IRC log of rdb2rdf on 2010-10-21

Timestamps are in UTC.

18:11:52 [RRSAgent]
RRSAgent has joined #rdb2rdf
18:11:52 [RRSAgent]
logging to http://www.w3.org/2010/10/21-rdb2rdf-irc
18:12:00 [betehess]
trackbot, start meeting
18:12:00 [trackbot]
Sorry... I don't know anything about this channel
18:12:00 [trackbot]
If you want to associate this channel with an existing Tracker, please say 'trackbot, associate this channel with #channel' (where #channel is the name of default channel for the group)
18:12:09 [betehess]
RRSAgent, draft minutes
18:12:09 [RRSAgent]
I have made the request to generate http://www.w3.org/2010/10/21-rdb2rdf-minutes.html betehess
18:12:15 [betehess]
scribe: betehess
18:12:29 [juansequeda]
http://www.w3.org/2001/sw/rdb2rdf/wiki/All_Cases_for_Default_Mapping
18:12:33 [juansequeda]
http://www.w3.org/2001/sw/rdb2rdf/directGraph/
18:12:59 [betehess]
@@: section 2
18:13:24 [betehess]
... what happens with the PrimaryKey?
18:14:15 [betehess]
... @enumerates 7 cases that have to be treated@
18:17:35 [betehess]
ericP: the FK points to a CandidateKey in another table, but ti does not mean it's a PK
18:17:47 [betehess]
... or even that there is a PK
18:18:22 [betehess]
@@: is that the only case you need both tables?
18:18:25 [betehess]
ericP: ye
18:19:01 [betehess]
@@: let's write down any single case in specific sections
18:19:19 [betehess]
Zakim, who is noisy?
18:19:31 [Zakim]
betehess, listening for 10 seconds I heard sound from the following: Marcelo (94%), EricP (44%)
18:20:09 [betehess]
ericP: the recipe should tell you what it has to do
18:20:21 [betehess]
... the examples give you a sense of what it does
18:21:27 [betehess]
... if you don't want to have separate examples for each case, you can just say: "this example covers these cases"
18:21:38 [betehess]
... because some cannot exist on their own
18:21:48 [betehess]
q+
18:22:32 [betehess]
+1
18:24:57 [betehess]
ericP: let's be sure we agree on the examples
18:26:15 [betehess]
Marcelo: the use-cases are ok, the examples too
18:26:23 [betehess]
... it's just difficult to read
18:26:40 [Zakim]
+OpenLink_Software
18:26:42 [betehess]
... I propose myself to re-organize for next week
18:26:48 [MacTed]
Zakim, OpenLink_Software is me
18:26:49 [Zakim]
+MacTed; got it
18:26:56 [betehess]
ericP: what about terminology?
18:27:03 [betehess]
... tables VS relations
18:27:23 [MacTed]
Zakim, mute me
18:27:23 [Zakim]
MacTed should now be muted
18:27:50 [betehess]
... SQL can give you multiple columns with the same name
18:28:00 [betehess]
... you cannot do it with SPARQL
18:28:06 [betehess]
... because @@
18:28:34 [betehess]
... so we have to decide what to do with headers and columns with the same name
18:29:40 [betehess]
juansequeda: can we vote to use SQL terminology instead of relational algebra terminology?
18:29:50 [MacTed]
Zakim, unmute me
18:29:50 [Zakim]
MacTed should no longer be muted
18:30:56 [betehess]
... choices are: relations/attributes OR tables/columns OR tables/attributes
18:31:17 [MacTed]
tables/columns.
18:31:36 [betehess]
ericP: and we can speak about fnames instead of order
18:31:59 [betehess]
[ betehess: tables/columns ]
18:32:12 [MacTed]
saying "relation = table" doesn't make sense to me. mixing tables/attributes makes even less.
18:32:36 [Marcelo]
relations/attributes
18:33:22 [betehess]
@@: in relation algebra, people use "relation"
18:33:56 [betehess]
@@: in any of my classes, I apologize because the names are not fully defined
18:34:29 [betehess]
RRSAgent, please draft minutes
18:34:29 [RRSAgent]
I have made the request to generate http://www.w3.org/2010/10/21-rdb2rdf-minutes.html betehess
18:34:53 [betehess]
@@: I would suggest to use SQL terminology because of the audience
18:35:33 [betehess]
PatH: we at least need to define the terms very well
18:35:58 [MacTed]
+1
18:36:03 [Marcelo]
+1
18:36:08 [betehess]
Consensus around SQL terminology
18:36:14 [betehess]
ericP: +1
18:36:23 [juansequeda]
+1 for using SQL terminology
18:36:45 [PatH]
abstain. I go with the flow.
18:37:37 [betehess]
@@: SQL people use column, not attributes
18:38:25 [betehess]
@@: can we say: tables + columns + columns are unique within the table?
18:38:26 [MacTed]
SQL identifiers == catalog.owner/schema.table.field
18:38:49 [MacTed]
s/field/column/
18:39:25 [betehess]
ericP: the schema has a unique mapping from to datatypes
18:39:26 [MacTed]
column names are unique within table; table names are unique within owner/schema; owner/schema names are unique within catalog
18:39:38 [betehess]
... tuple has unique mapping from names to value (or NULL)
18:40:41 [betehess]
... Marcelo and I propose a clearer version of the document
18:41:00 [betehess]
... expecting others will understand how genius we are :-)
18:41:21 [Marcelo]
http://www.w3.org/2001/sw/rdb2rdf/directGraph/
18:42:59 [betehess]
ericP proposed a strategy for editing the documents
18:46:57 [betehess]
betehess: what is semantics?
18:47:14 [betehess]
ericP: you take an RDB and then you have an RDF graph
18:47:53 [betehess]
... it's all about the formalism of the formal mapping
18:50:20 [betehess]
[[
18:50:21 [betehess]
given a SPARQL-to-SQL mapping to access SQL database, prove that the semantics of the translated SPARQL query executed against a particular RDB dataset is equivalent to the same SPARQL query executed against the same RDB dataset seen through the RDB2RDF mapping.
18:50:22 [betehess]
]]
18:50:31 [PatH]
+1 to whoever just spoke.
18:51:20 [PatH]
Yes, we should do this. BUt this is about the semantics of the inputs and outputs of the mapping, not of the mapping language itself.
18:51:31 [betehess]
PatH, yes :-)
18:51:48 [Marcelo]
+1 PatH
18:55:47 [Zakim]
-EricP
18:56:22 [betehess]
1. in the case of the Default Mapping, you have a function "mapping : RDB → RDF".
18:56:32 [betehess]
2. in the case of R2ML, you have a function "mapping : (RDB×R2ML) → RDF". The Default Mapping is just a particular case where you use the empty value as an inhabitant for R2ML.
18:57:19 [PatH]
We could (not on IRC) draw this as a 'square' of functors which we want to commute.
18:57:24 [MacTed]
+10000000 :-)
18:58:03 [PatH]
BTW, I think this might be quite tricky to establish, I think. It *ought* to be easy, but details, details...
19:00:24 [PatH]
Seems to me that there will be a big semantic mismatch arising from the 'open world' assumption (semantic: classical model thoery) of RDF. Have you guys discussed this?
19:00:51 [juansequeda]
PatH, we haven't
19:01:05 [PatH]
Oh.
19:02:45 [MacTed]
PatH - not sure what issue you're seeing....
19:02:45 [MacTed]
RDB ("schema first") is generally considered a 'closed world' model.
19:02:45 [MacTed]
RDF ("schema last") is generally considered an 'open world' model.
19:02:46 [MacTed]
'closed world' fits fine within 'open world', which is what we're doing. RDB exposed as *part of* the RDF GGG.
19:03:26 [PatH]
Yes, but if we try to establish semantic relationship using the semantics, we will find that RDF cannot express things that are implicit in the RDB semantic model.
19:03:40 [MacTed]
example, please?
19:04:19 [PatH]
Implied negatives for missing data, for example, which can be detected by SPARQL.
19:04:50 [PatH]
Semantically, closed world is much stronger than open.
19:05:09 [PatH]
stronger = fewer models, more implications.
19:05:43 [MacTed]
can you state the problem you see more explicitly? I'm not seeing the issue you apparently are.
19:07:09 [PatH]
Might take too long on IRC. Need a fully worked out example. Semantically, its that minimal models or algebraic semantics can express many things that cant be expressed in RDF wihtout negation.
19:10:17 [MacTed]
so ... they can be expressed in RDF *with* negation, yes? or in RDF with "affirmative statement of negative"?
19:10:19 [PatH]
Such things as: negative queries following from failure to find a match. And cardinality queries like 'how many'. None of this follows in the RDF semantics.
19:10:43 [PatH]
But RDF doesnt have negation!
19:10:54 [MacTed]
I think you're shifting from RDF to SPARQL, and from RDB to SQL
19:11:17 [MacTed]
<a> hasNo <b> -- affirmative statement of negation :-)
19:11:19 [PatH]
No, thats the problem. SPARQL can do things that arent supported by the RDF *semantics*.
19:11:35 [Zakim]
-MacTed
19:11:40 [Zakim]
-PatH
19:11:41 [Zakim]
-Alexandre
19:11:49 [Zakim]
-juansequeda
19:11:49 [betehess]
RRSAgent, draft minutes
19:11:49 [RRSAgent]
I have made the request to generate http://www.w3.org/2010/10/21-rdb2rdf-minutes.html betehess
19:11:54 [Zakim]
-Marcelo
19:11:55 [Zakim]
Team_(RDB2RDF)18:03Z has ended
19:11:57 [Zakim]
Attendees were PatH, Marcelo, Alexandre, juansequeda, EricP, MacTed
19:12:30 [betehess]
are the minutes of this WG public?
19:12:47 [betehess]
I mean, world readable?
19:12:59 [MacTed]
I'm clearly not grasping something.
19:13:06 [betehess]
do you guys need to authenticate to access the other minutes?
19:13:56 [MacTed]
RRSAgent, make minutes public
19:13:56 [RRSAgent]
I'm logging. I don't understand 'make minutes public', MacTed. Try /msg RRSAgent help
19:14:09 [betehess]
arf, I did it manually :-)
19:14:16 [MacTed]
*heh*
19:14:17 [betehess]
it's ok
19:14:41 [betehess]
and I will write briefly about the last part as I wans't scribing
20:34:10 [betehess]
minutes and semantics discussion available at http://lists.w3.org/Archives/Public/public-rdb2rdf-wg/2010Oct/0070.html
20:34:29 [betehess]
my theorem is not well-typed but you get it :-)
20:54:04 [Zakim]
Zakim has left #rdb2rdf
22:21:24 [MacTed]
MacTed has joined #RDB2RDF