[Date Prev] [Date Next] [Thread Prev] [Thread Next] Indexes: Main | Date | Thread | Author

Re: [ba-ohs-talk] rough draft of graph model paper


Oh boy, Eugene, it is evident that you can come up with brilliant ideas faster
than I can properly grasp them!    (01)

I read your paper, then decided to print it out for greater comfort and
increased rate of comprehension - as it is written in your doc's 1D (06).    (02)

But possibly more interesting to you and, I guess, quite relevant, is the
content of http://ojr.usc.edu/content/story.cfm?request=690    (03)

It is about news items written purely by computer, a technology developed at
Columbia University. The program extracts a coherent story from a variety of
sources detected on the Web. The method is outlined on the site.    (04)

Reason I bring this to your attention, it seems to me quite possible that a
message on a discussion forum and the responses it creates (and the responses
to the responses)  might by that technique be condensed into a single story.    (05)

Of course, those responses that have wandered off topic put some dirt in the
endproduct. Here is an example of boiling down a story from disparate
materials:    (06)

"The storm had given the Marquette area 17.6 inches of snow by Sunday
afternoon. Israeli troops shot dead three Palestinians during a raid in the
Gaza Strip and  stormed into a West Bank. A late-winter storm cruising through
the Midwest on  Saturday snarled airport traffic." (Example from The Economist)    (07)

But this mishap aside, things normally seem to work out quite well. Samples on
this site:  http://www.cs.columbia.edu/nlp/newsblaster/    (08)

The Columbia group have been working on this thing, called Newsblaster, for
about two decades and keep on refining it. They might be interested in  seeing
how well it will function with discussion threads.    (09)

Just a dumb thought.    (010)

Henry    (011)

Eugene Eric Kim wrote:    (012)

> I just posted a very rough, incomplete draft of a white paper entitled,
> "Interoperability Between Collaborative Knowledge Applications: Towards a
> Graph-Based Data Model for the Open Hyperdocument System."  It covers:
>
>   - motivation for a graph-based data model
>
>   - examples -- threaded forum apps, dialog maps
>
>   - discussion of existing graph-based modeling languages
>
> The draft is at:
>
>   http://www.eekim.com/ohs/papers/graphmodel/
>
> I would really appreciate feedback.  (As always, use the purple numbers
> for reference.)  Thanks!
>
> -Eugene
>
> --
> +=== Eugene Eric Kim ===== eekim@eekim.com ===== http://www.eekim.com/ ===+
> |       "Writer's block is a fancy term made up by whiners so they        |
> +=====  can have an excuse to drink alcohol."  --Steve Martin  ===========+    (013)