Hi Paola, (01)
I am indeed saying that knowledge must always be partial. But I'm not
saying it must be partial because there is too much. The limitation is
more interesting. It is in the nature of knowledge itself. (02)
The partial nature of knowledge is evident even in the smallest part of it. (03)
Take the analogy of a hologram again. It stores many different
perspectives of an object. But you can only see one perspective at
once. The limitation is not because there are too many perspectives.
It is because, by the nature of perspectives, only one is possible at
a time (you can only sum the diffraction pattern one way at a time.) (04)
If you like, I'm saying knowledge is different perspectives on data
(different ways of collecting it together) and only one is possible at
a time. (05)
To try and be a little concrete for a minute, what this means in terms
of the Web is that the Web itself (by virtue of its randomness) is the
most compact representation of the knowledge it contains. We should
keep the Web and expand out the knowledge it contains as different
combinations of its parts, as we need them. (06)
In fact, of course, that is what the most successful "knowledge
applications" of our day, search engines, do already. (07)
With a little theory to back it up, we could do better. Search engines
today find only very crude patterns in text. They find the patterns
identified by sets of words, but there is no structure to them. (08)
We have ways of finding structure, but we've been confused because the
structure we find is always partial. (09)
Our mistake has been to focus on the structure itself, rather than the
process of finding it. Understand we can only expect partial
structure, that this is the sign of a more powerful system, and it
shouldn't be too difficult to find the structure relevant to any given
problem, when we need it. (010)
-Rob (011)
On Feb 11, 2008 11:53 AM, <paola.dimaio@xxxxxxxxx> wrote:
> > But this "unpredictability" is not a bad thing. It means a hologram
> > can code much more information than an ordinary photographic plate. In
> > fact intrigued by Pat's analogy I chased up some vague references to
> > holographic memories and found that, indeed, technology is starting to
> > get hip to this. Among others I found this:
> >
> >
> > Equate resolution to predictability, and you have a sense for the
> > "randomness" of any grain of emulsion in a hologram.
> >
> Rob,
> I find your theory above interesting and challenging
>
> That random patterns of information , as unordered sets of assorted
> values, can reveal more details than ordered and defined bits. Like
> looking at the negative impressions of images reveals the shape of
> what is surrounding the image, than of the image itself, and makes one
> understand more about the object observed by highlighting its
> surrounding.
>
> Basically to me your idea corresponds to the notion that the
> knowledge that we are capable of storing, is always a limited subset
> of the knowledge that exists. Knowledge, like the universe of
> discourse, like information and the universe itself, is
> infinite (I know some theories purport otherwise), and what we can
> sort and apply is only some part thereof.
> Next, all we need is a method to query and apply all the knowledge
> contained in such infinite randomness, to make our defined information
> more useful (other than becoming a search engine addict and developing
> your own heuristics)
>
> I think what you say above points to the issue of 'context' already
> widely discussed on this list, although I am not sure we reached a
> conclusion on that one yet
>
> More to the meaning of words when I can wrap my mind around that
>
> Paola Di Maio
> School of IT
> www.mfu.ac.th
> ********************************************* (012)
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
Subscribe/Config: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/
To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx (013)
|