W3C home > Mailing lists > Public > public-sdw-wg@w3.org > May 2015

RE: UCR isssue: Is provenance in scope? [SEC=UNCLASSIFIED]

From: <Kerry.Taylor@csiro.au>
Date: Mon, 25 May 2015 11:42:06 +0000
To: <gil@isi.edu>, <andrea.perego@jrc.ec.europa.eu>
CC: <public-sdw-wg@w3.org>, <jlieberman@tumblingwalls.com>, <B.Bannerman@bom.gov.au>, <portele@interactive-instruments.de>
Message-ID: <3CD3C8BBF0D87B4D8154C3978732049C50EB9FBF@exmbx05-cdc.nexus.csiro.au>
Dear SDWWG provenance people,

I  agree that provenance is important for a lot of the things that this group's work will be used for.

It is certainly raised several times in our use cases.

And I agree wholeheartedly with Yolanda that iso19115 is not good enough.

It is not mentioned on our charter.

Andrea said 'It may be our job to ensure a consistent mapping from
ISO 19115 to PROV for the description of lineage"

This has been attempted already -a colleague and I did it  in 2013  for one of the PROV "implementations'" as required for a Recommendation (and as we will need to do for some of this group's deliverables)  (I presented it at the OGC Geosemantics dwg meeting in Barcelona this year).  See  www.w3.org/2001/sw/wiki/PROV<http://www.w3.org/2001/sw/wiki/PROV> under heading Uses of PROV.
However, it could do with some documentation and some informed critique. And , it may not align with this:
https://ontohub.org/socop/ISO19115.owl

A mapping from SSN (one of our deliverables)  to prov-o has also been done, too:  http://knoesis.org/ssn2014/paper_9.pdf

However, I am concerned that we may not have the collective energy to add this to the work we already have to do within the time frame we have. Which is why I have suggested we just convince ourselves that the deliverables we do create are well designed to work with prov-o, without actually saying *how* to encode  relevant provenance in prov-o?

If we were to take this on - exactly what would you see us doing? Who would do it? When would we do it (maybe after the FPWD of the other deliverables?)


Kerry



From: Yolanda Gil [mailto:gil@isi.edu]
Sent: Thursday, 21 May 2015 6:21 PM
To: Andrea Perego
Cc: SDW WG; Joshua Lieberman; Taylor, Kerry (Digital, Acton); Bruce Bannerman; Clemens Portele
Subject: Re: UCR isssue: Is provenance in scope? [SEC=UNCLASSIFIED]

+100

Last year in the context of OGC OWS-10 we used both PROV and ISO 19115 to document geospatial provenance.  The OGC technical report is here:

          https://portal.opengeospatial.org/files/?artifact_id=58967

In essence, what we learned is that 1) PROV-O provided a more flexible representation than the ISO standard, and 2) there are many open research challenges in geospatial provenance.

I'd be happy to discuss this work with the group.  My apologies that I have not been able to join the calls much this Spring, everything will change in June and I'd be very interested to pursue this.

Best,

Yolanda


Yolanda Gil
Director of Knowledge Technologies, USC/ISI
Associate Director for Research, Intelligent Systems Division, USC/ISI
Research Professor of Computer Science
Information Sciences Institute
University of Southern California
4676 Admiralty Way, Suite 1001
Marina del Rey, CA 90292 (USA)
+1-310-448-8794
http://www.isi.edu/~gil
@yolandagil

On May 21, 2015, at 12:54 AM, Andrea Perego <andrea.perego@jrc.ec.europa.eu<mailto:andrea.perego@jrc.ec.europa.eu>> wrote:


Just to mention that provenance is already implied in one of the
requirements ("5.2 Citizens as sensors" [1]), and related to a
requirement contributed by Clemens during the Barcelona meeting [2]
(but not included in the BP doc, as far as I can see), coming from UC
4.10 ("Publishing geospatial reference data") [3] - see also Josh's
comment.

This is also an implicit requirement for metadata, as far as lineage
is concerned. It may be our job to ensure a consistent mapping from
ISO 19115 to PROV for the description of lineage.

Andrea

----
[1]http://w3c.github.io/sdw/UseCases/SDWUseCasesAndRequirements.html#CitizensAsSensors
[2]https://www.w3.org/2015/spatial/wiki/BP_Requirements#Be_able_to_annotate_data_with_a_specification_of_what_the_information_is_.2F_where_do_you_find_the_geographic_information_for_the_wellknown_reference_like_a_zip_code
[3]http://w3c.github.io/sdw/UseCases/SDWUseCasesAndRequirements.html#PublishingGeospatialReferenceData

On Wed, May 20, 2015 at 1:46 PM, Joshua Lieberman
<jlieberman@tumblingwalls.com<mailto:jlieberman@tumblingwalls.com>> wrote:

Kerry,

I'll see what I can add this evening. Unfortunately more regrets for the
meeting today (entered on the wiki this time). I'm in a research consortium
meeting this morning.

I think that extensions of PROV-O to cover deriving a "new" feature by
linking to an existing / authoritative feature and/or geometry could be in
scope for Best Practice, but we'll see how well it fits.

-Josh

Joshua Lieberman, Ph.D.
Principal
Tumbling Walls
jlieberman*tumblingwalls.com<http://tumblingwalls.com>
+1 617 431 6431

On May 20, 2015, at 7:41 AM, <Kerry.Taylor@csiro.au<mailto:Kerry.Taylor@csiro.au>> <Kerry.Taylor@csiro.au<mailto:Kerry.Taylor@csiro.au>>
wrote:

Hi Bruce, Josh,

I, for one would love to see that use case! I will do what I can to hold the
presses for you - can you get it on the wiki in the next 24
hours?https://www.w3.org/2015/spatial/wiki/Working_Use_Cases    And also do
the analysis of requirements in the
spreadsheethttps://docs.google.com/spreadsheets/d/1PSnpJYQDgsdgZgPJEfUU0EhVfgFFYGc1WL4xUX9Dunk/edit?usp=sharing

I have done a lot of work on provenance in the context of Bioregional
assessments and other things with GA.
I also was part of that work in publishing BoM's  ACORN-SAT  as linked data
--  and it would have been lovely to do that with provenance too.

However, I do not think we are going to be "doing"  provenance in this
group, I would just like to know that what we are doing neatly docks to
PROV-O (the W3C prov ontology),
and I know that  will not be the case unless we make it so.  See for example
http://knoesis.org/ssn2014/paper_9.pdf. It would be great, too, if  Josh is
watching out for
"reference provenance of spatial data must address not only how a feature
and a spatial such as a geometry were formed, but how they were associated
and under what assumptions for representation of the physical world."
so that we can have some confidence that it will be possible to represent
this--- but I still don't see the doing of that as in scope (wrt our
charter). We should consider it for future work, which we can certainly
recommend coming out of this group.
Can I suggest that you, Josh, note it on the relevant "wish list" on the
main page of the wiki, so it does not get forgotten? Or, put it as an
"issue" on the tracker to ensure it gets more attention if you prefer. We
can put it on a meeting agenda, but can it wait for the UCR to stabilise
first?

Didn't  I meet you, Bruce,  in the Melbourne office  earlier this year? If
you are in Canberra some time it would be nice to catch up on these matters.

Kerry

From: Bruce Bannerman [mailto:B.Bannerman@bom.gov.au]
Sent: Tuesday, 19 May 2015 8:58 AM
To: Taylor, Kerry (Digital, Acton); jlieberman@tumblingwalls.com<mailto:jlieberman@tumblingwalls.com>
Cc: public-sdw-wg@w3.org<mailto:public-sdw-wg@w3.org>
Subject: Re: UCR isssue: Is provenance in scope? [SEC=UNCLASSIFIED]

Hi Kerry,

Provenance is particularly important for climate data related issues, and no
doubt for many more domains as well.

>From a climate perspective, when I publish a scientific paper, I need to be
able to reference all the data that underpins the analysis that the paper
was based on. So this may be:

Published paper
Claims in Published paper based on Analytical Data (perhaps a multi
dimensional array/grid/coverage)
Analytical data is derived from quality assured observations data (with
details as to why each change to the QA obs were made)
Quality assured observations data is derived from 'raw' observations data
which has details as to the conditions, sensors etc that the observation was
made under.

There are many nuances to provenance here. Including an understanding of
what algorithms were used to process the data and ideally a reference to the
source code of these algorithms as they were at the time of the analysis.

And to make things more interesting, the analysis and data is typically
time-series (observations and coverages).



This reminds me I posted on a potential climate use case several months ago,
but forgot to add it.

If there is still interest in this, let me know and I'll put something
together.

Bruce


From: "Kerry.Taylor@csiro.au<mailto:Kerry.Taylor@csiro.au>" <Kerry.Taylor@csiro.au<mailto:Kerry.Taylor@csiro.au>>
Date: Wednesday, 13 May 2015 23:59
To: "jlieberman@tumblingwalls.com<mailto:jlieberman@tumblingwalls.com>" <jlieberman@tumblingwalls.com<mailto:jlieberman@tumblingwalls.com>>
Cc: "public-sdw-wg@w3.org<mailto:public-sdw-wg@w3.org>" <public-sdw-wg@w3.org<mailto:public-sdw-wg@w3.org>>
Subject: RE: UCR isssue: Is provenance in scope?
Resent-From: <public-sdw-wg@w3.org<mailto:public-sdw-wg@w3.org>>
Resent-Date: Thursday, 14 May 2015 00:00


(Resending -missed the list cc)

From: Taylor, Kerry (Digital, Acton)
Sent: Wednesday, 13 May 2015 10:53 PM
To: 'Joshua Lieberman'
Subject: RE: UCR isssue: Is provenance in scope?

+1
I think we need only to make sure (and perhaps show how) our deliverables
can deal with provenance by attaching/linking  some W3C Prov-o. I would not
suggest we need to show to encode spatial data provenance in PROv-o  though.
Provenance is a first class issue in a great deal of spatial data
applications.

Kerry

From: Joshua Lieberman [mailto:jlieberman@tumblingwalls.com]
Sent: Wednesday, 13 May 2015 10:38 PM
To: Frans Knibbe
Cc: SDW WG Public List
Subject: Re: UCR isssue: Is provenance in scope?

Perhaps we can discuss the general issue of scope today on the call. There
are many aspects of spatiotemporal data that in general are similar to
issues with other data, but that clearly require specialization for our
case. For example, reference provenance of spatial data must address not
only how a feature and a spatial such as a geometry were formed, but how
they were associated and under what assumptions for representation of the
physical world. This is quite specialized to spatial and a significant
semantic interoperability issue. We will miss addressing critical points in
our work if we subsume them too often into general ones and deem them out of
scope.

Josh

Joshua Lieberman, Ph.D.
Principal
Tumbling Walls
jlieberman*tumblingwalls.com
+1 617 431 6431


On May 13, 2015, at 8:21 AM, Frans Knibbe <frans.knibbe@geodan.nl<mailto:frans.knibbe@geodan.nl>> wrote:

Hello all,

I have raised an issue for the UCR document: ISSUE-11.
Again, all help in getting this issue resolved is very welcome.

Regards,
Frans

--
Frans Knibbe
Geodan
President Kennedylaan 1
1079 MB Amsterdam (NL)

T +31 (0)20 - 5711 347
E frans.knibbe@geodan.nl<mailto:frans.knibbe@geodan.nl>
www.geodan.nl<http://www.geodan.nl>
disclaimer




--
Andrea Perego, Ph.D.
Scientific / Technical Project Officer
European Commission DG JRC
Institute for Environment & Sustainability
Unit H06 - Digital Earth & Reference Data
Via E. Fermi, 2749 - TP 262
21027 Ispra VA, Italy

https://ec.europa.eu/jrc/

----
The views expressed are purely those of the writer and may
not in any circumstances be regarded as stating an official
position of the European Commission.
Received on Monday, 25 May 2015 11:43:15 UTC

This archive was generated by hypermail 2.4.0 : Thursday, 24 March 2022 20:31:16 UTC