Harold showing Naitonal Data Resource Catalog Prototype

scribe: We have created in such a form, we are creating 11179 records
... loaded in as set of classes and relationship properties

Scott: If you can export RDF, how far is it away from exporting?

Harold: Would be good to convince semantic media wiki people

Scott: So this leads us to our next talk, SPARQL update

Harold: An earlier example, we used Prodigy to do a pull
... and then fed back to wiki

Eric: I have lots of questions...for later

Nagim: So how well does this scale?

Harold: Very well

Nagim: A solution chokes out at about a hundred million

Harold: Media wiki has to built on relational
... getting rest of community behind it
... to start writing SPARQL queries on back end
... If they are story queries in relational, just do the conversion
... I think that's how Simile does it

Scott: Because that would scale

EricP: I would like to welcome the SPARQL Working Group
... SPARQL is a spec that came out nearly two years ago
... it is one of the primary RDF query languages now; it is a stable standard
... First version didn't include some features
... Some of those are coming around this time
... What we have to do
... Like "not exist"

[Eric stands in front of screen]

Eric explains chart of clinical trial patient in a study

scribe: write "optional, not found"
... So if I want to find a patient and what type of medication he is taking
... So like SQLs grouped by...
... function that you can say, project out the some as some new variable total

Michael: If patient had two dates of birth?
... SQL does not allow that

Lee: We discussed that you will not be allowed ot project out variables

Michael: DoB

Lee: Create in aggregate function
... or if you wanted every DoB then include as part of group key

EricP: Sub queries
... looking for some patient taking some substance administration
... Some subselect where I'm finding their contra indications
... that disqualify them
... and I'm only taking one of them

Lee; Yes, exactly like sample

Eric: Concatonating
... creating contraURL
... Property paths
... little star is complicated

[Karen invites another scribe, please]

<LenaDeus> ericP: [slide 9] using "Service" as a url to SPARQL

<LenaDeus> the exact mathematical operation is executed with data coming from the SPARQL services as the one used to answer the query being executed

<LenaDeus> Kei: how to issue a query to find a node with the most connectivity?

<LenaDeus> lee: group by protein and do the counting operation using aggregates

<LenaDeus> [slide 10]

<LenaDeus> graph management protocol

<LenaDeus> SPARQL update: just like a construct, but inserting into a graph

<LenaDeus> (same for delete)

<LenaDeus> service description: each service describes itself based on teh version of the sparql protocol

<LenaDeus> each service can also specify which functions are accepted (for example reasoning)

<LenaDeus> [slode 13]

<LenaDeus> D2R service description:

<LenaDeus> using PUT <request_uri> ... RDF payload ... would be equivalente to putting a graph in a certain URI

<LenaDeus> [slide 15] given 2 models (2 universes) where the inference leads to two different assertions

<LenaDeus> using blank nodes in the query, both viable models can be queried

<LenaDeus> (a query that operates over the logical intersection)

<LenaDeus> [slide 16] inference

<LenaDeus> inference should be used sparringly

<LenaDeus> [slide 17]

<LenaDeus> michael: XOR?

<LenaDeus> RIF: the idea was to cover a large number of rule languages such that the engine could be agnostic to the specifici rule language beign user

<LenaDeus> RIF was not desgined to be a universal rule language

<LenaDeus> RIF is n-ary

<LenaDeus> there is a syntax in RIF for triples

<LenaDeus> rif was designed to be extensible

<LenaDeus> RDFS allows type derivation; owl allows much more inference of types, including identity

<LenaDeus> in owl, there can be multiple represetnations of the same individual (even with different URI)

<LenaDeus> RDFS gives us "cheap" inference

<LenaDeus> these inferences do not validate

<LenaDeus> rdfs is a language that allows you to learn more things

<LenaDeus> however, it does not tell me that a mistake has been made

<LenaDeus> when owl is translated to rdf, it reasults in a long verbose syntax

<LenaDeus> [slide 27] there are therefore reasons why people will not want to query the RDF form of this information

<LenaDeus> owl 1.1 allows some syntactic sugar

<LenaDeus> owl 2 has lots of cardinality

<LenaDeus> annotation properties produce no inferences

<LenaDeus> punning: a property that belongs to a specific individual belonging to a class but does not belong to the class

<LenaDeus> POWDER: define a class of things

This is scribe.perl Revision: 1.135  of Date: 2009/03/02 03:52:20  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

No ScribeNick specified.  Guessing ScribeNick: Karen_
Inferring Scribes: Karen_

WARNING: No "Topic:" lines found.

Default Present: +1.610.651.aaaa, +1.408.988.aabb, Yong, matthias_samwald, Susie, HCLSf2f, jun, ElgarPichler, andreassplendiani, Jonas
Present: +1.610.651.aaaa +1.408.988.aabb Yong matthias_samwald Susie HCLSf2f jun ElgarPichler andreassplendiani Jonas

WARNING: No meeting title found!
You should specify the meeting title like this:
<dbooth> Meeting: Weekly Baking Club Meeting


WARNING: No meeting chair found!
You should specify the meeting chair like this:
<dbooth> Chair: dbooth


WARNING: No date found!  Assuming today.  (Hint: Specify
the W3C IRC log URL, and the date will be determined from that.)
Or specify the date like this:
<dbooth> Date: 12 Sep 2002

Guessing minutes URL: http://www.w3.org/2009/11/03-hcls-minutes.html
People with action items: 

WARNING: No "Topic: ..." lines found!  
Resulting HTML may have an empty (invalid) <ol>...</ol>.

Explanation: "Topic: ..." lines are used to indicate the start of 
new discussion topics or agenda items, such as:
<dbooth> Topic: Review of Amy's report


WARNING: IRC log location not specified!  (You can ignore this 
warning if you do not want the generated minutes to contain 
a link to the original IRC log.)