<html xmlns:v="urn:schemas-microsoft-com:vml" xmlns:o="urn:schemas-microsoft-com:office:office" xmlns:w="urn:schemas-microsoft-com:office:word" xmlns:m="http://schemas.microsoft.com/office/2004/12/omml" xmlns="http://www.w3.org/TR/REC-html40"><head><meta http-equiv=Content-Type content="text/html; charset=us-ascii"><meta name=Generator content="Microsoft Word 14 (filtered medium)"><style><!--
/* Font Definitions */
@font-face
{font-family:Calibri;
panose-1:2 15 5 2 2 2 4 3 2 4;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
{margin:0in;
margin-bottom:.0001pt;
font-size:11.0pt;
font-family:"Calibri","sans-serif";}
a:link, span.MsoHyperlink
{mso-style-priority:99;
color:blue;
text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
{mso-style-priority:99;
color:purple;
text-decoration:underline;}
p.MsoPlainText, li.MsoPlainText, div.MsoPlainText
{mso-style-priority:99;
mso-style-link:"Plain Text Char";
margin:0in;
margin-bottom:.0001pt;
font-size:11.0pt;
font-family:"Calibri","sans-serif";}
span.PlainTextChar
{mso-style-name:"Plain Text Char";
mso-style-priority:99;
mso-style-link:"Plain Text";
font-family:"Calibri","sans-serif";}
.MsoChpDefault
{mso-style-type:export-only;
font-family:"Calibri","sans-serif";}
@page WordSection1
{size:8.5in 11.0in;
margin:1.0in 1.0in 1.0in 1.0in;}
div.WordSection1
{page:WordSection1;}
--></style><!--[if gte mso 9]><xml>
<o:shapedefaults v:ext="edit" spidmax="1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext="edit">
<o:idmap v:ext="edit" data="1" />
</o:shapelayout></xml><![endif]--></head><body lang=EN-US link=blue vlink=purple><div class=WordSection1><p class=MsoPlainText>Dear John,<o:p></o:p></p><p class=MsoPlainText><o:p> </o:p></p><p class=MsoPlainText>You wrote:<o:p></o:p></p><p class=MsoPlainText style='margin-left:.5in'><span style='font-size:10.0pt'>Lenat made the mistake of assuming that you need to hand-code<o:p></o:p></span></p><p class=MsoPlainText style='margin-left:.5in'><span style='font-size:10.0pt'>a huge amount of knowledge before a system can start to read<o:p></o:p></span></p><p class=MsoPlainText style='margin-left:.5in'><span style='font-size:10.0pt'>by itself. But that's wrong. You need to design a system that<o:p></o:p></span></p><p class=MsoPlainText style='margin-left:.5in'><span style='font-size:10.0pt'>can automatically augment its ontology every step of the way.<o:p></o:p></span></p><p class=MsoPlainText style='margin-left:.5in'><span style='font-size:10.0pt'><o:p> </o:p></span></p><p class=MsoPlainText style='margin-left:.5in'><span style='font-size:10.0pt'>John<o:p></o:p></span></p><p class=MsoPlainText><o:p> </o:p></p><p class=MsoPlainText>Yes, I agree; the system has to build up knowledge by interacting with its user(s), including knowledge about words and word usage. But that isn’t Lenat’s only miss. From his thesis on, he has been writing about discovery systems, not about systems that learn from empirical evidence. And of course there are those who think that embodied systems are needed before they can learn like people do. <o:p></o:p></p><p class=MsoPlainText><o:p> </o:p></p><p class=MsoPlainText>-Rich<o:p></o:p></p><p class=MsoPlainText><o:p> </o:p></p><p class=MsoPlainText>Sincerely,<o:p></o:p></p><p class=MsoPlainText>Rich Cooper<o:p></o:p></p><p class=MsoPlainText>EnglishLogicKernel.com<o:p></o:p></p><p class=MsoPlainText>Rich AT EnglishLogicKernel DOT com<o:p></o:p></p><p class=MsoPlainText>9 4 9 \ 5 2 5 - 5 7 1 2<o:p></o:p></p><p class=MsoPlainText><o:p> </o:p></p><p class=MsoPlainText>-----Original Message-----<br>From: corpora-bounces@uib.no [mailto:corpora-bounces@uib.no] On Behalf Of John F Sowa<br>Sent: Thursday, August 07, 2014 9:11 PM<br>To: corpora@uib.no<br>Subject: Re: [Corpora-List] WordNet vs Ontology</p><p class=MsoPlainText><o:p> </o:p></p><p class=MsoPlainText>On 8/7/2014 10:57 PM, Ken Litkowski wrote:<o:p></o:p></p><p class=MsoPlainText>> It would seem to me that our goal should be a classification<o:p></o:p></p><p class=MsoPlainText>> of all existing things (not to exclude the narrower types).<o:p></o:p></p><p class=MsoPlainText><o:p> </o:p></p><p class=MsoPlainText>Yes, but note the slides I suggested in my first note:<o:p></o:p></p><p class=MsoPlainText><o:p> </o:p></p><p class=MsoPlainText> <a href="http://www.jfsowa.com/talks/kdptut.pdf"><span style='color:windowtext;text-decoration:none'>http://www.jfsowa.com/talks/kdptut.pdf</span></a><o:p></o:p></p><p class=MsoPlainText><o:p> </o:p></p><p class=MsoPlainText>Slides 7 to 9: Cyc project. 30 years of work (since 1984).<o:p></o:p></p><p class=MsoPlainText>After the first 25 years, 100 million dollars and 1000 person-years<o:p></o:p></p><p class=MsoPlainText>of work (one person-millennium!), 600,000 concepts, defined by<o:p></o:p></p><p class=MsoPlainText>5,000,000 axioms, organized in 6,000 microtheories -- and counting.<o:p></o:p></p><p class=MsoPlainText><o:p> </o:p></p><p class=MsoPlainText>Slide 10: 2300 years of universal ontology schemes -- and counting.<o:p></o:p></p><p class=MsoPlainText><o:p> </o:p></p><p class=MsoPlainText>> The Brandeis Shallow Ontology attempts to do this, and incidentally<o:p></o:p></p><p class=MsoPlainText>> is being used to characterize arguments of verbs in Patrick Hanks<o:p></o:p></p><p class=MsoPlainText>> corpus pattern analysis, i.e., in the imperfect world of language.<o:p></o:p></p><p class=MsoPlainText><o:p> </o:p></p><p class=MsoPlainText>I strongly believe in shallow, underspecified ontologies -- especially<o:p></o:p></p><p class=MsoPlainText>when they're supplemented with lots of lexical information about verbs<o:p></o:p></p><p class=MsoPlainText>and their characteristic patterns.<o:p></o:p></p><p class=MsoPlainText><o:p> </o:p></p><p class=MsoPlainText>But I also believe that the key to having an open-ended variety of<o:p></o:p></p><p class=MsoPlainText>specialized ontologies is to make the computers do what people do:<o:p></o:p></p><p class=MsoPlainText>extend their ontologies automatically by reading books.<o:p></o:p></p><p class=MsoPlainText><o:p> </o:p></p><p class=MsoPlainText>Lenat made the mistake of assuming that you need to hand-code<o:p></o:p></p><p class=MsoPlainText>a huge amount of knowledge before a system can start to read<o:p></o:p></p><p class=MsoPlainText>by itself. But that's wrong. You need to design a system that<o:p></o:p></p><p class=MsoPlainText>can automatically augment its ontology every step of the way.<o:p></o:p></p><p class=MsoPlainText><o:p> </o:p></p><p class=MsoPlainText>John<o:p></o:p></p><p class=MsoPlainText><o:p> </o:p></p><p class=MsoPlainText>_______________________________________________<o:p></o:p></p><p class=MsoPlainText>UNSUBSCRIBE from this page: <a href="http://mailman.uib.no/options/corpora"><span style='color:windowtext;text-decoration:none'>http://mailman.uib.no/options/corpora</span></a><o:p></o:p></p><p class=MsoPlainText>Corpora mailing list<o:p></o:p></p><p class=MsoPlainText><a href="mailto:Corpora@uib.no"><span style='color:windowtext;text-decoration:none'>Corpora@uib.no</span></a><o:p></o:p></p><p class=MsoPlainText><a href="http://mailman.uib.no/listinfo/corpora"><span style='color:windowtext;text-decoration:none'>http://mailman.uib.no/listinfo/corpora</span></a><o:p></o:p></p></div></body></html>