On 2/11/15 9:52 AM, John F Sowa wrote:
> Kingsley,
>
> I agree that Nanonotation is a good humanly readable notation, and I
> realize that current NLP tools still have a long way to go. I also
> admit that I do all my word processing by typing HTML (including tags)
> with an ASCII editor. (I use OpenOffice or LibreOffice to produce PDF.) (01)
Yes, so do I. (02)
I really think the industry has generally overlooked the fundamental
file create, save, and share pattern [1]. Net effect, the notion of a
tractable read-write web remains mercurial to comprehend at best. (03)
I use Nanotation as a means of demonstrate how the following is now
possible, at web-scale: (04)
1. Commentary about anything
2. Read-Write against commentary
3. Data creation and sharing in general. (05)
>>> That's a rather large amount of typing. Do you have automated tools
>>> to generate those annotations?
>> Yes, but I also see value in typing i.e., moderate automation as opposed
>> to over automation. In short, there's a meme emerging called "Slow Web"
>> [1] which actually sheds light on this issue.
> I agree that we need a lot of human thought and reflection to make
> sense of the overwhelming volume of data on the WWW. (06)
Yep! The "Fast Web" doesn't encourage "Critical Thinking" hence the
exponential growth of noise on the Web. (07)
> No automated
> tools will be able to replace human analysis and thought for a
> long, long time. (08)
Yep! Thus, we have to orient towards tools that are as close to natural
language sentence construction as possible. It is this line of though
(thanks to a lot of your writing and knowledge sharing) that lead me to
Nanotation. (09)
>
>>> As a general principle, I don't trust people to produce structured
>>> data of any kind -- unless they're well trained and have a very
>>> strong motivation.
>> isn't that like saying: I don't trust people to be able to express
>> themselves using sentences?
> No. It's the observation that only a tiny percentage of programs
> run correctly on the first try. Declarative specifications or
> annotations are just as error prone. (010)
And we know errors are features that will live with humanity forever.
Thus, we shouldn't let their prevalence impede tool development progress . (011)
>
> Successful communication in NLs always requires a dialog. If you
> say something to another person and the only response is "OK",
> it's unlikely that they'll do what you expect or hope. (012)
I agree. That's why we all need to be able to contribute commentary
about anything, in a manner native to the Web.
>
>> but isn't error-free data just a variation of error-free ontologies?
>> We all know these do not really exist, so we have to live with
>> somewhat-accurate data, viewed through a variety of context-lenses
> There are two distinct issues: reliable sources and reliable conversion
> of data from any source to the finished presentation.
>
> For the second point, businesses in the 1960s required critical data
> to be keypunched and verified by *two* typists.
>
> (See http://www.computerculture.org/2012/05/punched-card-verifiers/ )
>
> For the first point, a universal ontology would only be possible
> if everybody in the world had exactly the same point of view on
> every possible issue. (013)
The universal ontology (as you know) is logic, everything else boils
down to "context lenses" through which we attempt to make and share
experience. (014)
> But as we all know, you can't find two
> people who agree on everything -- or even one person who agrees
> with everything he or she said yesterday. (015)
Exactly! And that's a good thing -- heterogeneity is the spice of life :) (016)
Links: (017)
[1] http://www.shirky.com/writings/view_source.html -- View Source...
Lessons from the Web's massively parallel development. (018)
Kingsley
>
> 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
>
>
> (019)
--
Regards, (020)
Kingsley Idehen
Founder & CEO
OpenLink Software
Company Web: http://www.openlinksw.com
Personal Weblog 1: http://kidehen.blogspot.com
Personal Weblog 2: http://www.openlinksw.com/blog/~kidehen
Twitter Profile: https://twitter.com/kidehen
Google+ Profile: https://plus.google.com/+KingsleyIdehen/about
LinkedIn Profile: http://www.linkedin.com/in/kidehen
Personal WebID: http://kingsley.idehen.net/dataspace/person/kidehen#this (021)
smime.p7s
Description: S/MIME Cryptographic Signature
_________________________________________________________________
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)
|