Solvent and combining sources

From: Michael McDougall <mcdougall_at_grammatech.com>
Date: Thu, 17 Nov 2005 15:53:03 -0500

I've been playing with Solvent and trying to get scrapings from
different pages listed on the same map.
Here's some feedback and a question.

Issues
- In the Solvent UI, make it easier to set the 'type' of the node you
are creating. Make this at least as easy as associating an XPath
expression with 'address' -- put a big text box at the top of the right
hand pane saying "TYPE" or something.
- I couldn't get the "process text further" thing to do anything. It
complained about Piggy Bank not being installed when it actually was
installed.
- Solvent doesn't seem to be aware of multiple tabs or windows. This
gets confusing when you are trying to scrape 2 tabs at the same time. I
would like to be able to write a scraper on one tab, switch tabs, write
another scraper, and then switch back and forth so that each tab
preserves the javascript generated for that tab.

Question: I want to combine data from different pages in a single map.
It's easy to dump everything into Piggy Bank and then map it, but I want
to have data from one page be blue and data from another page to be red
(like David's demo at ISWC). I couldn't figure this out. I would create
an n3 page (using Solvent), click the RDF coin, click back, track the
page (Ctrl-shift-T and assign it a name), repeat for the other page, and
then select the "combine..." menu item but I just get a page saying " No
typed data found." Any ideas? What's the exact sequence of commands I
need to use to combine data like this?



Michael
Received on Thu Nov 17 2005 - 20:46:58 EST