W3C home > Mailing lists > Public > public-semweb-lifesci@w3.org > September 2007

Re: [hcls] User interfaces for writing / querying RDF: Leeet

From: Chris Mungall <cjm@fruitfly.org>
Date: Wed, 12 Sep 2007 02:11:43 +0100
Message-Id: <169EF44C-1465-4CA2-8891-CE033EF8B1D5@fruitfly.org>
Cc: public-semweb-lifesci <public-semweb-lifesci@w3.org>
To: Matthias Samwald <samwald@gmx.at>

Hi Matthias

This is very nice. Not sure about the name though. Shouldn't it be  
L33TSP34K?

What group are you targeting? SPARQL syntax isn't bad enough to put  
off hardcore data miners. And for the less techie folks I think the  
main stumbling block isn't so much the syntax as the semantics.

For example, those of us with a knowledge of OWL, ABoxes, TBoxes,  
entailment and reasoning know why

	CL:astrocyte develops_from ?

has an empty result set, and fails to return glioblast, NC cell,  
neurectodermal cell... But I'm not sure I could explain this to the  
average user in a succinct way. Anyone who can understand the  
explanation is more than capable of handling SPARQL syntax.

Note that it's impossible to answer the intended query above without  
SPARQL-DL - and the most intuitive syntax for this kind of query in  
SPARQL-DL may not be triple-based, cosmeticised or not. E.g. "ALL  
astrocyte develops_from SOME ?"

I am heavily biased towards TBox queries - for ABox queries, a  
syntactic patina over SPARQL may be very welcome.

On Sep 7, 2007, at 11:13 AM, Matthias Samwald wrote:

>
> It seems to be general consensus that the lack of good user interfaces
> for authoring / querying RDF and OWL is one of the major  
> difficulties we
> are facing at the moment. Solving this problem should be one of our  
> top
> priorities for extending the 'HCLS Demo' (besides improving the
> ontologies and adding new data sources, of course).
>
> The following is a prototype of a simple, easily deployable GUI  
> based on
> Flash and Sparql endpoints:
>
> http://neuroscientific.net/leeet/
>
> While the user is typing, the tool offers a list of auto-complete
> options, based on Sparql queries to one or more Sparql endpoints.  
> Scroll
> down the page to find some examples and documentation.
>
> The tool uses a simple RDF syntax called 'Leeetspeak'. Here is an
> example of simple biological statements (two RDF triples):
> ----
> acetylcholine                 <http://example.org/rdf#ach
>   binds to                    <http://example.org/rdf#binds
>     acetylcholine receptor    <http://example.org/rdf#ach_receptor
>     acetylcholinesterase      <http://example.org/rdf#ache
> ----
>
>
> Example of a simple Sparql query (add a third '?' in the Leeet console
> to initiate query):
> ----
> exocrine cell         <http://example.org/celltype#CL_0000152
>   is a                <urn:leeet:is_a
>     ??
> ---
>
> The user face is deliberately focused on typing text instead of fancy
> graphics and menus, because this can speed up the creation of RDF
> tremendously. Rearranging / templating is just a matter of copying and
> pasting text.
>
> I guess such a tool surely is still too 'technical' for most  
> people, but
> it could at least attract the interest of researchers with a  
> background
> in informatics (e.g. bioinformaticians, computational biologists  
> etc.).
>
> If anyone wants to use this tool for his/her project or would be
> interested in somehow supporting its further development, please  
> reply!
>
> cheers,
> Matthias Samwald
>
>
>
>
>
>
Received on Wednesday, 12 September 2007 01:12:00 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 10 December 2014 20:09:36 UTC