[Top] [All Lists]

Re: [ontolog-forum] FW: Google Knowledge Graph

To: "[ontolog-forum]" <ontolog-forum@xxxxxxxxxxxxxxxx>
From: Duane Nickull <duane@xxxxxxxxxxx>
Date: Fri, 18 May 2012 03:52:02 +0000
Message-id: <CBDB143B.1198AB%duane@xxxxxxxxxxx>
For background on how it works, please see my 2009 presentation:    (01)

c-behind-flash    (02)

How Google actually works is not how you  probably think it works.    (03)

Duane    (04)

Consulting and Contracting; Proven Results!
i.  Neo4J, Java, LiveCycle ES, Flex, AIR, CQ5 & Mobile
b. http://technoracle.blogspot.com
t.  @duanechaos
"Don't fear the Graph!  Embrace Neo4J"    (05)

On 12-05-17 6:16 AM, "David Eddy" <deddy@xxxxxxxxxxxxx> wrote:    (06)

>John -
>On May 17, 2012, at 7:56 AM, John F Sowa wrote:
>>  But there is a huge difference between
>> a search engine that looks for exactly those strings you are searching
>> for and an "intelligent" search engine that second-guesses what it
>> thinks you want.
>Interesting how I hear more & more people expressing dissatisfaction
>with Google results.
>My Turing test for search...
>Context is software source code so the corpus is very small, less
>than 10 million lines.
>The project "standard" abbreviation for "CODE" is "CD" which in
>practice is most often used as a suffix.  (We're using IBM's "OF
>Language" approach for PRIME-MODIFIER-CLASS words & "CD" is the class
>word.  Folks seem to have forgotten "OF", but most programmers still
>use it... witness Objective-C's example splitView:
>Unfortunately since the "standards" are manually enforced, over the
>years, variety happens.
>When the database is expanded, an unknowing/new DBA uses "CDE" as the
>abbreviation for "CODE."
>Therefore searching for "-CD " (notice the space after "D"... this is
>a suffix), with normal string search "-CDE " will NEVER be found.
>A human, by painful experience, will discover that mostly "CD" is the
>abbreviation, but in this particular corner "CDE" means the same thing.
>As systems evolve into 3rd-4th-5th-nth generation staff, such search
>anomalies get to be expensive.  I believe such decay would fall into
>the "technical debt" bucket.
>Since retrofitting "CDE" to "CD" is typically not an option (mass
>spelling changes to production code is far too risky), I would
>recommend an annotation approach that would say:  CODE = {CD, CDE}.
>Search for one, get all.
>David Eddy
>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
>    (07)

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    (08)

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