See also: IRC log
<egonw> moin
http://www.ncbi.nlm.nih.gov/pubmed/22449719
<ericP> test
<egonw> pong
<mikel_egana> that's mikel
http://www.ncbi.nlm.nih.gov/pubmed/22449719
Processors could, for example, offer a virtual SPARQL endpoint over the
mapped relational data, or generate RDF dumps, or offer a Linked Data
interface.' Michael and Simon pointed out that the use of the word interface can be confusing here.
Scott: Integrated the Translating
Standards article (pubmed URI above) into an answer to Question
6: Does all data need to be mapped to the same ontologies in
order to achieve interoperability?
... can read a draft of that in the html attachment.
... can read a draft of that in the html attachment sent with
the announcement today.
<Simon asks about the mapping using SWObjects, Scott, then Eric P. answer>
<michel> greetings
Eric: The value of linked data to
a SPARQL user is that it provides a good mechanism for people
to reuse identifiers when they should.
... Linked data can provide information about a URI term by
just pasting it into their browser.
<egonw> agreed...
Eric: Linked data means I can do a GET on any of the URLs in it.
<egonw> dereferencable versus sparql
<egonw> right, and the GET returns at least RDF (in some form) but could also return HTML (e.g. depending on content negotiation)
I think that LD people will insist on the LD returning RDF (that is possibly rendered for humans).
Eric explains what is meant by the term "Linked Data" - doesn't necessarily mean "connected data", but refers to the http behavior of links (see description of GET above).
mikel - seems like we can hear you typing.
mikel - seems like we can hear you typing - please mute.
<egonw> ummm... likes rather than needs...
<mikel_egana> sorry :-(
no prob, can't hear it anymore :)
<michel> i think you're both right.
<michel> linked data is both providing data on the web, and creating links between previously unlinked data
<egonw> re Q11: there the thing was in the paper, AFAIK, about whether you could 'spider' it?
<egonw> re Q11: there the thing was in the paper, AFAIK, about whether you could 'spider' it
<egonw> SPARQL end points cannot be spidered
<egonw> because resources cannot be pointed to in a SPARLQ end point
<egonw> breaking the web of LOD
Thanks Egon. Bingo. Yes, we did.
<michel> egon, you can easily setup an web interface over a SPARQL endpoint
<michel> and it can then be spidered
I think that those are two different types of spidering (or crawling). One uses GET, the other uses SPARQL queries, non?
<egonw> michel: indeed, but that would not be a SPARQL end point anymore
<egonw> but a linked data layer *in addition* to...
SPARQL Select results are tabular and could result from querying linked data or a SPARQL endpoint without Linked Data (TM).
<ericP> SPARQL SELECT returns the results of searching over an RDF graph which ideally complies with the Linked Data Principles
<ericP> http://www.w3.org/DesignIssues/LinkedData.html
<egonw> btw, SPARQL can also host RDF which is 100% incompatible with linked data. E.g. this triple: http://localhost:8091/foo/bar owl:sameAs http://localhost:8099/foo/cafe
<egonw> :)
Scott (at beginning of call): noticed that our correction to have the CKAN - The Data Hub reference point to http://thedatahub.org were lost somewhere..
<egonw> yes (I think it will be hard for our audience anyway... that's at least my observation in Open PHACTS)
<egonw> thanx all
<mikel_egana> thx
<egonw> yes
<mikel_egana> bye
bye
<egonw> last time for me was at SARA/CWI in 2009 ?
<scribe> ACTION: Eric to add a bit to intro defining Linked Data vs. SPARQL [recorded in http://www.w3.org/2012/10/01-hcls-minutes.html#action01]
<scribe> ACTION: Michael to add a piece to Q6 mentioning the situation where RDF data have some unplanned overlap (example given was disease and drug). [recorded in http://www.w3.org/2012/10/01-hcls-minutes.html#action02]
<scribe> ACTION: Scott will keep adding straws to his strawman extension to Q6 (incorporating the mapping approach described in "Translating.."). [recorded in http://www.w3.org/2012/10/01-hcls-minutes.html#action03]
This is scribe.perl Revision: 1.137 of Date: 2012/09/20 20:19:01 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: mscottm Inferring Scribes: mscottm WARNING: No "Topic:" lines found. Default Present: +1.510.705.aaaa, +31.62.427.aabb, +1.715.389.aacc, +1.206.732.aadd, EricP, [IPcaller], +1.603.727.aaee, +34.65.672.aaff, mikel_egana, +1.613.293.aagg, +31.62.427.aahh Present: +1.510.705.aaaa +31.62.427.aabb +1.715.389.aacc +1.206.732.aadd EricP [IPcaller] +1.603.727.aaee +34.65.672.aaff mikel_egana +1.613.293.aagg +31.62.427.aahh 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 Got date from IRC log name: 01 Oct 2012 Guessing minutes URL: http://www.w3.org/2012/10/01-hcls-minutes.html People with action items: adding eric keep michael scott straws will 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[End of scribe.perl diagnostic output]