Re: [semanticweb] how to explain to humans the term ontology or the name of the rose

Martin,

I like the fact that you substitute to 'personal ontology' a more
suitable term, namely a 'conceptualization'. I would go further and
say that this has nothing to do with ontology but only belief systems
which may or may not be in line with an ontology.

Marja,

I'm not trying to chastise or split hairs for the sake of it, but I'm
truly puzzled by what you say here.

I don't have principled reasons against ontological pluralism, but I'm
a bit leery about your talk of 'realities'. That makes me think that
it's not ontologies that you are looking for but as Martin remarks,
conceptualizations (in the sense of mind representations), and
possibly only a vocabulary or glossary (linguistic representations).

What's puzzling is that you speak so casually of your reality, others'
reality and many realities. Another thing that you say and that
doesn't make sense to me is that you like to 'simplify reality'. Just
to be sure that I understand you well, are we in agreement that what
you call 'reality' is something like a view or an understanding of or
maybe a perspective on reality, the world, maybe of that portion of
the world to which you have access? If your talk is metaphorical in
that sort of sense, it seems to me very important to not use that sort
of metaphor, simply because it is dangerously misleading.

I doubt that you or anyone else has the ability to create reality or
to simplify it (I'm not speaking of creating or destroying things, but
reality itself). You are part of reality and you take stances (as you
say, you perceive, name, classify, etc... as much as you touch or kick
things in reality). The rough point here is that there are relations
between you and things (it doesn't matter that you create these
relations, now they count as things). This cognitive dimension belongs
to but one small domain of reality, in other words, there is an
ontology of cognition (a domain ontology) and taking it into account
is but making an extension of non cognitive ontology.

A risk which I see in what I take to be metaphorical talk is to
obliterate that relational aspect and turn a conceptualization
abstracted from these relations into a self-standing ontology. (Maybe
there are good technical reasons to do this, although it looks like
this also brings in troubles, e.g. of interoperability.) But these are
no, imho, ontologies -- or only metaphorically if you want, or by
analogy of structure of their formal representations with that of an
ontology. No more than there are multiple realities, it seems to me
that there are no multiple ontologies defined by cognitive agents, but
multiple belief systems if you'd like or world views.

Other than that, I find ontological pluralism a credible philosophical
stance. But not because there might be multiple agents with their
peculiar representation, rather because it is not incredible that
reality could be broken down in different ways, each of these breaking
down being adequate. Sorry for being so vague, the only point I want
to stress is that if there is ontological pluralism it is rooted in
reality, not in cognitive agents which root, let's call it, conceptual
or representational pluralism.

To me these distinctions are unavoidable, but too easily obscured. The
interesting questions are whether we can do after all with multiple
representations or whether in addition to these we have to have an
ontology or again whether we can do just with an ontology. I don't
have a principled answer, it's credible that we could get different
answers depending on the applications (compare, I don't know, plain
weather information and gift advisors).

Now,  if your talk of multiple realities was not metaphorical...

Cheers,
Pierre

On 1/12/06, Martin Hepp (DERI extern) <martin.hepp@deri.org> wrote:
> Marja:
> Good point! The reason why I stressed the community aspect of an ontology is
> that it needs commitment by the community necessary for achieving the given
> purpose, and that an ontology created by someone in his/her room but *for a
> bigger audience* is unlikely to achieve such consensus.
>
> However, I agree that even a local conceptualization (= personal ontology)
> makes a lot of sense, since then you have at least a conceptualization that
> can be a starting point for mediation. Fausto Giunchiglia is doing a lot of
> work in this direction, see http://dit.unitn.it/~accord/.
>
> A personal ontology has full commitment of the target community, which is
> solely its creator.
>
> In the continuum of "one ontology for the world" and "one individual
> ontology for everyone", I am much more on your side.
>
> Martin
> ---------------------------
> martin.hepp@deri.org, phone: +43 512 507 6465
> http://www.heppnetz.de / http://www.deri.org
>
> -----Original Message-----
> From: semantic-web-request@w3.org [mailto:semantic-web-request@w3.org] On
> Behalf Of Marja Koivunen
> Sent: Donnerstag, 12. Jänner 2006 19:55
> To: semantic-web@w3.org
> Cc: semanticweb@yahoogroups.com
> Subject: Re: [semanticweb] how to explain to humans the term ontology or the
> name of the rose
>
>
> For a long time I have talked about the value of "personal ontologies"
> in addition to having more standard widely agreed upon ontologies and let
> people define their own concepts (they will do it anyway) until they are
> ready to share them with larger groups and/or try to learn the more standard
> concepts and what they mean in each person's own reality and link from their
> concepts to the standard concepts.
>
> This is all because in my reality this makes sense and I readily accept that
> it may not make sense in some other people's reality and that's fine. My
> reality is not only dependent on my abilities to see things, hear things,
> feel things, smell and taste things and understand things based on the
> current wiring in my brain (with all my cultural bias) but also what I think
> is important for my current goals. I often like to simplify the reality when
> it makes sense. However, in addition I also would like to know and
> understand the exact reality with all the bits and pieces in their place,
> and so do a lot of other researchers who unfortunately have only come up
> with several models, none of which explains everything, some of which do fit
> together and some of which are simplified to make their use easier as they
> are close enough in most of the cases.
>
> So what's an ontology? I don't really know exactly, I just have my own
> approximations. Is it impossible to say "personal ontology" as I have been
> told many times? I have already said it many times so it was not totally
> impossible although it has created many discussions during the years. Now
> I'm starting to like the folksonomy concept for describing what I called
> "personal ontology" because in my reality it makes sense but most probably
> that's totally wrong too in some other realities.
>
> Marja
>
>
>
>
>
>
>
> Yahoo! Groups Links
>
> <*> To visit your group on the web, go to:
>     http://groups.yahoo.com/group/semanticweb/
>
> <*> To unsubscribe from this group, send an email to:
>     semanticweb-unsubscribe@yahoogroups.com
>
> <*> Your use of Yahoo! Groups is subject to:
>     http://docs.yahoo.com/info/terms/
>
>
>
>

Received on Friday, 13 January 2006 13:42:19 UTC