Formal ontology chaosΒΆ

Tags: aec

Ben Hyde has a nice thing to say about committee-developed ontologies versus massively volunteer-developed folksonomies.

Folksonomies is a pretty new term for stuff like del.icio.us, where everybody can tag all sorts of information with tags of their own devising. Yeah, that's chaos - but pretty chaos. I watch what people add to the tag startup and despite the occasional page about starting MS windows, I've gleaned a couple of useful pages for starting up a company from there.

Chaos! Nobody in control! This can't be right!

Ben is right: Folksonomies - nobody in control? What! that's silly, the site designers are in control. If the ontologists wish to remain in control they had best start building sites, and fast.

I'm working in the building and construction industry: there are a couple of committee-based ontology efforts there. Any extra project won't add anything major. So: add a folksonomy-like project! Let the committee try to keep up with the unwashed masses. (Or laugh at the stinking pool of utter chaos in which the masses slowly drown themselves :-) Note: the bcoWeb building-construction ontology we've started at Delft University of Technology is intended for the unwashed masses.

I've attached part of chapter four of my upcoming PhD thesis ("New instruments for dynamic Building-Construction"). The part about "lightweight development" is the most relevant here. Development and open source

The key question then seems to be: how to develop a standard that is semantically rich, can be useful and even commercial attractive within a short period of time (less than one year) and allows the owners of the information and knowledge to keep control and ownership.

The suggestion put forward in this research is fourfold. The cost for use should be minimal; there should be as much application support as possible; development should be lightweight; the development should be supply/demand-oriented.

Low cost of use
All other things being equal, a lower price increases demand. The suggestion for this concept solution is to make the usage free of charge. Free of charge does not equate to low quality, as is exemplified by the free, online encyclopedia wikipedia .
Much application support
Application support will depend on individual economic decisions on the part of the software vendors. What can be done, though, is to do away with all barriers for the software vendors. So: no mandatory membership of an organisation, no fees payable before you're allowed to access the object library, no risk by tying yourself to one platform vendor. This all can be achieved by making the object library open source. This makes the object library free in the sense of `gratis' and free in the sense of freedom to use, freedom to change, freedom to redistribute. One restriction that might be a good idea is to demand that the redistributed or changed object library contents be freely available under the same conditions, this as a safeguard against an embrace-and-extend attack by one market party. A different position is possible on this point, though.
Lightweight development
Heavyweight committee-based development is slow. Some of this work is rightfully hard and should take a long time. The solution concept, though, suggests to develop the object library as multiple small pieces that can be made by just a few interested individuals, without having to resort to formal organisations and meetings. The next generation internet possibilities allow you to tie the various independently developed parts together in a web-based network.
Supply/demand oriented
To achieve immediate relevance and interest in the building and construction industry, a new development has to attach itself directly to the money-making process. If an ontology can be build on basis of a supply/demand distinction, it can start lubricating the market mechanism in the building and construction industry. At the least it can provide a few innovating new ways to connect supply and demand.

Concluding, the proposal is for an open source web-based object library as a very interesting new effort at creating a useful basis for web-based communication in the Building-Construction.

blog comments powered by Disqus
 
vanrees.org logo

About me

My name is Reinout van Rees and I work a lot with Python (programming language) and Django (website framework). I live in The Netherlands and I'm happily married to Annie van Rees-Kooiman.

Weblog feeds

Most of my website content is in my weblog. You can keep up to date by subscribing to the automatic feeds (for instance with Google reader):