16:00:13 <RRSAgent> RRSAgent has joined #rdb2rdf
16:00:13 <RRSAgent> logging to http://www.w3.org/2010/08/10-rdb2rdf-irc
16:00:30 <Ashok> zakim, this will be RDB2RDF
16:00:30 <Zakim> ok, Ashok, I see SW_RDB2RDF()12:00PM already started
16:00:32 <Zakim> +[IPcaller]
16:00:37 <Ashok> chair: Ashok
16:00:45 <Ashok> meeting:  RDB2RDF
16:01:08 <Ashok> regrets: Boris, Marcelo
16:01:10 <Zakim> +soeren
16:01:36 <Zakim> +[IPcaller.a]
16:01:42 <Souri> Souri has joined #rdb2rdf
16:01:49 <juansequeda> Zakim, IPCaller.a is me
16:01:49 <Zakim> +juansequeda; got it
16:02:24 <Zakim> +OpenLink_Software
16:02:28 <Zakim> +??P11
16:02:30 <cygri> Zakim, IPcaller is me
16:02:30 <Zakim> +cygri; got it
16:02:32 <Zakim> +Ashok_Malhotra
16:02:36 <MacTed> Zakim, OpenLink_Software is temporarily me
16:02:36 <Zakim> +MacTed; got it
16:02:37 <hhalpin> Zakim, ??P11 is hhalpin
16:02:37 <Zakim> +hhalpin; got it
16:02:38 <MacTed> Zakim, mute me
16:02:39 <Zakim> MacTed should now be muted
16:02:42 <MacTed> Zakim, who's here?
16:02:42 <Zakim> On the phone I see Souri, cygri, soeren, juansequeda, MacTed (muted), hhalpin, Ashok_Malhotra
16:02:44 <Zakim> On IRC I see Souri, RRSAgent, soeren, Zakim, Ashok, hhalpin, boris, juansequeda, cygri, LeeF, MacTed, iv_an_ru, ericP, trackbot
16:02:54 <hhalpin> Zakim, mute me
16:02:54 <Zakim> hhalpin should now be muted
16:03:19 <hhalpin> Zakim, unmute me
16:03:19 <Zakim> hhalpin should no longer be muted
16:03:23 <hhalpin> Zakim, mute me
16:03:23 <Zakim> hhalpin should now be muted
16:03:32 <Zakim> +LeeF
16:04:11 <Ashok> scribenick: hhalpin
16:04:24 <Ashok> scribe: Harry
16:04:54 <Ashok>  PROPOSAL: Accept the minutes of last meeting, see > http://www.w3.org/2010/08/03-rdb2rdf-minutes.html >
16:06:14 <ericP> Zakim, please dial ericP-office
16:06:14 <Zakim> ok, ericP; the call is being made
16:06:14 <hhalpin> Ashok: The plan of the meeting will be for Souri to finish the proposal, and then address comments.
16:06:15 <Zakim> +EricP
16:06:32 <Souri> comments on Richard's comments: http://lists.w3.org/Archives/Public/public-rdb2rdf-wg/2010Aug/0013.html
16:06:41 <hhalpin> ... I think we have a reasonable agreement on the SQL-based approach, so we should see the next steps
16:08:14 <hhalpin> Souri: Would prefer to go over the comments in detail first.
16:08:19 <hhalpin> cygri: that's fine
16:08:43 <hhalpin> topic: Comments on SQL-based approach 
16:09:09 <hhalpin> cygri: I have a number of nonfunctional improvements
16:09:18 <juansequeda> juansequeda has joined #RDB2RDF
16:09:18 <hhalpin> ... and then some expressivity requirements
16:09:34 <hhalpin> ... then some things for author convenience
16:09:47 <hhalpin> cygri: 1.1 don't include URI in pointed brackets
16:10:06 <hhalpin> ... souri seems to agree
16:11:08 <hhalpin> 1.2 a renaming away from property URI to instance URI map
16:11:19 <juansequeda> +1 for InstanceURIMap
16:11:20 <hhalpin> cygri: I would prefer it to be called URI instance map
16:11:23 <hhalpin> souri: I agree
16:12:45 <hhalpin> cygri: 1.3 URI graph name
16:13:38 <hhalpin> souri: let me explain, for each column, the graph URI property name is one way of identifying the graph with the column. 
16:13:48 <hhalpin> ... we could avoid it as in previous treatment
16:14:12 <hhalpin> ... but richard points out you could have more property maps that generate the same names
16:14:23 <hhalpin> ... so we should need to keep the graph name
16:14:40 <hhalpin> ... for each property name we should have a backpointer that explains which graph URI name it goes with
16:14:51 <hhalpin> ... but once you have that, you really a name for the graph property
16:15:56 <hhalpin> cygri: it appears that I may have been confused
16:16:02 <hhalpin> ... I thought it referred to a RDF property
16:16:10 <hhalpin> ... but this may not have been your attention
16:16:26 <hhalpin> ... but what you just explained, that a property mapping has a mapping to a graph URI mapping
16:16:35 <hhalpin> ... then that's a design that I would be happy with.
16:16:51 <hhalpin> juan: Could you give us an example?
16:17:01 <Ashok> present: Ashok, Richard, EricP, Harry, Ivan, Juan, LeeF, Ted, Soeren, Souri
16:17:05 <Souri> http://www.w3.org/2001/sw/rdb2rdf/wiki/Revised_Example_of_SQL-Query_based_Approach
16:19:54 <hhalpin> juan: so the semantics of that are specified inside the URI?
16:20:33 <hhalpin> Zakim, who's making noise?
16:20:46 <Zakim> hhalpin, listening for 10 seconds I heard sound from the following: Souri (47%), soeren (90%)
16:20:55 <hhalpin> zakim, mute soeren
16:20:55 <Zakim> soeren should now be muted
16:21:18 <hhalpin> Souri: It doesn't access the URI, it's just the name of the graph
16:21:34 <hhalpin> souri: so the syntax of the URI is not at all important.
16:22:51 <Zakim> -soeren
16:23:00 <hhalpin> cygri: my comment is that we can easily duplicate information
16:23:29 <hhalpin> ... using keymaps
16:23:29 <Zakim> +soeren
16:23:36 <hhalpin> ericP: I'm projecting out 2 or 3 tables
16:23:40 <hhalpin> ... if a particular column in one of the tables
16:23:46 <hhalpin> ... and that acts as a key
16:23:56 <hhalpin> ... in the regular database
16:24:21 <hhalpin> souri: it's a very simple approach, in fact, we've already simplified it in our prototype
16:24:27 <hhalpin> ... so we identify a foreign key
16:24:36 <hhalpin> ... and then the key sequence also references this primary key
16:24:48 <hhalpin> ... now, foreign key can be any expression
16:25:05 <hhalpin> ... so in general, all we need to know that these columns are available
16:25:16 <hhalpin> ... that are used in some expression (x+y+2)
16:25:24 <hhalpin> ... but I realized we do not need the whole key definition
16:25:27 <hhalpin> ... all we really need
16:25:33 <hhalpin> ... is the name of the foreign key
16:25:43 <hhalpin> ... an object property name
16:26:23 <hhalpin> ... so we need to translate that.
16:26:33 <hhalpin> ... all we really need is the corresponding join condition
16:26:44 <hhalpin> ... we need the corresponding consequent.
16:26:47 <hhalpin> ... we need a template.
16:27:19 <hhalpin> ... but if it's a five column key, then we have five conditions, and so all we need from the mapping specifier is that template.
16:27:27 <hhalpin> ... which gives us a join condition for the foreign key
16:27:35 <hhalpin> ... which allows us to work through the aliases
16:29:16 <hhalpin> ericP: So there are a class of functions that are reversible, but for those that aren't we have to materialize them
16:29:23 <hhalpin> souri: that's not the issue
16:30:21 <hhalpin> ... imagine if x is exposed as f(x), but only x is exposed, but if the foreign key is f(x)=d(y), where d may not be reversible, but all we need is the info that f(x)=d(y) as that needs to be put in join condition
16:32:00 <ericP> Souri has convinced me that reversibility is not required for creating a join constraint; the engine will be evaluating this on specific tuples
16:32:01 <hhalpin> cygri: I will study this more
16:32:09 <hhalpin> ... Second set of points
16:32:15 <hhalpin> ... we need to have blank nodes as subjects
16:32:34 <ericP> (of course, non-reversibility leads to table scans, but that's not our problem)
16:34:33 <hhalpin> cygri: different literals genreated from same property name could have different langauge tags
16:34:37 <hhalpin> souri: I agree, good point
16:35:48 <hhalpin> ... the question is can we have it in the expression that generate the literal value that give it a language tag
16:35:58 <Souri> title || "@en-us"
16:37:06 <hhalpin> cygri: my preference would be to keep it separate
16:37:09 <hhalpin> souri: not sure if I agree
16:37:57 <hhalpin> cygri: we don't have this in d2rq
16:38:06 <hhalpin> ... but no-one has ever asked for it.
16:39:00 <hhalpin> ericP: For example, SPARQL 1.1 doesnt include dates but does handle datetimes
16:39:26 <hhalpin> cygri: So it would be nice if typed literals could be contracted and specify an override for a datatype
16:40:14 <ericP> "2010-08-10"@date => "2010-08-10"^^xsd:dateTime
16:40:55 <hhalpin> souri: but for computed data types how do we handle this?
16:41:02 <hhalpin> cygri: we do need some kind of casting
16:41:08 <soeren> zakim, unmute me
16:41:08 <Zakim> soeren was not muted, soeren
16:41:12 <hhalpin> ... numbers in string columns
16:42:13 <Zakim> +Souri_Das
16:42:15 <Zakim> -Souri
16:42:24 <Zakim> -Souri_Das
16:42:55 <Zakim> +Souri_Das
16:43:20 <hhalpin> souri: I now agree with the importance, but not sure how to handle computed types, that's all.
16:43:34 <hhalpin> ericP: we do not necessarily need to validate the types
16:43:41 <hhalpin> cygri: flag this as open issue?
16:46:30 <hhalpin> cygri: Entity value schema
16:46:41 <hhalpin> ... here are the attributes that the column actually represents
16:46:50 <hhalpin> ... we would want to map this attribute to different properties
16:47:07 <hhalpin> ... you seem to think this approach is OK.
16:47:46 <hhalpin> ... I remember
16:48:10 <hhalpin> ... there was a long e-mail from ?? that having computed property URIs is not a good idea.
16:48:39 <hhalpin> s/??/Orri from OpenLink
16:48:45 <hhalpin> cygri: 2.5 
16:49:08 <hhalpin> ... we might have a column with an extended hospitality
16:49:39 <hhalpin> ... we want numbers that identify the particular extent that specify the particular extent
16:49:52 <Zakim> -soeren
16:50:32 <Zakim> +soeren
16:50:36 <hhalpin> s/extent/some readable text
16:50:54 <hhalpin> s/hospitality/data in another table
16:51:06 <hhalpin> ... so having a lookup table is a good idea
16:51:10 <hhalpin> souri: not sure if this is a good idea
16:51:19 <hhalpin> ... as imagine if the lookup table gets too big
16:52:28 <hhalpin> ... this may have impacts on the computation 
16:52:43 <hhalpin> cygri: we implemented it in d2rq
16:52:51 <hhalpin> ... and it worked in a straightforward manner
16:54:16 <hhalpin> souri: I will update the wiki
16:54:24 <Zakim> -Souri_Das
16:54:56 <hhalpin> Ashok: let's leave the rest for next week
16:55:05 <hhalpin> cygri: he agreed with most of the rest of these.
16:55:45 <hhalpin> Ashok: is this a reasonable direction?
16:55:55 <hhalpin> ... are we headed towards a first public draft document?
16:56:26 <hhalpin> sequada: Looks great, but we need to work on semantics
16:56:32 <hhalpin> Ashok: No strong negatives
16:56:36 <LeeF> I'm wondering if we're going to talk about syntax before FPWD
16:56:39 <LeeF> because I think that's very important
16:56:40 <hhalpin> cygri: I'm happy with expressivity and general design
16:56:50 <hhalpin> ... some question about syntax
16:56:57 <hhalpin> Ashok: let's talk about syntax last
16:57:22 <hhalpin> juan: datatypes and generating URIs
16:58:11 <hhalpin> juan: myself and Marcelo are working on a semantics document
16:58:11 <Zakim> -LeeF
16:58:22 <Zakim> -soeren
16:58:22 <hhalpin> Meeting Adjourned
16:58:23 <Zakim> -EricP
16:58:24 <Zakim> -Ashok_Malhotra
16:58:27 <Zakim> -MacTed
16:58:27 <hhalpin> trackbot, end meeting
16:58:27 <trackbot> Zakim, list attendees
16:58:28 <trackbot> RRSAgent, please draft minutes
16:58:28 <RRSAgent> I have made the request to generate http://www.w3.org/2010/08/10-rdb2rdf-minutes.html trackbot
16:58:28 <Zakim> As of this point the attendees have been Souri, soeren, juansequeda, cygri, Ashok_Malhotra, MacTed, hhalpin, LeeF, EricP, Souri_Das
16:58:29 <trackbot> RRSAgent, bye
16:58:31 <Zakim> -cygri
16:58:33 <Zakim> -juansequeda
17:00:35 <Ashok> rrsagent, make logs public
17:00:56 <Ashok> rrsagent, draft minutes
17:00:56 <RRSAgent> I have made the request to generate http://www.w3.org/2010/08/10-rdb2rdf-minutes.html Ashok
17:01:14 <hhalpin> its done Ashok.
17:01:19 <hhalpin> Keep up the good work, signing off!
17:21:06 <Zakim> -hhalpin
17:21:07 <Zakim> SW_RDB2RDF()12:00PM has ended
17:21:09 <Zakim> Attendees were Souri, soeren, juansequeda, cygri, Ashok_Malhotra, MacTed, hhalpin, LeeF, EricP, Souri_Das
18:59:06 <Zakim> Zakim has left #rdb2rdf