Hello again, UoMers and Ontologgers in general! (01)
The activity building up to the UoM ontology face-to-face tomorrow
(30 October) prompts this pebble-in-the-pond. (02)
I have been following those fascinating and instructive doings
with great personal interest (and not surprisingly so, given my
own present situation, namely, being in mid-flight in my own
ontology-related development which I have sporadically and
clumsily been trying to set out on Ontolog since December 2007, as
further set out below). (03)
I have been particularly struck by one general feature of your
present doings, which I perhaps rather unfairly yet usefully
oversimplify in this way: (04)
There is no commonly-agreed set of use cases for the UoM ontology.
The vexed matter of "scope" crops up regularly. The most
significant progress towards a definition of scope may have been
the position staked out by Pat Hayes, and generally accepted (I
believe...), that for something to be achieved in our lifetimes we
had better leave out uncertainty and tolerance! (05)
Resonating with that angst is the most active present thread on
the ontolog forum itself: "Just What Is an Ontology, Anyway?". (06)
Is it not remarkable, after nearly a decade of SUO list and
Ontolog Forum existence, that such basic questions are still so
preoccupying? (07)
Anyway, this situation, whether it really is ridiculous or I
completely misunderstand it, pushes me to some further jottings on
the very clear position as to use cases, scope and the nature of
ontology/ies I have adopted in my own present project, that is,
the RTM or "Ride The Mainstream!" project to establish MACK, the
"Mainstream Architecture for Common Knowledge", as the basis for
the "new web" or next generation of Internet-based media. The
project's present home page, http://TheMainstream.info, dated
January 2009, even opens by calling it "a tightly circumscribed
project" (with the last 3 words highlighted in bold). (08)
It is true that that website is very incomplete, even as it
announces itself. But that merely reflects my present scheduling
of my two parallel activities, namely programming the initial
proof-of-concept "AOS" or "Application Operating System" (to
replace the web browser and in general act as the universal
front-end), and trying to describe it all in such a way that one
or more of you will want to help set up a good, small team to
expedite that programming. It is the former or programming track
which I am trying hardest to advance at the moment. Anyway, here's
a new little shove along the other track, namely the team
advertising one. (09)
As set out in greater detail on the web elsewhere (Try googling
Scylla metaset spottiswoode), I started the RTM project in 1963
with the set of use cases which were brought together in 1966
under the classic rubric of steering between Scylla and Charybdis.
That was before I had had any involvement with Information
Systems. In present terms, Scylla the many-headed monster of
oversimplification is exemplified by failed doctrines, dogmas,
paradigms, ideologies, theories and, of course, by any conceptual
schemas when glorified with the pretentious and misleading name
"ontology" (lower-case "o", and cf. George Box's "all models are
wrong but some are useful", as Andreas Tolk's recent Ontolog post
reminded us). So the use case here is that we should always
follow Homer's Circe's advice to Odysseus: "Hug Scylla's rock!"
We have to simplify the complexity of unstructured reality, even
at the expense of the inevitable losses to Scylla the man-eating
monster, who is destined always to devour some of the best of the
crew of every passing ship. Concretely that lands us in the most
general market, where products represent simplifications and
eventual oversimplifications of needs, with facilitating and
stimulating market processes, all round the recurring marketing
cycle from needs awareness to product deployment, being the
relevant set of use cases. (010)
So, the first contrast between the RTM and the UoM projects is
that the latter is always talking about "an" ontology. MACK on
the other hand, as we'll see more of it below, and as realized by
an AOS, deals with ever-changing combinations of compatible
"Forms" ("conceptual Forms"), leveraging the conceptual
components that our intellectual history has distilled out as more
or less mutually orthogonal and therefore composable. They
encapsulate our discoveries of invariances abstracted from the
flux of ever-changing reality. (011)
Here some of the most relevant details of MACK are in the 2nd, 3rd
and 4th instalments of my "MACK basics" series on Ontolog, as
follows:
http://ontolog.cim3.net/forum/ontolog-forum/2008-02/msg00291.html#nid06
http://ontolog.cim3.net/forum/ontolog-forum/2008-03/msg00249.html#nid032
http://ontolog.cim3.net/forum/ontolog-forum/2008-04/msg00109.html#nid010 . (012)
Set across the strait from Scylla, Charybdis the whirlpool of
complexity is the Homeric package of the next set of use cases.
Its most interesting feature is the great figtree growing over
complex reality and tending to hide its unstructured and often
destructive nature, Homer's "dark sands of the sea bottom" which
one can only glimpse from afar. The figtree's branches, onto one
of which Odysseus clung for a while to save himself from
annihilation by the vortex, represent the mainstreams of
established cultures and institutions. The prime function of "The
Mainstream", then, is to enable us to meet the recurrent
challenges of escaping the immobility of such social predicaments.
That, Odysseus achieved when Charybdis regurgitated his raft
consisting of the mast and keel of his wrecked ship, "though they
came up very late, not till such time as a judge with long list of
disputes to settle between obstinate litigants rises from court
for his evening meal." (And what an enormously significant
metaphor Homer chose to employ in that scene of elemental physical
catastrophe!) (013)
Thus the Homeric keel is the structured nature of our conceptual
knowledge, in logical space. And the Homeric mast is that which
holds the sails which catch the winds of progress: human
creativity and its products operating in time dimensions,
mediating process. That represents the second contrast between
the RTM and UoM or other conventional notions of ontology: in a
MACK-architected digital environment _all_ processing takes place
within structured and Form-delineated contexts. Thus "Forms" (ex-"ontologies",
renovated) and their local instances, under the impact of outside
events, fully circumscribe the states, recognizable events and
resulting state-transitions of IS applications. (014)
I have a new metaphor which portrays that dynamic process. It
further builds on the Koestler view of creativity as "bisociation"
from one reference-frame to another. (I had introduced that view
on slide 24 of the slideshow I had prepared for my once-scheduled
contribution to the Standards and Ontologies Summit last April and
which, with its many added notes, is now at
http://TheMainstream.info/RTM.html.) (015)
Thus the streams of state-transitions as managed by a
MACK-conformant AOS can be compared with the metabolic pathways in
the living cell (see e.g.
http://en.wikipedia.org/wiki/Metabolic_pathway), which consist of
specific steps or state-transitions from one coherent and
relatively stable state to a new one. (Integrating
gene-expression and other DNA-related pathways in the nucleus most
evocatively enriches the metaphor by involving the meme concept!) (016)
(Here we do well to consider how metaphors are great unifiers of
the otherwise-disparate facts and features with which a new domain
easily overwhelms us, even though they can blind or mislead. But
of course our Homeric metaphor is there to remind us in turn of
such phenomena too, and help us resist their depredations!) (017)
But how does that whole picture relate to our present Units of
Measure exercise? (018)
Well, MACK handles them fully canonically! And modularly too, of
course, in the MACK-canonical way that has very little to do with
namespaces, that historically-created diversion from the real
modularity that the omnipresent processes of abstraction towards
orthogonality as properly understood make available to us. It's
really very simple: there is nothing that has been discussed on
the uom-ontology-std forum - not even uncertainty and tolerance! -
that does not fit in easily, reusably, controllably, and
interoperablely. Quite frankly, I have been amazed at the
confusion and fuss. Well, what's the solution then? My
suggestion is brief enough: go back and piece together the story
as set out here and on the relevant pages linked-to above. (019)
However, I am afraid that there I have set you a well-nigh
impossible task. I really should do better with my whole
exposition. But until the first AOS makes it plain to any kind of
user (which will be thanks also to the foreseeable effects of the
thereby-created marketplace...) it is very difficult to portray
the outcome of a process which diverged from conventional
professional pathways nearly half a century ago. But my past
career as applications programmer, system programmer, database
designer, and application development tool seller and trainer
makes it plain enough to me that correct, efficient, trustable and
flexible programmability is not the problem. The problem, I
believe, is that I need the input from - and particularly the
stimulus of - some small team to help bring the existing pieces
together. Do you perhaps have a role to play there? Or do you
know someone who might? (020)
Sorry, but that's the best I can do on this front right now, and,
hopefully, thanks in anticipation of your role, sooner or later! (021)
Christopher (022)
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
Config Subscr: 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 join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J
To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx (023)
|