Re: Semantic Web as Pragmatic Web from Alexander Zelitchenko on 2007-12-25 (semantic-web@w3.org from December 2007)

From: Alexander Zelitchenko <zelitchenk@yahoo.com>
Date: Tue, 25 Dec 2007 02:11:09 -0800 (PST)
To: editor@content-wire.com
Cc: semantic-web@w3.org
Message-ID: <284999.27848.qm@web36914.mail.mud.yahoo.com>

Let us continue.

I share your vision, that sw uses too old ideas of AI and invests in them too big efforts. Of course, the initial aim was to cover applications, where there are no explicit formal (mathematical, logical etc.) models of reality, but after now near half-century of experience with AI it had to be more or less clear from the very beginning - where its methods work and how they work. But you are right also in your second observation - big investments in old ideas make it difficult (and not only psychologically) to be unfaithful to them.

What about the future of Pragmatic Web, this will be determined only by its success in resolving practical tasks and developing application in the scopes, where traditional approach did not work. But I see pw as continuation of sw, next step in our understanding what methods work and what does not work with transformation of texts into meanings. And in this perspective I do not see real big problem if some efforts with modern mode in sw will demonstrate a limitations which are more than their authors expected.

editor@content-wire.com wrote:
Thanks

One more thought from me: re.the fact that sw is appropriate for high formality environments
(I think you mean data which has been tagged/formatted in a certain way) I am not sure if thats is the way it is intended in the first place, or it is just happening like that because that's what people have managed to come up with so far. Admittedly, a lot of what is done in sw today is heavily influenced by
computer science paradigms, which tend to become obsolete so quickly these days... the technology 'markets' however function following rather lengthy 'commercial cycles', so after investing so much on some things, they are very reluctant to scrap their plans before their products come to maturity etc...
..even if we all agree that is what needs to be done... somehow I think the IT industry is between
the vision and the visionaries.....so I understand if the pw want to cut out a little corner for themselves where they want to invent their own world anew where they do not have to live with the 'real world' constraints - it would be a pity if all the efforts being put in sw today would not benefit from the freshest thinking because of some structural systemic limitation of the way things are done

Surely if the technologies that will be coming ahead will enable us to do without all this rdf/owl business trouble, and be able to reason and infer and support logic using natural languages, I think that would be a great leap forward - whatever it is going to be called..sw/pw whatnot

I ll start thinking of a defintion for pw,
best wishes
PDM

----- Original Message -----
From: Alexander Zelitchenko
To: editor@content-wire.com
Cc: semantic-web@w3.org
Sent: Sunday, December 23, 2007 3:17 PM
Subject: Re: Semantic Web as Pragmatic Web

Thank you, Paola;

To continue our exchange of views, I'd like to say that I see natural development of PW as part of natural development of SW. In fact, most important thing is the set of tasks which may be resolved in framework of given approach. Obviously, 'classic' approach of SW is appropriate for some class of problems with high formality and low uncertainty and variability, which are usually introduced by the factor of "natural human psychology'. Obviously, out of this class SW need to develop other and perhaps more specific approaches. Some of them must be called 'Pragmatic Web'. I do not afraid (at least for the present moment) fuzzy definitions of what you call 'my pragmatic web', because the value of each approach and each definition here depend on how powerful web tool this approach creates. In other words, each approach must prove its usefulness and must find its own scope of application. I see this is as the only way to clarify what is 'real PW".

But in any case, I think it would be very useful to exchange by preliminary personal definitions of 'what must be called PW'

P.S. By the way I also have no yet moderator approval from PW web-site and mailing list. Maybe Christmas time is reason?

editor@content-wire.com wrote:
Alexander
thanks a lot for your thoughts
My idea behind the semantic web corresponds to what we are now starting to call the pragmatic web
except that 'semantic web' was the name given at the time
Also my ideas of what the pragmatic web is/can be/will be is slightly different from your description, and slightly different from what the manifesto says

I would say that we could start by creating a definition, or set of definitions that we can all agree on
(otherwise it will be 'my pragmatic web')

Regarding the technology, I do think that futures gen web technologies will evolve towards what is currently defined as pw, but I do perceive that as a natural evolution of the sw and maybe therefore
will require a terminology adaptation ie name change

finally I do think that from the pw vision a few pointers should come to help untangle the current struggles and knots that we are in the sw - as a lot of the problems that we may get stuck into today may not be central to what is ahead - but this can only happen if we open up the debate to all interested parties
(I still have not received the mail confirmation from the pw mailing list! is it implicit? is anyone else on that list who could kindly nudge the list moderators on my behalf? thanks)

more to talk I guess...
cheers

PDM

----- Original Message -----
From: Alexander Zelitchenko
To: editor@content-wire.com
Cc: semantic-web@w3.org
Sent: Saturday, December 22, 2007 3:16 PM
Subject: Re: Semantic Web as Pragmatic Web

I'll simply try to explain my vision of your questions.

1. What does 'pragmatic web' mean, as opposed to 'semantic web?'
(is it just another buzzword, or does it better convey the meaning of
what is intended?

I use 'pragmatic web' not like synonym for 'semantic web'. Pragmatic web is subset of semantic web, where the practically infinite set of classes, traits, relations, attributes and so on is limited in effective way by intentions of web-participants.

The second moment seems to be very important is that 'real ontology' is determined in pragmatic web in process of INTERACTION (between 2 human participants, as it is in "Manifest of pragmatic Web group" or between dialog wizard and one human participant in the article, which I propose to your attention).

2. Are semantic web technologies, as we are developing them today,
going to be capable of delivering 'pragmatic' capabilities (that is
are they going to be capable of supporting context sensitive
reasoning?If not, what is missing)

I do not think, that they are capable. There are a lot of things, which are well known to cognitive psychologists, but are missed in present approaches. First is experience. All human semantic is based on human experience (internal images of the world, as the world is represented namely by this person). The second thing is that human semantic is always semantic of acting human being with his aims, intentions, repertoir of actions and so on. The third thing is that human system for processing of meanings is multilevel one, where different levels use not exactly same principles of processing. There is also even more metaphysical 'fourth' principle of human semantic - "making of senses", but they must be discussed separately.

3. What kind of vision is there for the pragmatic web, what technologies?

Pragmatic Web may include some thing (like seeing Web participant as acting persons, for example) into consideration and actively use man-machine interaction to clarifying 'real semantic'.

4. How would the vision for the pragmatic web fit into the vision for
the semantic web?

Semantic Web today is philosophy and technology. Pragmatic Web is part of the same philosophy, but propose quite different technology. Technology of semantic web works well only in very formal application. The more application informal the less semantic web technology is appropriate. On the contrary, pragmatic web is addressed to wide range of 'non-formal' applications. Thus relationships between these 2 classes of technologies must resulted in each of them will be used for its own tasks.