Reverse Template Engine?

[Dan Brickley-2]

Is the TAP RTE system still used, developed? I found a link to
it on kb.alpiri.com but that site now seems dead...

What's new in TAP this year?

Dan

[RobM-2]

> Is the TAP RTE system still used, developed? I found a link to
> it on kb.alpiri.com but that site now seems dead...
>
> What's new in TAP this year?

The short answers are, no, no, and not much. TAP is in maintenance mode
at Stanford, since Guha went to Google in 2005 and I went to Yahoo later
that year.

The RTE system in particular was replaced by a script-based system that
resembled LISP. The newer system was much faster and more flexible than
RTE turned out to be. Nonetheless, scale never went over 30-40 sites,
though the RDF itself grew to the limit of what you could load on a
1.5GB machine even when compressed.

The semantic search demo on http://sp11.stanford.edu/ is likely to be
the final iteration of TAP as a system.

[Dan Brickley-2]

* robm@... <robm@...> [2006-02-14 12:47-0800]
>
> > Is the TAP RTE system still used, developed? I found a link to
> > it on kb.alpiri.com but that site now seems dead...
> >
> > What's new in TAP this year?
>
> The short answers are, no, no, and not much. TAP is in maintenance mode
> at Stanford, since Guha went to Google in 2005 and I went to Yahoo later
> that year.

Yeah, I could see that inhibiting collaboration a little...

> The RTE system in particular was replaced by a script-based system that
> resembled LISP. The newer system was much faster and more flexible than
> RTE turned out to be. Nonetheless, scale never went over 30-40 sites,
> though the RDF itself grew to the limit of what you could load on a
> 1.5GB machine even when compressed.

I'm looking at possibility of extractors for medical/health content
on the Web. Did the code for the RTE successor ever get released?


> The semantic search demo on http://sp11.stanford.edu/ is likely to be
> the final iteration of TAP as a system.

How about the management of the core TAP KB? Do stanford have plans
for it? Or should it be considered abandonware of some kind?

cheers,

Dan

[RobM-2]

> I'm looking at possibility of extractors for medical/health content
> on the Web. Did the code for the RTE successor ever get released?

Yes, it's downloadable here:

http://sp11.stanford.edu/getinvolved.html

It's Java based. There is text documentation talking about the clauses
that are available to the scrapers, i.e. what does (get-targets ...)
mean. If you don't find those documents, let me know and I'll poke
around. There was a student interested in the system so I wrote
documentation for him.

> > The semantic search demo on http://sp11.stanford.edu/ is likely to be
> > the final iteration of TAP as a system.
> How about the management of the core TAP KB? Do stanford have plans
> for it? Or should it be considered abandonware of some kind?

The license is pretty flexible, so I don't think it would be a problem
if you or someone picked it up. Stanford may have some plans for it, a
good person to ask is selene@....