Jason Foster wrote:
> Having chosen RDF over Topic Maps for organizing my research (thanks
> Stefano!) I'm in the process of setting up my system. I'm hoping that
> you can help me get started converting my resources into Simile-
> compatible forms, in exchange for the promise of my writing a summary
> "Getting Started" document for the wiki.
Awesome! that would be very useful.
> I've downloaded and built Longwell (which should mention somewhere that
> by default you have to be online to build) and am working on Haystack.
> My impression from browsing the Simile website is that the current
> focus is on read-only display of existing data.
For Simile, yes, we have not touched the editing side of the story
(yet), but haystack does.
Note, however, that if we get the funding that we are asking for, we
would be moving in the editing direction.
> Given that all I have
> right now are handwritten research notes, some vaguely- structured text
> files, and an EndNote database, my next step seems to be coming up with
> a way to add information to an RDF store.
>
> The Longwell documentation lists 5 different RDF stores and provides
> configuration information for each. Unfortunately (for me) I don't
> have a feel for which store makes the most sense. I have a sneaking
> suspicion that I should look at DSpace, but a cursory look hasn't
> revealed any documentation on integrating Longwell with DSpace.
DSpace is not a triples store (well, it contains one for its use of the
history system but that's not connected to the rest of the data and
therefore is pretty much useless for our purposes).
Simile's objective is to work with the dspace project to find a way to
enable arbitrary metadata, but how this is going to be possible is both
a technical and political issue.... and we are not sure ourselves of
what is the best option out there.
> If you have a minute, can you outline the steps that a new "RDF
> Convert" would follow to incorporate their research notes, web
> bookmarks, z39.50 results, etc., into a browsable *and update-able* RDF
> store?
Hmmm, honestly, nobody knows what is the best way... and I suspect
nobody will know for a while, until we try out things and find out if
they work on not.
The beauty of RDF versus XML, for example, is the mixability of the
data. If you use things like SRW or an XML representation of a x39.50
result, you end up with xml data that you can't just pour together in
the single container and, for example, browse facetted.
So, one solution is to do something like
http://www.marcont.org/mediations.jsp
(which is for MARC records) but I'm not sure what would be the best
architecture for mediation, trasnformation and aggregation of external
services.
> Thanks for your help, and for providing the only open source, facet-
> based browser I've been able to find!
You are welcome.
--
Stefano Mazzocchi
Research Scientist Digital Libraries Research Group
Massachusetts Institute of Technology location: E25-131C
77 Massachusetts Ave telephone: +1 (617) 253-1096
Cambridge, MA 02139-4307 email: stefanom at mit . edu
-------------------------------------------------------------------
Received on Mon Aug 15 2005 - 18:09:06 EDT