ontolog-forum
[Top] [All Lists]

Re: [ontolog-forum] UIMA

To: "[ontolog-forum]" <ontolog-forum@xxxxxxxxxxxxxxxx>
From: John Bottoms <john@xxxxxxxxxxxxxxxxxxxx>
Date: Wed, 07 Dec 2011 10:11:57 -0500
Message-id: <4EDF823D.9040708@xxxxxxxxxxxxxxxxxxxx>
re: The Birth of a Word

Carl,

I agree, but the most important design requirement is "what is the problem you are trying to solve?".
Of all the examples of intelligence we find in nature we can see two key ingredients in the development of higher levels of understanding. First, knowledge progresses from simple understandings to more complex, it is not possible to integrate knowledge that has no understanding underpinning it.

I have had multiple discussions where i discuss this view and people; parents tell me: "I always talk to my kids with 'grown-up' language. It would be demeaning to talk to them as children." But as Deb Roy explained in his Ted lecture, the acquisition of speech corresponds directly to a dip in the complexity of a speech pattern by the speaking caregivers.

It is an interesting discussion of a complex topic that involves teacher & learner in a bonded feedback loop. The video relates the tracking of a single word, "water", captured from early babbling to the point of correct pronunciation (point 4:50 in the video). The graph is shown at 6:30. If you want to see the video it is at:
   http://www.ted.com/talks/deb_roy_the_birth_of_a_word.html

And Second, all knowledge, speech and intelligence develops within an ecological niche. If you don't understand the niche and it's threats and resources, then it is impossible to define a system architecture to address survival within that niche. Our learning and knowledge is about those resources and my philosophy professor repeatedly told me, "There are no rice gods where there is no rice".

-John Bottoms
 FirstStar Systems
 Concord, MA

On 12/7/2011 9:39 AM, carl mattocks wrote:
Duane:

I was on the UIMA TC and experimented with some open source code .. rather than starting from scratch ... it is all good .. but before you dive into using the code you really need to absorb the model and have a target outcome in mind .

cheers
Carl Mattocks 

On Tue, Dec 6, 2011 at 10:27 PM, <ontolog-forum-request@xxxxxxxxxxxxxxxx> wrote:
Send ontolog-forum mailing list submissions to
       ontolog-forum@xxxxxxxxxxxxxxxx

To subscribe or unsubscribe via the World Wide Web, visit
       http://ontolog.cim3.net/mailman/listinfo/ontolog-forum
or, via email, send a message with subject or body 'help' to
       ontolog-forum-request@xxxxxxxxxxxxxxxx

You can reach the person managing the list at
       ontolog-forum-owner@xxxxxxxxxxxxxxxx

When replying, please edit your Subject line so it is more specific
than "Re: Contents of ontolog-forum digest..."

Today's Topics:

  1. Re: RDF vs. EAR (Duane Nickull)
  2. Re: RDF vs. EAR (Duane Nickull)
  3. UIMA (Duane Nickull)
  4. If you can't beat 'em, join 'em. (John F. Sowa)
  5. Re: If you can't beat 'em, join 'em. (Kingsley Idehen)
  6. Re: RDF vs. EAR (Kingsley Idehen)
  7. Re: UIMA (Simon Spero)
  8. Re: If you can't beat 'em, join 'em. (Rich Cooper)
  9. Re: UIMA (Peter Yim)
 10. Re: RDF vs. EAR (Duane Nickull)
 11. Re: If you can't beat 'em, join 'em. (AzamatAbdoullaev)
 12. Re: If you can't beat 'em, join 'em. (Kingsley Idehen)
 13. Re: RDF vs. EAR (Ed Barkmeyer)
 14. Re: RDF vs. EAR (John F. Sowa)
 15. Re: RDF vs. EAR (Duane Nickull)
 16. Re: RDF vs. EAR (John F. Sowa)
 17. Re: RDF vs. EAR (Duane Nickull)
 18. Re: RDF vs. EAR (Zhuk, Yefim)
 19. Re: RDF vs. EAR (John F. Sowa)
 20. Re: RDF vs. EAR (Duane Nickull)


---------- Forwarded message ----------
From: Duane Nickull <duane@xxxxxxxxxxx>
To: "[ontolog-forum]" <ontolog-forum@xxxxxxxxxxxxxxxx>
Date: Mon, 05 Dec 2011 20:15:22 -0800
Subject: Re: [ontolog-forum] RDF vs. EAR

On 11-12-05 5:45 PM, "Kingsley Idehen" <kidehen@xxxxxxxxxxxxxx> wrote:

>Yes, but for bootstrap at InterWeb scale, it hasn't worked. Today's
>developer aren't ready to absorb the cost of parser development
>especially in the Open Source era where purchasing commercial quality
>grade libraries isn't the preferred options. People really like the
>"Free" part of the "Freedom" aspect of Open Source. Basically, they want
>the "Free Beer" and that's where XML parser problems start.

DN: I have to disagree with this.  I've worked in the commercial software
industry since XML was a recommendation in 1998 and have worked on
numerous solutions around B2B.   The most widely used XML parsers are
FOSS.  Xerces, JDOM et al for Java are driving most large businesses XML
parsing requirements. Almost every language has good XML capabilities.

The main cost of XML interchange is mapping the instance data to the
business services based on its' semantics.

Duane
_

Überity.com
Principal Data Architect &
Adobe LiveCycle ES Consultant Services
http://www.uberity.com
Blog | http://technoracle.blogspot.com
Twitter | @duanechaos

---------- Forwarded message ----------
From: Duane Nickull <duane@xxxxxxxxxxx>
To: "[ontolog-forum]" <ontolog-forum@xxxxxxxxxxxxxxxx>
Date: Mon, 05 Dec 2011 20:22:42 -0800
Subject: Re: [ontolog-forum] RDF vs. EAR
Very valid comments.

Duane
________________________________________

Überity.com
Principal Data Architect &
Adobe LiveCycle ES Consultant Services
http://www.uberity.com
Blog | http://technoracle.blogspot.com
Twitter | @duanechaos

On 11-12-05 7:05 PM, "John F. Sowa" <sowa@xxxxxxxxxxx> wrote:

>Duane,
>
>I have a very high regard for the *ML family of languages, which I have
>been using since the 1970s.  I used GML for formatting documents for
>most of my career at IBM -- that includes creating camera-ready
>copy for my 1984 book and many papers that I published externally.
>
>In the late 1980s, I participated in some workshops in computational
>linguistics, where I recommended SGML for annotating documents.  I
>wasn't the only one who did so, but I was one of the early adopters.
>
>When HTML came out, it was easy for me to convert GML documents to
>HTML documents (although I missed many of the GML formatting tags
>that had no HTML counterparts).  I still use HTML for doing almost
>all of my word processing.  Then I use OpenOffice to do the final
>formatting and translation to PDF.  (And OpenOffice uses XML under
>the covers for their document formats).
>
>I also use Mozilla tools (which are also based on XML).
>
>> XML is an "_expression_" of data.  This can be metadata, abstract
>> models, UML or just about any data imaginable.
>
>When it's appropriate, I definitely approve.  But the "sweet spot"
>for *ML is in marking up documents.  That includes inserting tags,
>such as the RDFa tags or the Microdata used by schema.org.
>
>But for anything beyond tags that are immediately related to the
>document (as in RDFa), XML is so far out of its sweet spot, that
>it very quickly turns sour.  _javascript_ uses the script-tag for
>inserting language-like material into web pages.  That is still
>the best way to embed languages into a web page.
>
>> I think that your assertion that Tim did it for political pressure
>
>I didn't say that Tim B-L exerted the pressure.  And I wasn't the
>one who introduced the word 'political'.  But just listen to Guha's
>talk.  He wanted to use LISP notation for RDF.  So did Ora Lassila,
>Pat Hayes, and many others who had long experience in AI.  I wasn't
>there at the time, so I don't know who to blame.
>
>> XSLT, SML Schema etc all come from the same basic model
>
>XSLT is another disaster.  There have been good tools for processing
>languages since the 1960s.  I developed some myself.  But XSLT is
>a horribly inefficient example.  I'll admit that XSLT was designed
>for processing languages whose syntax uses XML.  But for most of those
>languages, XML notation was a very bad choice -- for both human *and*
>computer efficiency.  You cannot improve a bad syntax by using a bad
>tool to process it.
>
>John
>
>_________________________________________________________________
>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
>

---------- Forwarded message ----------
From: Duane Nickull <duane@xxxxxxxxxxx>
To: "[ontolog-forum]" <ontolog-forum@xxxxxxxxxxxxxxxx>
Date: Mon, 05 Dec 2011 20:26:56 -0800
Subject: [ontolog-forum] UIMA
Has anyone on this group got experience with UIMA?  We are contemplating a
project at Uberity that involves using it.

http://uima.apache.org/

Duane
________________________________________

Überity.com



--
CEO CheckMi
Chair OASIS BCM Tech Committee
President BTUSAR Dive Team
Producer Shore Adventure
Cell (usa) (732) 497-CARL {2275}
CarlMattocks@xxxxxxxxxxx
Information Risk Checks > Secure Continuous Service




_________________________________________________________________
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    (01)

<Prev in Thread] Current Thread [Next in Thread>