom-up merging ? Maybe they lack tools to do so. Maybe they lack time. Maybe they lack imagination. I don't believe most of them remain insular on purpose.

<Steve> Can I invite one other person, in particular Dave Weinberger <http://www.hyperorg.com/)>; to this conversation? </Steve>

You are welcome to do so
11
Steve Yost
11-07-2000
01:27 PM ET (US)
I've just read all our posts again, and I want to make it clear that I agree with your philosophy here. I find that I'm focussing a lot on tools, which seems to be a little divergent from the thrust of this discussion. Bernard, let me know where you'd like this to go from here.

But for now I'll forge ahead. I wonder if we'd agree that something freeform like Wiki Wiki might work at a small enough scale. Certainly structure does "evolve" there by the work of intelligent people to create it, though it's a loose network structure rather than a sort of Aristotelean hierarchy.

But I recognize the work you're interested in is at the boundaries between disciplines and between scales of organization, where certain a priori structures are needed for interchange. Is this an accurate statement?

Searching for Henri Atlan (whose work sounds like interesting reading), I find this quote from Entre le Cristal et la Fumée:
"So then it would suffice to look at organization as an uninterrupted process of disorganization-organization, and not as a state, so that order and disorder, the organized and the contingent, construction and destruction, life and death, are no longer so distinct..."
(http://www.uia.org/uiadocs/values93.htm)

This to me sounds like a recipe for primordial soup. Or just a recognition that our superimposition of order on a chaotic world needs constant adaptation. If you're familiar with his work, I'd love to hear more about this.

I'm being doggedly tool-oriented, but here's a picture that comes to mind:

Within small-scale disciplines, information is loosely organized as a network of continuously changing and appearing/disappearing links and content. A Wiki Wiki Web, or the web itself, serves to represent this.

At a higher level, there are the librarians and ontologists that continually seek to recognize keywords and structures that evolve in this soup. (One such example is "Jerry's Brain" -- http://www.sociate.com/My_Brain/my_brain.html -- by Jerry Michalski. It's an instance of The Brain, assiduously maintained by Jerry and covering everything in his realm of interests). These structures and keywords change at a slower rate, since they by definition reflect discernible summarizing patterns.

These ontologists/librarians are motivated to allow tools (and ultimately people's understanding) to span disciplines by defining a mapping between the keywords and structures of one discipline and another, and maybe to define super-structures that span disciplines.

For example, one area I've seen this need is the fields of "Evolutionary X". There's evolutionary psychology (a.k.a. human sociobiology, human ethology, human behavioral ecology, or evolutionary anthropology: http://www.egroups.com/message/evolutionary-psychology/8246), and there's evolutionary economics (for example) with an apparently nearly disjoint set of experts. I recently read a paper in the latter subject that had many parallel ideas to papers and books in the former, but didn't cite *one* author in the former field. Was this because they're being insular on purpose, or because there's no overt mapping of concepts?
____________________
Can I invite one other person, in particular David Weinberger (http://www.hyperorg.com/) to this conversation?
Edited 11-07-2000 01:35 PM
10
Bernard Vatant
11-06-2000
04:44 PM ET (US)
Wiki Wiki Web : more amusing than amazing, I'd say ;-)
I am not that confident anyway in the auto-organizing capacity of ideas and "memetics" at large, to believe something consistent will come out of such playgrounds. Following Henri Atlan's word : "Entre le cristal et la fumée" - between crystal and smoke - living complexity is a strange mix-up of order and disorder. We are far from understanding yet how it has appeared and evolved on Earth. What we more and more presume is there has been a good deal of extraordinary coincidences behind all that story, since our quest to find it anywhere else has failed so far.
So I wonder if there is any more chance to get emerging intelligent structures out of such chaotic encounters of raw ideas than to see the Web auto-organize spontaneously or see life emerge out of Jupiter's atmosphere.
On the other hand, what we know well is that higher levels of organisation may come out of integration and merging of lower ones. So I believe firmly we'd better start to build from what is already organized knowledge than reinventing all from scratch. We have all this complex material at hand in data banks and thesauri and libraries and on the Web. That's what we have to deal with and organize, there is more in it that in any sandbox.
Remember Newton : To see far away is a question of climbing on giant's shoulders ...
9
jean delahousse
11-02-2000
04:33 PM ET (US)
Hello,
Sorry, Mondeca's Topic Navigator will only be on our web site end of december. We make the first presentation of the product during the Topic Map Interest day (december 4th) preceding XML 2000 in Washington.

Sincerely

Jean Delahousse
CEO Mondeca
Mobile : +33(0)6 16 90 73 95
Tel : +33(0)1 47 46 18 89
Fax : +33(0)1 47 46 01 09
www.mondeca.com
8
Steve Yost
10-31-2000
10:17 AM ET (US)
Mondeca's Topic Navigator looks like it might be interesting -- the demo is down, so I can't see what it looks like. Its description, and your great explanation of the problem you're trying to solve (thanks for that), bring to mind some other tools:

Wiki Wiki Web (http://c2.com/cgi/wiki?WikiWikiWeb) is an amazingly open tool for collaboration. Its success relies on a community of trust -- anyone can edit/add/delete any page. It has clever ways of automatically generating links. It's at the opposite end of the spectrum from a tool that requires a predefined formal "ontology" (I'll adopt your terminology here, since it's your discussion :). Structure would tend to be "emergent". It might be an interesting social experiment to try it in this domain, though maybe the scope of your project is too large to support something this freeform. Or maybe it's a good parallel means for information sharing.

The Brain (http://www.thebrain.com/) is just a clever way of presenting a network of hyperlinks, along with your own annotation. But the presentation is nice.

http://www.trellix.com . Trellix's history is an interesting note on the difficulty of predefining structure. Their original tool was for the creating a web of linked documents. It had you first create a detailed outline of the structure of this web (the "trellis" I suppose), before you began editing content. This product didn't sell well, and they've since changed their focus to have Trellix be a simple way to create web sites.
^     All messages            8-23 of 23  1-7 >>

Print | RSS Views: 1445 (Unique: 750 ) / Subscribers: 7 | What's this?