W3C home > Mailing lists > Public > public-prov-wg@w3.org > April 2013

PROV-Dictionary update for telecon 4.11.2013

From: Tom De Nies <tom.denies@ugent.be>
Date: Thu, 11 Apr 2013 12:00:52 +0200
Message-ID: <CA+=hbbewsQqJ9Tbty-XOxb7q93kERFnoipzhVLM2iU+VshOgYQ@mail.gmail.com>
To: Provenance Working Group <public-prov-wg@w3.org>
Cc: Sam Coppens UGent <sam.coppens@ugent.be>
Hi all,

Since neither Sam nor I can make the telecon today, we'll provide a status
update in this email.

There are 4 issues remaining on the tracker:
https://www.w3.org/2011/prov/track/products/19
2 are pending review, and I expect they can be closed after today.

The two remaining issues are:

- Constraints: see issue-660 https://www.w3.org/2011/prov/track/issues/660
This is something that came up during the last round of review. We have the
possibility to simplify inference D8 considerably by replacing it with the
constraints suggested in the issue. The goal is still to guarantee that
insertion and removal don't introduce or remove any key-entity pairs other
than those specified.
I don't see any negative consequences, but I might be wrong. Could this be
discussed during the telecon, and possibly voted on?
Proposed: Drop inference
D8<https://dvcs.w3.org/hg/prov/raw-file/default/dictionary/Overview.html#insertion-removal-membership-inference>(insertion-removal-membership)
and create a new inference
(membership-removal-membership) consisting of the following sub-inferences:
1. IF prov:hadDictionaryMember(d1, e, k) and prov:derivedByRemovalFrom(d2,
d1, K1) and k ∉ K1 THEN prov:hadDictionaryMember(d2, e, k)
2. IF prov:hadDictionaryMember(d2, e, k) and prov:derivedByRemovalFrom(d2,
d1, K1) THEN prov:hadDictionaryMember(d1, e, k)


- Example: see issue-642 https://www.w3.org/2011/prov/track/issues/642
Paul would like to see an example of the provenance of a dictionary added
to the document. We are currently working on this. The idea is to take the
players of the Boston Red Sox baseball team at the start of the 2011 season
and the 2012 season, as well as their field positions during the first two
games of 2012, and track the provenance of this data. For the first
dictionaries, the uniform number is the key, and the players the entities.
For the second, the field positions are used as keys, and the players as
entities. In other words, we would have an example of insertion/removal
differences between season 2011 and 2012, and a clear example of update
semantics for the field positions. We would show the raw data and the
prov-n, prov-xml and prov-o serializations.
The example would be added to the appendix of the document. Does this seem
acceptable to everyone?
Proposed: use Boston Red Sox 2011 and 2012 team and field positions as
example for PROV-Dictionary


Other various things:

- Luc and Simon's reviews were very helpful, thanks again for those. All of
Simon's remarks have been addressed, and most of Luc's (except the
constraints, discussed earlier).

- I've changed the title of the document to "PROV-Dictionary: Modeling
Provenance for Dictionary Data Structures" instead of just "PROV
Dictionary", to be more consistent with the other documents.

- the change log since FPWD can be viewed here:
https://dvcs.w3.org/hg/prov/raw-file/default/dictionary/Overview.html#change-log


Regards,
Tom and Sam
Received on Thursday, 11 April 2013 10:03:13 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 16:51:35 UTC