Christopher S, Amanda, Nicola, Bobbin, and Matthew, (01)
As I told Peter, I have scheduling conflicts on most Thursdays that
prevent me from calling in to the telecons. But I'd like to mention
some issues that have not been addressed in the deluge of messages
about the summit. (02)
The following comment was sent to Ontolog Forum, but it asks the
fundamental questions that have not been discussed: (03)
CS
> But does the whole effort have the support of the bio-ontology community?
> Or does the latter community not have parallel and perhaps competitive
> facilities addressing the same needs? (04)
The article cited by the ACM announcement does not say a word about
ontologies of any kind. Following is the concluding passage from
http://www.physorg.com/news/2012-01-oxford-harvard-scientists-data-sharing-effort.html (05)
> We are working with this framework to help harmonizing and presenting
> as many large-data types as possible in a common standardized and
> usable form publishing it in the associated GigaScience journal. (06)
Note that it's called a "data sharing effort". There are people at
Harvard and Oxford who have heard about ontologies and the Semantic
Web. But apparently they did not find anything they would consider
"a common standardized and usable form". (07)
On a related note, my morning mailbox stuffing included an announcement
for a webinar on "Integrating Big Data Technologies" (abstract below).
It mentions various technologies, but not ontology or the SW. (08)
AV
> The purpose of this email is two-fold: to let you know of (seed)
> material now populating the Quality Cross-track Community Input wiki page at
>
>http://ontolog.cim3.net/cgi-bin/wiki.pl?OntologySummit2012_Quality_CommunityInput
> and to initiate email discussion under the "[Quality]" subject header (09)
That is a good list of issues concerning the quality of an ontology. (010)
But many of us on this list have been discussing those issues in
the SRKB project (started in 1991), the ANSI ontology project
(mid to late 1990s), the SUO list (starting in 2000), and Ontolog
Forum. But when we compare the quality of the ontologies that came
out of those efforts with what is published in OWL on the WWW and
with the Big Data projects mentioned above, there is a common thread: (011)
The use of an ontology is inversely proportional to its quality. (012)
If you examine the OWL ontologies on the WWW, the overwhelming majority
are nothing more than terminologies with angle brackets. The only
definitions are stated in English comments, and the only operators
are subclass and disjoint. For computation, they are nothing more
than a tree of undefined terms. That's certainly decidable. (013)
For schema.org, the founding members -- Google, Microsoft, and Yahoo!
-- took the obvious next step: throw away the bloated angle brackets
and express the hierarchy in JSON. (014)
NG
> I believe that the questions below ("Why haven't semantic technologies
> been adopted as an integral part of the IT mainstream?" and similar),
> although very relevant, are out of scope for the present Summit. Indeed,
> these issues have been discussed rather extensively in the latest Summit, (015)
Last year's summit did not get into those issues. It consisted of talks
that said "Ontologies are wonderful. You should be using them." A year
later, we see that the major web companies have abandoned any pretense
of using ontologies, and people who need to share Big Data ignore them. (016)
BT
> We keep dancing around what I would call process (rules/algorithms are
> sort of 'micro-processes'), and it would be great to hear thoughts about
> an eXecutable ontology -- one that combines in one structure, data with
> the processes that use them; that combine the 'what' with the 'how'; that
> might combine the wave view (process) with the particle view (graph of
>structure). (017)
This gets us to the heart of the matter: All usable systems provide
a clean, well-defined, easy-to-use interface between the methods for
defining the semantics of the data and the software that processes it. (018)
But the ontology languages ranging from OWL to CycL and many others
in between are isolated from mainstream IT. (019)
BT
> Would that bring us to a tTOE -- a technical TOE (Theory of Everything
> as in physics), and is it logically possible. (020)
Yes. It's a very simple tTOE: (021)
Any technology that's isolated from mainstream IT will be ignored. (022)
MW
> Ontology is like the air we breathe. In information systems it is all
> around us, but we often do not notice it. However, if there are problems
> with it can choke us. (023)
I agree with that point, but I'd like to elaborate on the problems
and the cause of the choking. (024)
In your work at Shell, you used Express, which is a highly expressive
ontology language (equivalent in power to FOL) that you and your
colleagues used successfully for large-scale projects. The tools
that processed it were thoroughly integrated with the development
process for the applications used at Shell (and at many other
businesses around the world). (025)
But very expressive languages like CycL, which could define anything
that is definable in Express, lack the kinds of interfaces and
methodologies that made Express successful. (026)
The weakly expressive languages like OWL DL also lack the tools and
methodologies you had with Express. But they are hamstrung by
a constraint that makes them unusable for defining the kinds of
structures you specified in Express. For example, OWL DL is
deliberately constrained to be "decidable". The OWL designers
chose a brute-force way of making it decidable: all models are
limited to tree structures. (027)
That constraint implies that you can't define a triangle or a
a benzene ring in OWL DL because they happen to be connected
in a cycle. That is one reason why published OWL ontologies
put the definitions in English comments. Another reason is
that no tools for doing anything more with OWL have any
connection to any kind of computation that people actually do. (028)
Short summary: Mainstream IT ignores ontologies for a simple
reason: they are isolated from anything and everything that
mainstream IT is doing. If the summit does not address
that issue, it will be totally irrelevant. (029)
John
_________________________________________________________________________ (030)
Integrating Big Data Technologies (031)
Presenter: Krish Krishnan, Sixth Sense Advisors Inc. (032)
The hottest topic of discussion around any data related subject today is
Big Data. We all know what is Big Data, the related technology wave it
has brought in its wake Hadoop, Cassandra, NoSQL and more. This is where
the confusion starts, what are these technologies and how do they fit
your environment? How do you go about trying them out and how will you
integrate them? Along with the technical issues, you also need to ponder
on do I need quants to build algorithms or do I download them from open
source? Who can guide me and how can I go about these tasks. (033)
This webinar is focused on presenting an information filled session on
the topic with tips and techniques around the same. We will discuss real
life applicability of the technologies, success criteria evaluation etc. (034)
_________________________________________________________________
Msg Archives: http://ontolog.cim3.net/forum/ontology-summit/
Subscribe/Config: http://ontolog.cim3.net/mailman/listinfo/ontology-summit/
Unsubscribe: mailto:ontology-summit-leave@xxxxxxxxxxxxxxxx
Community Files: http://ontolog.cim3.net/file/work/OntologySummit2012/
Community Wiki: http://ontolog.cim3.net/cgi-bin/wiki.pl?OntologySummit2012
Community Portal: http://ontolog.cim3.net/wiki/ (035)
|