W3C home > Mailing lists > Public > public-prov-wg@w3.org > March 2012

Re: {Disarmed} Feedback about DM draft 4

From: Luc Moreau <l.moreau@ecs.soton.ac.uk>
Date: Fri, 02 Mar 2012 17:01:14 +0100
Message-ID: <EMEW3|3437d5d9508a7d0f2e0e75fa66e9f5a9o22F6Z08l.moreau|ecs.soton.ac.uk|4F50EECA.2010300@ecs.soton.ac.uk>
To: Daniel Garijo <dgarijo@delicias.dia.fi.upm.es>
CC: Paolo Missier <Paolo.Missier@ncl.ac.uk>, "<public-prov-wg@w3.org>" <public-prov-wg@w3.org>
Hi Daniel

Paolo and I have made changes following your feedback.
Our responses can be found below.

This now completes WD4. Notes have been inserted in the document,
which we will tackle as part of WD5.

We are proposing to close ISSUE-274. Let us know if this is fine with you.
Regards,
Luc

 > Hi all,
 > here are my comments after reading part 1:
 >
 > Objectives:
 >
 >    - decide whether the new documents are inline with the simplification
 >    objective recommend whether they become the new editor's draft.
 >
 >           ---> YES, it is much more simple and easy to read now. I would
 > take it as the new editor's draft.
 >
 >
 >    -    Decide whether ISSUE-145, ISSUE-183, ISSUE-215, ISSUE-225 and
 >    ISSUE-234 (all relating to identifiers) can be closed
 >
 >           ---> 145: No accounts anymore, just bundles (or AccountEntity),
 > so it could be closed.
 >           ---> 215: It has to do with the distinction between records,
 > accounts and mitning ids. Since we don't have records and accounts, then
 > the issue could be closed.
 >           ---> 225: All objects in the universe of discourse have been
 > clarified. Can be closed.
 >           ---> 234: The term "record" has been dropped. Therefore, 
this can
 > be closed.
 >
 > ***Comments from
 > 
http://dvcs.w3.org/hg/prov/raw-file/default/model/working-copy/towards-wd4.html
 > ***
 >
 > - Button "Hide ASN" does actually do anything?

Removed

 >
 > 2.3
 > - AccountEntity? I thought it was Bundle, but ok.

No name had been agreed.
Waiting to see how discussion on account progresses.

 >
 > -Three types of agents are recognized by PROV-DM because they are 
commonly
 > encountered in applications making data and documents available on 
the Web:
 > persons, software agents, and organizations.--> Wasn't software 
supposed to
 > be system/computingSystem?

No consensus was reached. Document mentions both
  People and Human

We reverted back to the previous definiitions.



 >
 > 2.5: there are arrows missing: Activity wasStartedBy Activity. Entity:
 > alternateOf, specializationOf

It is an overview, intentionally, simplifying the presentation,
including not showing some edges.
With the component-based presentation, further small diagrams will be 
produced.

TODO: add a sentence saying that it is not complete coverage of the dm 
in diagram.


 >
 > 3.1: It would be helpful to see the properties labelled in the figure.
 >

Figures to be drawn by hand.  Note added.


 > 3.2: Here I would suggest to simplify the figure (leave just 2 
authors (as
 > in the example), or the editors), and label the edges as well.

Same.

 >
 > 3.3: Ah finally a reference to metadata provenance :) This is what 
Kai and
 > some of the DC community were asking for.

Nothing to do.
 >
 > 4.1.2: "In contrast, an activity is something that happens, unfolds or
 > develops through time, but is typically not identifiable by the
 > characteristics it exhibits at any point during its duration". What about
 > the activity's ID. Why isn't that enough to characterize the activity
 > enough to become an entity or an agent?


This is ISSUE-94, which is now closed.

 >
 > 4.2: wasStartedBy between activities is missing in the table. In fact I
 > haven't seen wasStartedBy between activities in the doc. It certainly was
 > an overloaded property in the WD4. Has it been removed?

This is proposed to be reorganized for WD5.


 >
 > 4.2.1.2:There is a note that refers to Usage record's id. It should 
be just
 > usage.

Text of notes has not been updated.
Hopefully, this note can be removed soon.

 >
 > 4.2.3.2: I got the feeling from discussions on the mailing list that we
 > were going to reduce one of the derivation types (Imprecise-1 
derivation).
 > Am I wrong?

Yes, proposal to come for WD5.

 >
 > 4.3.3.5: I don't understand how a path in a computer or a row and a 
column
 > are a geographic place.

Updated definition to allow for non-geographic places.

 >
 > 5.5: Example missing
 >
 > 5.7: Example missing.

Added to the notes, these sections need rewriting.

 >
 > 5.8: If collections are just a kind of entity and they have their custom
 > relationships (afterInsertion, afterRemoval), would it make sense to
 > separate them from the core? (In a profile, best practice or example of
 > extensibility)
 >

In WD5, they will constitute one of the components.

 > *********
 > - One question that came into my mind when reading the model: How would I
 > model a usage that lasted for 20 min? (Right now we only have the 
beggining
 > of the usage). Example: My activity uses 2 files. The first one is parsed
 > for 20 mins and the other one instantly, and I want to model this 
with DM.
 > Unless I create 2 activities (which is not what happened) I don't see 
how.
 >

To represent something that happens, then one needs to use an activity, to
which one can attach duration.

This could go in a best practice document, but ... nobody is taking charge
of such a document. Would you volunteer? ;-)

 > Thanks,
 > Daniel
 >


On 24/02/2012 12:00, Daniel Garijo wrote:
> Hi all,
> here are my comments after reading part 1:
>
> Objectives:
>
>     * decide whether the new documents are inline with the
>       simplification objective recommend whether they become the new
>       editor's draft.
>
>           ---> YES, it is much more simple and easy to read now. I 
> would take it as the new editor's draft.
>
>     *    Decide whether ISSUE-145, ISSUE-183, ISSUE-215, ISSUE-225 and
>       ISSUE-234 (all relating to identifiers) can be closed
>
>           ---> 145: No accounts anymore, just bundles (or 
> AccountEntity), so it could be closed.
>           ---> 215: It has to do with the distinction between records, 
> accounts and mitning ids. Since we don't have records and accounts, 
> then the issue could be closed.
>           ---> 225: All objects in the universe of discourse have been 
> clarified. Can be closed.
>           ---> 234: The term "record" has been dropped. Therefore, 
> this can be closed.
>
> ***Comments from 
> http://dvcs.w3.org/hg/prov/raw-file/default/model/working-copy/towards-wd4.html***
>
> - Button "Hide ASN" does actually do anything?
>
> 2.3
> - AccountEntity? I thought it was Bundle, but ok.
>
> -Three types of agents are recognized by PROV-DM because they are 
> commonly encountered in applications making data and documents 
> available on the Web: persons, software agents, and organizations.--> 
> Wasn't software supposed to be system/computingSystem?
>
> 2.5: there are arrows missing: Activity wasStartedBy Activity. Entity: 
> alternateOf, specializationOf
>
> 3.1: It would be helpful to see the properties labelled in the figure.
>
> 3.2: Here I would suggest to simplify the figure (leave just 2 authors 
> (as in the example), or the editors), and label the edges as well.
>
> 3.3: Ah finally a reference to metadata provenance :) This is what Kai 
> and some of the DC community were asking for.
>
> 4.1.2: "In contrast, an activity is something that happens, unfolds or 
> develops through time, but is typically not identifiable by the 
> characteristics it exhibits at any point during its duration". What 
> about the activity's ID. Why isn't that enough to characterize the 
> activity enough to become an entity or an agent?
>
> 4.2: wasStartedBy between activities is missing in the table. In fact 
> I haven't seen wasStartedBy between activities in the doc. It 
> certainly was an overloaded property in the WD4. Has it been removed?
>
> 4.2.1.2:There is a note that refers to Usage record's id. It should be 
> just usage.
>
> *MailScanner has detected a possible fraud attempt from "4.2.3.2" 
> claiming to be* 4.2.3.2 <http://4.2.3.2>: I got the feeling from 
> discussions on the mailing list that we were going to reduce one of 
> the derivation types (Imprecise-1 derivation). Am I wrong?
>
> *MailScanner has detected a possible fraud attempt from "4.3.3.5" 
> claiming to be* 4.3.3.5 <http://4.3.3.5>: I don't understand how a 
> path in a computer or a row and a column are a geographic place.
>
> 5.5: Example missing
>
> 5.7: Example missing.
>
> 5.8: If collections are just a kind of entity and they have their 
> custom relationships (afterInsertion, afterRemoval), would it make 
> sense to separate them from the core? (In a profile, best practice or 
> example of extensibility)
>
> *********
> - One question that came into my mind when reading the model: How 
> would I model a usage that lasted for 20 min? (Right now we only have 
> the beggining of the usage). Example: My activity uses 2 files. The 
> first one is parsed for 20 mins and the other one instantly, and I 
> want to model this with DM. Unless I create 2 activities (which is not 
> what happened) I don't see how.
>
> Thanks,
> Daniel
Received on Saturday, 3 March 2012 15:10:04 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 26 April 2012 13:06:58 GMT