Slides and Notes from Joint Meeting at Tokyo P7

Dear MIG,

At yesterday's joint metadata session, I felt that a major barrier to effectively discussing the proposed elements--i.e. evaluating as to whether the elements were "the right set"--was the lack of definitions for concepts they are to represent. In an effort improve the discussion, I offer a strawman set of definitions for each of the terms in slide 9 ("Open Data: Purposing the elements") in a spreadsheet posted at https://rd-alliance.org/metadata-p7-tokyo.html.  This was a useful exercise at least for me as I realized that I really didn't understand the intended concept behind a few of the elements. 

Another thing that I think is important to clarify is the overall model for the things we are trying to describe with these elements.  This is covered in part by block diagram in the slides that were presented, but I suspect more should be spelled out.  As context for my posted defiitions, I assumed that these elements are (complex) properties describing a thing I refer to as a Resource; a more precise definition is given within that spreadsheet.  I will also note that my own thinking comes from work I'm currently doing as part of the Materials Science Interest Group (and the proposed WG on resource registries for Material Science); a conceptual model is describe here:  http://wiki.nationaldataservice.org/ResourceConceptDataModel. 

hope this is helpful,
Ray

On definitions for things data-related, I recently ran across the
following website (from a German PhD thesis of a few years ago by Jakob
Voß):
http://aboutdata.org/patterns.html
It defines a number of standard "patterns" he found repeated across a
very wide variety of data systems. There is some clear overlap with your
list ("Schema", "Identifier", "title"/"label"); maybe there are some
additional elements in his analysis that would be helpful in defining
the metadata needed to understand a particular collection of data.
Arthur

Ray—thanks for compiling this. I wasn’t at the session yesterday, but looking at the slides I had the same reaction. I’ve taken the liberty of making a google doc spreadsheet from your Excel to make it easier to collect comments and discussion. The resource definition is on the first tab, metadata element definitions on the second tab. At this point if people want, I’d suggest adding a new column for each reviewer. Comments can also be added using the google comments.
the spreadsheet is at
https://docs.google.com/spreadsheets/d/18Wi2GNM10d3e_iDWgDS-w4M4Zeaei1wO...
permissions are set up so anyone with the link can edit.
the doc is also published at
https://docs.google.com/spreadsheets/d/18Wi2GNM10d3e_iDWgDS-w4M4Zeaei1wO...
with that link you can only view, so you can share for read-only access.
steve
- Show quoted text -From: raymond.plante=***@***.***-groups.org [mailto:***@***.***-groups.org] On Behalf Of RayPlante
Sent: Wednesday, March 02, 2016 6:45 AM
To: Metadata IG <***@***.***-groups.org>
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
Dear MIG,
At yesterday's joint metadata session, I felt that a major barrier to effectively discussing the proposed elements--i.e. evaluating as to whether the elements were "the right set"--was the lack of definitions for concepts they are to represent. In an effort improve the discussion, I offer a strawman set of definitions for each of the terms in slide 9 ("Open Data: Purposing the elements") in a spreadsheet posted at https://rd-alliance.org/metadata-p7-tokyo.html. This was a useful exercise at least for me as I realized that I really didn't understand the intended concept behind a few of the elements.
Another thing that I think is important to clarify is the overall model for the things we are trying to describe with these elements. This is covered in part by block diagram in the slides that were presented, but I suspect more should be spelled out. As context for my posted defiitions, I assumed that these elements are (complex) properties describing a thing I refer to as a Resource; a more precise definition is given within that spreadsheet. I will also note that my own thinking comes from work I'm currently doing as part of the Materials Science Interest Group (and the proposed WG on resource registries for Material Science); a conceptual model is describe here: http://wiki.nationaldataservice.org/ResourceConceptDataModel.
hope this is helpful,
Ray
--
Full post: https://rd-alliance.org/group/metadata-ig/post/slides-and-notes-joint-me...
Manage my subscriptions: https://rd-alliance.org/mailinglist
Stop emails for this post: https://rd-alliance.org/mailinglist/unsubscribe/51502

We (that is in STFC) have also been working on similar models in the context of facilities and nano-materials data.
We have just completed a deliverable for the project NFFA-Europe (www.nffa.eu) where we have
defined a core set of concepts and fields for experiments. The document is not quite released as yet –
I’ll send a link as soon as it is, but I’ll be talking about it at the joint Materials session this afternoon.
It’s based on our earlier work on Core Scientific MetaData for facilities data.
http://purl.org/net/csmd
thanks
Brian
- Show quoted text -From: steve.richard=***@***.***-groups.org [mailto:***@***.***-groups.org] On Behalf Of srichardUSGIN
Sent: 01 March 2016 22:52
To: RayPlante; Metadata IG
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
Ray—thanks for compiling this. I wasn’t at the session yesterday, but looking at the slides I had the same reaction. I’ve taken the liberty of making a google doc spreadsheet from your Excel to make it easier to collect comments and discussion. The resource definition is on the first tab, metadata element definitions on the second tab. At this point if people want, I’d suggest adding a new column for each reviewer. Comments can also be added using the google comments.
the spreadsheet is at
https://docs.google.com/spreadsheets/d/18Wi2GNM10d3e_iDWgDS-w4M4Zeaei1wO...
permissions are set up so anyone with the link can edit.
the doc is also published at
https://docs.google.com/spreadsheets/d/18Wi2GNM10d3e_iDWgDS-w4M4Zeaei1wO...
with that link you can only view, so you can share for read-only access.
steve
From: raymond.plante=***@***.***-groups.org [mailto:***@***.***-groups.org] On Behalf Of RayPlante
Sent: Wednesday, March 02, 2016 6:45 AM
To: Metadata IG <***@***.***-groups.org>
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
Dear MIG,
At yesterday's joint metadata session, I felt that a major barrier to effectively discussing the proposed elements--i.e. evaluating as to whether the elements were "the right set"--was the lack of definitions for concepts they are to represent. In an effort improve the discussion, I offer a strawman set of definitions for each of the terms in slide 9 ("Open Data: Purposing the elements") in a spreadsheet posted at https://rd-alliance.org/metadata-p7-tokyo.html. This was a useful exercise at least for me as I realized that I really didn't understand the intended concept behind a few of the elements.
Another thing that I think is important to clarify is the overall model for the things we are trying to describe with these elements. This is covered in part by block diagram in the slides that were presented, but I suspect more should be spelled out. As context for my posted defiitions, I assumed that these elements are (complex) properties describing a thing I refer to as a Resource; a more precise definition is given within that spreadsheet. I will also note that my own thinking comes from work I'm currently doing as part of the Materials Science Interest Group (and the proposed WG on resource registries for Material Science); a conceptual model is describe here: http://wiki.nationaldataservice.org/ResourceConceptDataModel.
hope this is helpful,
Ray
--
Full post: https://rd-alliance.org/group/metadata-ig/post/slides-and-notes-joint-me...
Manage my subscriptions: https://rd-alliance.org/mailinglist
Stop emails for this post: https://rd-alliance.org/mailinglist/unsubscribe/51502
We (that is in STFC) have also been working on similar models in the context of facilities and nano-materials data.
We have just completed a deliverable for the project NFFA-Europe (www.nffa.eu) where we have
defined a core set of concepts and fields for experiments. The document is not quite released as yet –
I’ll send a link as soon as it is, but I’ll be talking about it at the joint Materials session this afternoon.
It’s based on our earlier work on Core Scientific MetaData for facilities data.
http://purl.org/net/csmd
thanks
Brian
From: steve.richard=***@***.***-groups.org [mailto:***@***.***-groups.org] On Behalf Of srichardUSGIN
Sent: 01 March 2016 22:52
To: RayPlante; Metadata IG
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
Ray—thanks for compiling this. I wasn’t at the session yesterday, but looking at the slides I had the same reaction. I’ve taken the liberty of making a google doc spreadsheet from your Excel to make it easier to collect comments and discussion. The resource definition is on the first tab, metadata element definitions on the second tab. At this point if people want, I’d suggest adding a new column for each reviewer. Comments can also be added using the google comments.
the spreadsheet is at
https://docs.google.com/spreadsheets/d/18Wi2GNM10d3e_iDWgDS-w4M4Zeaei1wO...
permissions are set up so anyone with the link can edit.
the doc is also published at
https://docs.google.com/spreadsheets/d/18Wi2GNM10d3e_iDWgDS-w4M4Zeaei1wO...
with that link you can only view, so you can share for read-only access.
steve
- Show quoted text -From: raymond.plante=***@***.***-groups.org [mailto:***@***.***-groups.org] On Behalf Of RayPlante
Sent: Wednesday, March 02, 2016 6:45 AM
To: Metadata IG <***@***.***-groups.org>
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
Dear MIG,
At yesterday's joint metadata session, I felt that a major barrier to effectively discussing the proposed elements--i.e. evaluating as to whether the elements were "the right set"--was the lack of definitions for concepts they are to represent. In an effort improve the discussion, I offer a strawman set of definitions for each of the terms in slide 9 ("Open Data: Purposing the elements") in a spreadsheet posted at https://rd-alliance.org/metadata-p7-tokyo.html. This was a useful exercise at least for me as I realized that I really didn't understand the intended concept behind a few of the elements.
Another thing that I think is important to clarify is the overall model for the things we are trying to describe with these elements. This is covered in part by block diagram in the slides that were presented, but I suspect more should be spelled out. As context for my posted defiitions, I assumed that these elements are (complex) properties describing a thing I refer to as a Resource; a more precise definition is given within that spreadsheet. I will also note that my own thinking comes from work I'm currently doing as part of the Materials Science Interest Group (and the proposed WG on resource registries for Material Science); a conceptual model is describe here: http://wiki.nationaldataservice.org/ResourceConceptDataModel.
hope this is helpful,
Ray
--
Full post: https://rd-alliance.org/group/metadata-ig/post/slides-and-notes-joint-me...
Manage my subscriptions: https://rd-alliance.org/mailinglist
Stop emails for this post: https://rd-alliance.org/mailinglist/unsubscribe/51502

Brian—Excellent! How does your analysis compare with what’s in the spreadsheet?
From: ***@***.*** [mailto:***@***.***]
Sent: Wednesday, March 02, 2016 8:14 AM
To: Steve Richard <***@***.***>; ***@***.***; ***@***.***-groups.org
Cc: ***@***.***
Subject: RE: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
We (that is in STFC) have also been working on similar models in the context of facilities and nano-materials data.
We have just completed a deliverable for the project NFFA-Europe (www.nffa.eu) where we have
defined a core set of concepts and fields for experiments. The document is not quite released as yet –
I’ll send a link as soon as it is, but I’ll be talking about it at the joint Materials session this afternoon.
It’s based on our earlier work on Core Scientific MetaData for facilities data.
http://purl.org/net/csmd
thanks
Brian
- Show quoted text -From: steve.richard=***@***.***-groups.org [mailto:***@***.***-groups.org] On Behalf Of srichardUSGIN
Sent: 01 March 2016 22:52
To: RayPlante; Metadata IG
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
Ray—thanks for compiling this. I wasn’t at the session yesterday, but looking at the slides I had the same reaction. I’ve taken the liberty of making a google doc spreadsheet from your Excel to make it easier to collect comments and discussion. The resource definition is on the first tab, metadata element definitions on the second tab. At this point if people want, I’d suggest adding a new column for each reviewer. Comments can also be added using the google comments.
the spreadsheet is at
https://docs.google.com/spreadsheets/d/18Wi2GNM10d3e_iDWgDS-w4M4Zeaei1wO...
permissions are set up so anyone with the link can edit.
the doc is also published at
https://docs.google.com/spreadsheets/d/18Wi2GNM10d3e_iDWgDS-w4M4Zeaei1wO...
with that link you can only view, so you can share for read-only access.
steve
From: raymond.plante=***@***.***-groups.org [mailto:***@***.***-groups.org] On Behalf Of RayPlante
Sent: Wednesday, March 02, 2016 6:45 AM
To: Metadata IG <***@***.***-groups.org>
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
Dear MIG,
At yesterday's joint metadata session, I felt that a major barrier to effectively discussing the proposed elements--i.e. evaluating as to whether the elements were "the right set"--was the lack of definitions for concepts they are to represent. In an effort improve the discussion, I offer a strawman set of definitions for each of the terms in slide 9 ("Open Data: Purposing the elements") in a spreadsheet posted at https://rd-alliance.org/metadata-p7-tokyo.html. This was a useful exercise at least for me as I realized that I really didn't understand the intended concept behind a few of the elements.
Another thing that I think is important to clarify is the overall model for the things we are trying to describe with these elements. This is covered in part by block diagram in the slides that were presented, but I suspect more should be spelled out. As context for my posted defiitions, I assumed that these elements are (complex) properties describing a thing I refer to as a Resource; a more precise definition is given within that spreadsheet. I will also note that my own thinking comes from work I'm currently doing as part of the Materials Science Interest Group (and the proposed WG on resource registries for Material Science); a conceptual model is describe here: http://wiki.nationaldataservice.org/ResourceConceptDataModel.
hope this is helpful,
Ray
--
Full post: https://rd-alliance.org/group/metadata-ig/post/slides-and-notes-joint-me...
Manage my subscriptions: https://rd-alliance.org/mailinglist
Stop emails for this post: https://rd-alliance.org/mailinglist/unsubscribe/51502
Brian—Excellent! How does your analysis compare with what’s in the spreadsheet?
From: ***@***.*** [mailto:***@***.***]
Sent: Wednesday, March 02, 2016 8:14 AM
To: Steve Richard <***@***.***>; ***@***.***; ***@***.***-groups.org
Cc: ***@***.***
Subject: RE: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
We (that is in STFC) have also been working on similar models in the context of facilities and nano-materials data.
We have just completed a deliverable for the project NFFA-Europe (www.nffa.eu) where we have
defined a core set of concepts and fields for experiments. The document is not quite released as yet –
I’ll send a link as soon as it is, but I’ll be talking about it at the joint Materials session this afternoon.
It’s based on our earlier work on Core Scientific MetaData for facilities data.
http://purl.org/net/csmd
thanks
Brian
From: steve.richard=***@***.***-groups.org [mailto:***@***.***-groups.org] On Behalf Of srichardUSGIN
Sent: 01 March 2016 22:52
To: RayPlante; Metadata IG
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
Ray—thanks for compiling this. I wasn’t at the session yesterday, but looking at the slides I had the same reaction. I’ve taken the liberty of making a google doc spreadsheet from your Excel to make it easier to collect comments and discussion. The resource definition is on the first tab, metadata element definitions on the second tab. At this point if people want, I’d suggest adding a new column for each reviewer. Comments can also be added using the google comments.
the spreadsheet is at
https://docs.google.com/spreadsheets/d/18Wi2GNM10d3e_iDWgDS-w4M4Zeaei1wO...
permissions are set up so anyone with the link can edit.
the doc is also published at
https://docs.google.com/spreadsheets/d/18Wi2GNM10d3e_iDWgDS-w4M4Zeaei1wO...
with that link you can only view, so you can share for read-only access.
steve
- Show quoted text -From: raymond.plante=***@***.***-groups.org [mailto:***@***.***-groups.org] On Behalf Of RayPlante
Sent: Wednesday, March 02, 2016 6:45 AM
To: Metadata IG <***@***.***-groups.org>
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
Dear MIG,
At yesterday's joint metadata session, I felt that a major barrier to effectively discussing the proposed elements--i.e. evaluating as to whether the elements were "the right set"--was the lack of definitions for concepts they are to represent. In an effort improve the discussion, I offer a strawman set of definitions for each of the terms in slide 9 ("Open Data: Purposing the elements") in a spreadsheet posted at https://rd-alliance.org/metadata-p7-tokyo.html. This was a useful exercise at least for me as I realized that I really didn't understand the intended concept behind a few of the elements.
Another thing that I think is important to clarify is the overall model for the things we are trying to describe with these elements. This is covered in part by block diagram in the slides that were presented, but I suspect more should be spelled out. As context for my posted defiitions, I assumed that these elements are (complex) properties describing a thing I refer to as a Resource; a more precise definition is given within that spreadsheet. I will also note that my own thinking comes from work I'm currently doing as part of the Materials Science Interest Group (and the proposed WG on resource registries for Material Science); a conceptual model is describe here: http://wiki.nationaldataservice.org/ResourceConceptDataModel.
hope this is helpful,
Ray
--
Full post: https://rd-alliance.org/group/metadata-ig/post/slides-and-notes-joint-me...
Manage my subscriptions: https://rd-alliance.org/mailinglist
Stop emails for this post: https://rd-alliance.org/mailinglist/unsubscribe/51502

Steve –
Thanks for the google doc spreadsheet. I have added a column in an attempt to clarify (but I may well have confused!). I suspect we need an e-meeting after a suitable time interval since it is easier to express things that way than in a spreadsheet – although the spreadsheet is a great reference document.
Best
Keith
Keith G Jeffery Consultants
Prof Keith G Jeffery
E: ***@***.***
T: +44 7768 446088
S: keithgjeffery
Past President ERCIM www.ercim.eu (***@***.***)
Past President euroCRIS www.eurocris.org
Past Vice President VLDB www.vldb.org
Fellow (CITP, CEng) BCS www.bcs.org
Co-chair RDA MIG https://rd-alliance.org/internal-groups/metadata-ig.html
Co-chair RDA MSDWG https://rd-alliance.org/working-groups/metadata-standards-directory-work...
Co-chair RDA DICIG https://rd-alliance.org/internal-groups/data-context-ig.html
----------------------------------------------------------------------------------------------------------------------------------
The contents of this email are sent in confidence for the use of the
intended recipient only. If you are not one of the intended
recipients do not take action on it or show it to anyone else, but
return this email to the sender and delete your copy of it.
----------------------------------------------------------------------------------------------------------------------------------
- Show quoted text -From: steve.richard=***@***.***-groups.org [mailto:***@***.***-groups.org] On Behalf Of srichardUSGIN
Sent: 01 March 2016 22:52
To: RayPlante; Metadata IG
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
Ray—thanks for compiling this. I wasn’t at the session yesterday, but looking at the slides I had the same reaction. I’ve taken the liberty of making a google doc spreadsheet from your Excel to make it easier to collect comments and discussion. The resource definition is on the first tab, metadata element definitions on the second tab. At this point if people want, I’d suggest adding a new column for each reviewer. Comments can also be added using the google comments.
the spreadsheet is at
https://docs.google.com/spreadsheets/d/18Wi2GNM10d3e_iDWgDS-w4M4Zeaei1wO...
permissions are set up so anyone with the link can edit.
the doc is also published at
https://docs.google.com/spreadsheets/d/18Wi2GNM10d3e_iDWgDS-w4M4Zeaei1wO...
with that link you can only view, so you can share for read-only access.
steve
From: raymond.plante=***@***.***-groups.org [mailto:***@***.***-groups.org] On Behalf Of RayPlante
Sent: Wednesday, March 02, 2016 6:45 AM
To: Metadata IG <***@***.***-groups.org>
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
Dear MIG,
At yesterday's joint metadata session, I felt that a major barrier to effectively discussing the proposed elements--i.e. evaluating as to whether the elements were "the right set"--was the lack of definitions for concepts they are to represent. In an effort improve the discussion, I offer a strawman set of definitions for each of the terms in slide 9 ("Open Data: Purposing the elements") in a spreadsheet posted at https://rd-alliance.org/metadata-p7-tokyo.html. This was a useful exercise at least for me as I realized that I really didn't understand the intended concept behind a few of the elements.
Another thing that I think is important to clarify is the overall model for the things we are trying to describe with these elements. This is covered in part by block diagram in the slides that were presented, but I suspect more should be spelled out. As context for my posted defiitions, I assumed that these elements are (complex) properties describing a thing I refer to as a Resource; a more precise definition is given within that spreadsheet. I will also note that my own thinking comes from work I'm currently doing as part of the Materials Science Interest Group (and the proposed WG on resource registries for Material Science); a conceptual model is describe here: http://wiki.nationaldataservice.org/ResourceConceptDataModel.
hope this is helpful,
Ray
--
Full post: https://rd-alliance.org/group/metadata-ig/post/slides-and-notes-joint-me...
Manage my subscriptions: https://rd-alliance.org/mailinglist
Stop emails for this post: https://rd-alliance.org/mailinglist/unsubscribe/51502
Steve –
Thanks for the google doc spreadsheet. I have added a column in an attempt to clarify (but I may well have confused!). I suspect we need an e-meeting after a suitable time interval since it is easier to express things that way than in a spreadsheet – although the spreadsheet is a great reference document.
Best
Keith
Keith G Jeffery Consultants
Prof Keith G Jeffery
E: ***@***.***
T: +44 7768 446088
S: keithgjeffery
Past President ERCIM www.ercim.eu (***@***.***)
Past President euroCRIS www.eurocris.org
Past Vice President VLDB www.vldb.org
Fellow (CITP, CEng) BCS www.bcs.org
Co-chair RDA MIG https://rd-alliance.org/internal-groups/metadata-ig.html
Co-chair RDA MSDWG https://rd-alliance.org/working-groups/metadata-standards-directory-work...
Co-chair RDA DICIG https://rd-alliance.org/internal-groups/data-context-ig.html
----------------------------------------------------------------------------------------------------------------------------------
The contents of this email are sent in confidence for the use of the
intended recipient only. If you are not one of the intended
recipients do not take action on it or show it to anyone else, but
return this email to the sender and delete your copy of it.
----------------------------------------------------------------------------------------------------------------------------------
From: steve.richard=***@***.***-groups.org [mailto:***@***.***-groups.org] On Behalf Of srichardUSGIN
Sent: 01 March 2016 22:52
To: RayPlante; Metadata IG
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
Ray—thanks for compiling this. I wasn’t at the session yesterday, but looking at the slides I had the same reaction. I’ve taken the liberty of making a google doc spreadsheet from your Excel to make it easier to collect comments and discussion. The resource definition is on the first tab, metadata element definitions on the second tab. At this point if people want, I’d suggest adding a new column for each reviewer. Comments can also be added using the google comments.
the spreadsheet is at
https://docs.google.com/spreadsheets/d/18Wi2GNM10d3e_iDWgDS-w4M4Zeaei1wO...
permissions are set up so anyone with the link can edit.
the doc is also published at
https://docs.google.com/spreadsheets/d/18Wi2GNM10d3e_iDWgDS-w4M4Zeaei1wO...
with that link you can only view, so you can share for read-only access.
steve
- Show quoted text -From: raymond.plante=***@***.***-groups.org [mailto:***@***.***-groups.org] On Behalf Of RayPlante
Sent: Wednesday, March 02, 2016 6:45 AM
To: Metadata IG <***@***.***-groups.org>
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
Dear MIG,
At yesterday's joint metadata session, I felt that a major barrier to effectively discussing the proposed elements--i.e. evaluating as to whether the elements were "the right set"--was the lack of definitions for concepts they are to represent. In an effort improve the discussion, I offer a strawman set of definitions for each of the terms in slide 9 ("Open Data: Purposing the elements") in a spreadsheet posted at https://rd-alliance.org/metadata-p7-tokyo.html. This was a useful exercise at least for me as I realized that I really didn't understand the intended concept behind a few of the elements.
Another thing that I think is important to clarify is the overall model for the things we are trying to describe with these elements. This is covered in part by block diagram in the slides that were presented, but I suspect more should be spelled out. As context for my posted defiitions, I assumed that these elements are (complex) properties describing a thing I refer to as a Resource; a more precise definition is given within that spreadsheet. I will also note that my own thinking comes from work I'm currently doing as part of the Materials Science Interest Group (and the proposed WG on resource registries for Material Science); a conceptual model is describe here: http://wiki.nationaldataservice.org/ResourceConceptDataModel.
hope this is helpful,
Ray
--
Full post: https://rd-alliance.org/group/metadata-ig/post/slides-and-notes-joint-me...
Manage my subscriptions: https://rd-alliance.org/mailinglist
Stop emails for this post: https://rd-alliance.org/mailinglist/unsubscribe/51502

What we have done identifies a number of base concepts which are either:
- Part of the experimental context – e.g. people (in various roles) facility, instrument, project, sample (a key entity in our world), measurements or
- Describe “Data Assets” and their management – e.g. data sets, data files, archives
These conceptual entities then can have attributes such as titles, identifiers (universally!), timestamps, formats etc.
The core list in the spreadsheet mixes up the “conceptual elements” and the “attribute elements” – understandably as this is a list.
One of Keith’s diagrams does distinguish these.
When I get a moment I’ll comment on the spreadsheet.
I’ll try to arrange a summary of our draft specification to distribute.
One thing that we have done also is begin to give an “abstract workflow for experimental science”.
I don’t believe that you can clearly define (contextual) metadata without a model of what you are describing.
Thanks
Brian
- Show quoted text -From: steve.richard=***@***.***-groups.org [mailto:***@***.***-groups.org] On Behalf Of srichardUSGIN
Sent: 01 March 2016 23:45
To: Matthews, Brian (STFC,RAL,SC); ***@***.***-groups.org
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
Brian—Excellent! How does your analysis compare with what’s in the spreadsheet?
From: ***@***.*** [mailto:***@***.***]
Sent: Wednesday, March 02, 2016 8:14 AM
To: Steve Richard <***@***.***>; ***@***.***; ***@***.***-groups.org
Cc: ***@***.***
Subject: RE: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
We (that is in STFC) have also been working on similar models in the context of facilities and nano-materials data.
We have just completed a deliverable for the project NFFA-Europe (www.nffa.eu) where we have
defined a core set of concepts and fields for experiments. The document is not quite released as yet –
I’ll send a link as soon as it is, but I’ll be talking about it at the joint Materials session this afternoon.
It’s based on our earlier work on Core Scientific MetaData for facilities data.
http://purl.org/net/csmd
thanks
Brian
From: steve.richard=***@***.***-groups.org [mailto:***@***.***-groups.org] On Behalf Of srichardUSGIN
Sent: 01 March 2016 22:52
To: RayPlante; Metadata IG
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
Ray—thanks for compiling this. I wasn’t at the session yesterday, but looking at the slides I had the same reaction. I’ve taken the liberty of making a google doc spreadsheet from your Excel to make it easier to collect comments and discussion. The resource definition is on the first tab, metadata element definitions on the second tab. At this point if people want, I’d suggest adding a new column for each reviewer. Comments can also be added using the google comments.
the spreadsheet is at
https://docs.google.com/spreadsheets/d/18Wi2GNM10d3e_iDWgDS-w4M4Zeaei1wO...
permissions are set up so anyone with the link can edit.
the doc is also published at
https://docs.google.com/spreadsheets/d/18Wi2GNM10d3e_iDWgDS-w4M4Zeaei1wO...
with that link you can only view, so you can share for read-only access.
steve
From: raymond.plante=***@***.***-groups.org [mailto:***@***.***-groups.org] On Behalf Of RayPlante
Sent: Wednesday, March 02, 2016 6:45 AM
To: Metadata IG <***@***.***-groups.org>
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
Dear MIG,
At yesterday's joint metadata session, I felt that a major barrier to effectively discussing the proposed elements--i.e. evaluating as to whether the elements were "the right set"--was the lack of definitions for concepts they are to represent. In an effort improve the discussion, I offer a strawman set of definitions for each of the terms in slide 9 ("Open Data: Purposing the elements") in a spreadsheet posted at https://rd-alliance.org/metadata-p7-tokyo.html. This was a useful exercise at least for me as I realized that I really didn't understand the intended concept behind a few of the elements.
Another thing that I think is important to clarify is the overall model for the things we are trying to describe with these elements. This is covered in part by block diagram in the slides that were presented, but I suspect more should be spelled out. As context for my posted defiitions, I assumed that these elements are (complex) properties describing a thing I refer to as a Resource; a more precise definition is given within that spreadsheet. I will also note that my own thinking comes from work I'm currently doing as part of the Materials Science Interest Group (and the proposed WG on resource registries for Material Science); a conceptual model is describe here: http://wiki.nationaldataservice.org/ResourceConceptDataModel.
hope this is helpful,
Ray
--
Full post: https://rd-alliance.org/group/metadata-ig/post/slides-and-notes-joint-me...
Manage my subscriptions: https://rd-alliance.org/mailinglist
Stop emails for this post: https://rd-alliance.org/mailinglist/unsubscribe/51502
What we have done identifies a number of base concepts which are either:
- Part of the experimental context – e.g. people (in various roles) facility, instrument, project, sample (a key entity in our world), measurements or
- Describe “Data Assets” and their management – e.g. data sets, data files, archives
These conceptual entities then can have attributes such as titles, identifiers (universally!), timestamps, formats etc.
The core list in the spreadsheet mixes up the “conceptual elements” and the “attribute elements” – understandably as this is a list.
One of Keith’s diagrams does distinguish these.
When I get a moment I’ll comment on the spreadsheet.
I’ll try to arrange a summary of our draft specification to distribute.
One thing that we have done also is begin to give an “abstract workflow for experimental science”.
I don’t believe that you can clearly define (contextual) metadata without a model of what you are describing.
Thanks
Brian
From: steve.richard=***@***.***-groups.org [mailto:***@***.***-groups.org] On Behalf Of srichardUSGIN
Sent: 01 March 2016 23:45
To: Matthews, Brian (STFC,RAL,SC); ***@***.***-groups.org
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
Brian—Excellent! How does your analysis compare with what’s in the spreadsheet?
From: ***@***.*** [mailto:***@***.***]
Sent: Wednesday, March 02, 2016 8:14 AM
To: Steve Richard <***@***.***>; ***@***.***; ***@***.***-groups.org
Cc: ***@***.***
Subject: RE: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
We (that is in STFC) have also been working on similar models in the context of facilities and nano-materials data.
We have just completed a deliverable for the project NFFA-Europe (www.nffa.eu) where we have
defined a core set of concepts and fields for experiments. The document is not quite released as yet –
I’ll send a link as soon as it is, but I’ll be talking about it at the joint Materials session this afternoon.
It’s based on our earlier work on Core Scientific MetaData for facilities data.
http://purl.org/net/csmd
thanks
Brian
- Show quoted text -From: steve.richard=***@***.***-groups.org [mailto:***@***.***-groups.org] On Behalf Of srichardUSGIN
Sent: 01 March 2016 22:52
To: RayPlante; Metadata IG
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
Ray—thanks for compiling this. I wasn’t at the session yesterday, but looking at the slides I had the same reaction. I’ve taken the liberty of making a google doc spreadsheet from your Excel to make it easier to collect comments and discussion. The resource definition is on the first tab, metadata element definitions on the second tab. At this point if people want, I’d suggest adding a new column for each reviewer. Comments can also be added using the google comments.
the spreadsheet is at
https://docs.google.com/spreadsheets/d/18Wi2GNM10d3e_iDWgDS-w4M4Zeaei1wO...
permissions are set up so anyone with the link can edit.
the doc is also published at
https://docs.google.com/spreadsheets/d/18Wi2GNM10d3e_iDWgDS-w4M4Zeaei1wO...
with that link you can only view, so you can share for read-only access.
steve
From: raymond.plante=***@***.***-groups.org [mailto:***@***.***-groups.org] On Behalf Of RayPlante
Sent: Wednesday, March 02, 2016 6:45 AM
To: Metadata IG <***@***.***-groups.org>
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
Dear MIG,
At yesterday's joint metadata session, I felt that a major barrier to effectively discussing the proposed elements--i.e. evaluating as to whether the elements were "the right set"--was the lack of definitions for concepts they are to represent. In an effort improve the discussion, I offer a strawman set of definitions for each of the terms in slide 9 ("Open Data: Purposing the elements") in a spreadsheet posted at https://rd-alliance.org/metadata-p7-tokyo.html. This was a useful exercise at least for me as I realized that I really didn't understand the intended concept behind a few of the elements.
Another thing that I think is important to clarify is the overall model for the things we are trying to describe with these elements. This is covered in part by block diagram in the slides that were presented, but I suspect more should be spelled out. As context for my posted defiitions, I assumed that these elements are (complex) properties describing a thing I refer to as a Resource; a more precise definition is given within that spreadsheet. I will also note that my own thinking comes from work I'm currently doing as part of the Materials Science Interest Group (and the proposed WG on resource registries for Material Science); a conceptual model is describe here: http://wiki.nationaldataservice.org/ResourceConceptDataModel.
hope this is helpful,
Ray
--
Full post: https://rd-alliance.org/group/metadata-ig/post/slides-and-notes-joint-me...
Manage my subscriptions: https://rd-alliance.org/mailinglist
Stop emails for this post: https://rd-alliance.org/mailinglist/unsubscribe/51502
What we have done identifies a number of base concepts which are either:
- Part of the experimental context – e.g. people (in various roles) facility, instrument, project, sample (a key entity in our world), measurements or
- Describe “Data Assets” and their management – e.g. data sets, data files, archives
These conceptual entities then can have attributes such as titles, identifiers (universally!), timestamps, formats etc.
The core list in the spreadsheet mixes up the “conceptual elements” and the “attribute elements” – understandably as this is a list.
One of Keith’s diagrams does distinguish these.
When I get a moment I’ll comment on the spreadsheet.
I’ll try to arrange a summary of our draft specification to distribute.
One thing that we have done also is begin to give an “abstract workflow for experimental science”.
I don’t believe that you can clearly define (contextual) metadata without a model of what you are describing.
Thanks
Brian
From: steve.richard=***@***.***-groups.org [mailto:***@***.***-groups.org] On Behalf Of srichardUSGIN
Sent: 01 March 2016 23:45
To: Matthews, Brian (STFC,RAL,SC); ***@***.***-groups.org
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
Brian—Excellent! How does your analysis compare with what’s in the spreadsheet?
From: ***@***.*** [mailto:***@***.***]
Sent: Wednesday, March 02, 2016 8:14 AM
To: Steve Richard <***@***.***>; ***@***.***; ***@***.***-groups.org
Cc: ***@***.***
Subject: RE: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
We (that is in STFC) have also been working on similar models in the context of facilities and nano-materials data.
We have just completed a deliverable for the project NFFA-Europe (www.nffa.eu) where we have
defined a core set of concepts and fields for experiments. The document is not quite released as yet –
I’ll send a link as soon as it is, but I’ll be talking about it at the joint Materials session this afternoon.
It’s based on our earlier work on Core Scientific MetaData for facilities data.
http://purl.org/net/csmd
thanks
Brian
From: steve.richard=***@***.***-groups.org [mailto:***@***.***-groups.org] On Behalf Of srichardUSGIN
Sent: 01 March 2016 22:52
To: RayPlante; Metadata IG
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
Ray—thanks for compiling this. I wasn’t at the session yesterday, but looking at the slides I had the same reaction. I’ve taken the liberty of making a google doc spreadsheet from your Excel to make it easier to collect comments and discussion. The resource definition is on the first tab, metadata element definitions on the second tab. At this point if people want, I’d suggest adding a new column for each reviewer. Comments can also be added using the google comments.
the spreadsheet is at
https://docs.google.com/spreadsheets/d/18Wi2GNM10d3e_iDWgDS-w4M4Zeaei1wO...
permissions are set up so anyone with the link can edit.
the doc is also published at
https://docs.google.com/spreadsheets/d/18Wi2GNM10d3e_iDWgDS-w4M4Zeaei1wO...
with that link you can only view, so you can share for read-only access.
steve
- Show quoted text -From: raymond.plante=***@***.***-groups.org [mailto:***@***.***-groups.org] On Behalf Of RayPlante
Sent: Wednesday, March 02, 2016 6:45 AM
To: Metadata IG <***@***.***-groups.org>
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
Dear MIG,
At yesterday's joint metadata session, I felt that a major barrier to effectively discussing the proposed elements--i.e. evaluating as to whether the elements were "the right set"--was the lack of definitions for concepts they are to represent. In an effort improve the discussion, I offer a strawman set of definitions for each of the terms in slide 9 ("Open Data: Purposing the elements") in a spreadsheet posted at https://rd-alliance.org/metadata-p7-tokyo.html. This was a useful exercise at least for me as I realized that I really didn't understand the intended concept behind a few of the elements.
Another thing that I think is important to clarify is the overall model for the things we are trying to describe with these elements. This is covered in part by block diagram in the slides that were presented, but I suspect more should be spelled out. As context for my posted defiitions, I assumed that these elements are (complex) properties describing a thing I refer to as a Resource; a more precise definition is given within that spreadsheet. I will also note that my own thinking comes from work I'm currently doing as part of the Materials Science Interest Group (and the proposed WG on resource registries for Material Science); a conceptual model is describe here: http://wiki.nationaldataservice.org/ResourceConceptDataModel.
hope this is helpful,
Ray
--
Full post: https://rd-alliance.org/group/metadata-ig/post/slides-and-notes-joint-me...
Manage my subscriptions: https://rd-alliance.org/mailinglist
Stop emails for this post: https://rd-alliance.org/mailinglist/unsubscribe/51502

Hey Steve -- thanks for the Google docs version--very helpful.
Hi Brian--nice model for the CSMD. Borrowing some of that detail would useful for the more general model.
________________________________
From: brian.matthews=***@***.***-groups.org <***@***.***-groups.org> on behalf of brianmatthews <***@***.***>
Sent: Tuesday, March 1, 2016 8:33 PM
To: ***@***.***; ***@***.***-groups.org
Cc: ***@***.***
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
What we have done identifies a number of base concepts which are either:
- Part of the experimental context – e.g. people (in various roles) facility, instrument, project, sample (a key entity in our world), measurements or
- Describe “Data Assets” and their management – e.g. data sets, data files, archives
These conceptual entities then can have attributes such as titles, identifiers (universally!), timestamps, formats etc.
The core list in the spreadsheet mixes up the “conceptual elements” and the “attribute elements” – understandably as this is a list.
One of Keith’s diagrams does distinguish these.
When I get a moment I’ll comment on the spreadsheet.
I’ll try to arrange a summary of our draft specification to distribute.
One thing that we have done also is begin to give an “abstract workflow for experimental science”.
I don’t believe that you can clearly define (contextual) metadata without a model of what you are describing.
Thanks
Brian
- Show quoted text -From: steve.richard=***@***.***-groups.org [mailto:***@***.***-groups.org] On Behalf Of srichardUSGIN
Sent: 01 March 2016 23:45
To: Matthews, Brian (STFC,RAL,SC); ***@***.***-groups.org
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
Brian—Excellent! How does your analysis compare with what’s in the spreadsheet?
From: ***@***.*** [mailto:***@***.***]
Sent: Wednesday, March 02, 2016 8:14 AM
To: Steve Richard <***@***.***>; ***@***.***; ***@***.***-groups.org
Cc: ***@***.***
Subject: RE: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
We (that is in STFC) have also been working on similar models in the context of facilities and nano-materials data.
We have just completed a deliverable for the project NFFA-Europe (www.nffa.eu) where we have
defined a core set of concepts and fields for experiments. The document is not quite released as yet –
I’ll send a link as soon as it is, but I’ll be talking about it at the joint Materials session this afternoon.
It’s based on our earlier work on Core Scientific MetaData for facilities data.
http://purl.org/net/csmd
thanks
Brian
From: steve.richard=***@***.***-groups.org [mailto:***@***.***-groups.org] On Behalf Of srichardUSGIN
Sent: 01 March 2016 22:52
To: RayPlante; Metadata IG
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
Ray—thanks for compiling this. I wasn’t at the session yesterday, but looking at the slides I had the same reaction. I’ve taken the liberty of making a google doc spreadsheet from your Excel to make it easier to collect comments and discussion. The resource definition is on the first tab, metadata element definitions on the second tab. At this point if people want, I’d suggest adding a new column for each reviewer. Comments can also be added using the google comments.
the spreadsheet is at
https://docs.google.com/spreadsheets/d/18Wi2GNM10d3e_iDWgDS-w4M4Zeaei1wO...
permissions are set up so anyone with the link can edit.
the doc is also published at
https://docs.google.com/spreadsheets/d/18Wi2GNM10d3e_iDWgDS-w4M4Zeaei1wO...
with that link you can only view, so you can share for read-only access.
steve
From: raymond.plante=***@***.***-groups.org [mailto:***@***.***-groups.org] On Behalf Of RayPlante
Sent: Wednesday, March 02, 2016 6:45 AM
To: Metadata IG <***@***.***-groups.org>
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
Dear MIG,
At yesterday's joint metadata session, I felt that a major barrier to effectively discussing the proposed elements--i.e. evaluating as to whether the elements were "the right set"--was the lack of definitions for concepts they are to represent. In an effort improve the discussion, I offer a strawman set of definitions for each of the terms in slide 9 ("Open Data: Purposing the elements") in a spreadsheet posted at https://rd-alliance.org/metadata-p7-tokyo.html. This was a useful exercise at least for me as I realized that I really didn't understand the intended concept behind a few of the elements.
Another thing that I think is important to clarify is the overall model for the things we are trying to describe with these elements. This is covered in part by block diagram in the slides that were presented, but I suspect more should be spelled out. As context for my posted defiitions, I assumed that these elements are (complex) properties describing a thing I refer to as a Resource; a more precise definition is given within that spreadsheet. I will also note that my own thinking comes from work I'm currently doing as part of the Materials Science Interest Group (and the proposed WG on resource registries for Material Science); a conceptual model is describe here: http://wiki.nationaldataservice.org/ResourceConceptDataModel.
hope this is helpful,
Ray
--
Full post: https://rd-alliance.org/group/metadata-ig/post/slides-and-notes-joint-me...
Manage my subscriptions: https://rd-alliance.org/mailinglist
Stop emails for this post: https://rd-alliance.org/mailinglist/unsubscribe/51502
Hey Steve -- thanks for the Google docs version--very helpful.
Hi Brian--nice model for the CSMD. Borrowing some of that detail would useful for the more general model.
________________________________
From: brian.matthews=***@***.***-groups.org <***@***.***-groups.org> on behalf of brianmatthews <***@***.***>
Sent: Tuesday, March 1, 2016 8:33 PM
To: ***@***.***; ***@***.***-groups.org
Cc: ***@***.***
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
What we have done identifies a number of base concepts which are either:
- Part of the experimental context – e.g. people (in various roles) facility, instrument, project, sample (a key entity in our world), measurements or
- Describe “Data Assets” and their management – e.g. data sets, data files, archives
These conceptual entities then can have attributes such as titles, identifiers (universally!), timestamps, formats etc.
The core list in the spreadsheet mixes up the “conceptual elements” and the “attribute elements” – understandably as this is a list.
One of Keith’s diagrams does distinguish these.
When I get a moment I’ll comment on the spreadsheet.
I’ll try to arrange a summary of our draft specification to distribute.
One thing that we have done also is begin to give an “abstract workflow for experimental science”.
I don’t believe that you can clearly define (contextual) metadata without a model of what you are describing.
Thanks
Brian
From: steve.richard=***@***.***-groups.org [mailto:***@***.***-groups.org] On Behalf Of srichardUSGIN
Sent: 01 March 2016 23:45
To: Matthews, Brian (STFC,RAL,SC); ***@***.***-groups.org
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
Brian—Excellent! How does your analysis compare with what’s in the spreadsheet?
From: ***@***.*** [mailto:***@***.***]
Sent: Wednesday, March 02, 2016 8:14 AM
To: Steve Richard <***@***.***>; ***@***.***; ***@***.***-groups.org
Cc: ***@***.***
Subject: RE: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
We (that is in STFC) have also been working on similar models in the context of facilities and nano-materials data.
We have just completed a deliverable for the project NFFA-Europe (www.nffa.eu) where we have
defined a core set of concepts and fields for experiments. The document is not quite released as yet –
I’ll send a link as soon as it is, but I’ll be talking about it at the joint Materials session this afternoon.
It’s based on our earlier work on Core Scientific MetaData for facilities data.
http://purl.org/net/csmd
thanks
Brian
- Show quoted text -From: steve.richard=***@***.***-groups.org [mailto:***@***.***-groups.org] On Behalf Of srichardUSGIN
Sent: 01 March 2016 22:52
To: RayPlante; Metadata IG
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
Ray—thanks for compiling this. I wasn’t at the session yesterday, but looking at the slides I had the same reaction. I’ve taken the liberty of making a google doc spreadsheet from your Excel to make it easier to collect comments and discussion. The resource definition is on the first tab, metadata element definitions on the second tab. At this point if people want, I’d suggest adding a new column for each reviewer. Comments can also be added using the google comments.
the spreadsheet is at
https://docs.google.com/spreadsheets/d/18Wi2GNM10d3e_iDWgDS-w4M4Zeaei1wO...
permissions are set up so anyone with the link can edit.
the doc is also published at
https://docs.google.com/spreadsheets/d/18Wi2GNM10d3e_iDWgDS-w4M4Zeaei1wO...
with that link you can only view, so you can share for read-only access.
steve
From: raymond.plante=***@***.***-groups.org [mailto:***@***.***-groups.org] On Behalf Of RayPlante
Sent: Wednesday, March 02, 2016 6:45 AM
To: Metadata IG <***@***.***-groups.org>
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
Dear MIG,
At yesterday's joint metadata session, I felt that a major barrier to effectively discussing the proposed elements--i.e. evaluating as to whether the elements were "the right set"--was the lack of definitions for concepts they are to represent. In an effort improve the discussion, I offer a strawman set of definitions for each of the terms in slide 9 ("Open Data: Purposing the elements") in a spreadsheet posted at https://rd-alliance.org/metadata-p7-tokyo.html. This was a useful exercise at least for me as I realized that I really didn't understand the intended concept behind a few of the elements.
Another thing that I think is important to clarify is the overall model for the things we are trying to describe with these elements. This is covered in part by block diagram in the slides that were presented, but I suspect more should be spelled out. As context for my posted defiitions, I assumed that these elements are (complex) properties describing a thing I refer to as a Resource; a more precise definition is given within that spreadsheet. I will also note that my own thinking comes from work I'm currently doing as part of the Materials Science Interest Group (and the proposed WG on resource registries for Material Science); a conceptual model is describe here: http://wiki.nationaldataservice.org/ResourceConceptDataModel.
hope this is helpful,
Ray
--
Full post: https://rd-alliance.org/group/metadata-ig/post/slides-and-notes-joint-me...
Manage my subscriptions: https://rd-alliance.org/mailinglist
Stop emails for this post: https://rd-alliance.org/mailinglist/unsubscribe/51502
Hey Steve -- thanks for the Google docs version--very helpful.
Hi Brian--nice model for the CSMD. Borrowing some of that detail would useful for the more general model.
________________________________
From: brian.matthews=***@***.***-groups.org <***@***.***-groups.org> on behalf of brianmatthews <***@***.***>
Sent: Tuesday, March 1, 2016 8:33 PM
To: ***@***.***; ***@***.***-groups.org
Cc: ***@***.***
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
What we have done identifies a number of base concepts which are either:
- Part of the experimental context – e.g. people (in various roles) facility, instrument, project, sample (a key entity in our world), measurements or
- Describe “Data Assets” and their management – e.g. data sets, data files, archives
These conceptual entities then can have attributes such as titles, identifiers (universally!), timestamps, formats etc.
The core list in the spreadsheet mixes up the “conceptual elements” and the “attribute elements” – understandably as this is a list.
One of Keith’s diagrams does distinguish these.
When I get a moment I’ll comment on the spreadsheet.
I’ll try to arrange a summary of our draft specification to distribute.
One thing that we have done also is begin to give an “abstract workflow for experimental science”.
I don’t believe that you can clearly define (contextual) metadata without a model of what you are describing.
Thanks
Brian
From: steve.richard=***@***.***-groups.org [mailto:***@***.***-groups.org] On Behalf Of srichardUSGIN
Sent: 01 March 2016 23:45
To: Matthews, Brian (STFC,RAL,SC); ***@***.***-groups.org
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
Brian—Excellent! How does your analysis compare with what’s in the spreadsheet?
From: ***@***.*** [mailto:***@***.***]
Sent: Wednesday, March 02, 2016 8:14 AM
To: Steve Richard <***@***.***>; ***@***.***; ***@***.***-groups.org
Cc: ***@***.***
Subject: RE: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
We (that is in STFC) have also been working on similar models in the context of facilities and nano-materials data.
We have just completed a deliverable for the project NFFA-Europe (www.nffa.eu) where we have
defined a core set of concepts and fields for experiments. The document is not quite released as yet –
I’ll send a link as soon as it is, but I’ll be talking about it at the joint Materials session this afternoon.
It’s based on our earlier work on Core Scientific MetaData for facilities data.
http://purl.org/net/csmd
thanks
Brian
From: steve.richard=***@***.***-groups.org [mailto:***@***.***-groups.org] On Behalf Of srichardUSGIN
Sent: 01 March 2016 22:52
To: RayPlante; Metadata IG
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
Ray—thanks for compiling this. I wasn’t at the session yesterday, but looking at the slides I had the same reaction. I’ve taken the liberty of making a google doc spreadsheet from your Excel to make it easier to collect comments and discussion. The resource definition is on the first tab, metadata element definitions on the second tab. At this point if people want, I’d suggest adding a new column for each reviewer. Comments can also be added using the google comments.
the spreadsheet is at
https://docs.google.com/spreadsheets/d/18Wi2GNM10d3e_iDWgDS-w4M4Zeaei1wO...
permissions are set up so anyone with the link can edit.
the doc is also published at
https://docs.google.com/spreadsheets/d/18Wi2GNM10d3e_iDWgDS-w4M4Zeaei1wO...
with that link you can only view, so you can share for read-only access.
steve
- Show quoted text -From: raymond.plante=***@***.***-groups.org [mailto:***@***.***-groups.org] On Behalf Of RayPlante
Sent: Wednesday, March 02, 2016 6:45 AM
To: Metadata IG <***@***.***-groups.org>
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
Dear MIG,
At yesterday's joint metadata session, I felt that a major barrier to effectively discussing the proposed elements--i.e. evaluating as to whether the elements were "the right set"--was the lack of definitions for concepts they are to represent. In an effort improve the discussion, I offer a strawman set of definitions for each of the terms in slide 9 ("Open Data: Purposing the elements") in a spreadsheet posted at https://rd-alliance.org/metadata-p7-tokyo.html. This was a useful exercise at least for me as I realized that I really didn't understand the intended concept behind a few of the elements.
Another thing that I think is important to clarify is the overall model for the things we are trying to describe with these elements. This is covered in part by block diagram in the slides that were presented, but I suspect more should be spelled out. As context for my posted defiitions, I assumed that these elements are (complex) properties describing a thing I refer to as a Resource; a more precise definition is given within that spreadsheet. I will also note that my own thinking comes from work I'm currently doing as part of the Materials Science Interest Group (and the proposed WG on resource registries for Material Science); a conceptual model is describe here: http://wiki.nationaldataservice.org/ResourceConceptDataModel.
hope this is helpful,
Ray
--
Full post: https://rd-alliance.org/group/metadata-ig/post/slides-and-notes-joint-me...
Manage my subscriptions: https://rd-alliance.org/mailinglist
Stop emails for this post: https://rd-alliance.org/mailinglist/unsubscribe/51502

Hey Vasily,
Very insightful post; +1 on your 2 realizations when approaching metadata design.
I was intrigued by your discussion of the annotation approach (I'll spend some time on the plane trying to digest your paper). I was wondering you might share a few sentences toward "annotations for dummies"--not so much that I'm completely unfamiliar with the term as much as I've encountered several variations on a theme bearing this name. You might comment, for instance, if this is an approach that allows people other than a provider of some resource to make statements about it and have those characterizations be part of the globally accessible description of it. Does it sit firmly in the ontological approach to metadata development? I gather from your discussion of "sentences" (akin to RDF statements), the answer is yes.
In as much as I understand your perspective, the annotation approach sounds like something that can exist and be effective as a complement to some "core" description provided by the resource provider, particularly if, as you suggest, that the core schema is not so expansive.
So I bring my own skepticism to metadata development efforts, despite the fact I'm engaged that very activity for Materials Science. I worry about whether a common metadata schema can simultaneously address the needs of high-level discovery and lower-level analysis. My interest in non-disruptive metadata extensibility is an attempt to address this concern: let's start off small and extend as needed by real use cases.
And though it appeals to my info-nerd side, I also worry about the practicality of extracting information (let alone "knowledge") from sentences (is this like RDF triple statements?) that obscures any structure that a mere scientist could wrap their head around. Does getting anything useful out of a big graph of statements require a) learning a still-obscure query language, b) understanding of a vast or numerous ontologies, or c) a really smart tool (that I've yet to see)?
That said, the sentence approach is probably more flexible to restructuring for addressing both high-level discovery and low-level analysis. (Did that make sense?)
Anyway, I hope you can pardon the post-conference dinner thoughts. FWIW, it helps to understand where we think we're going (and what kind of work it will take to get there).
cheers,
Ray
________________________________
From: ***@***.*** <***@***.***>
Sent: Wednesday, March 2, 2016 6:12 AM
To: Plante, Raymond L. (Fed); ***@***.***-groups.org
Cc: ***@***.***
Subject: RE: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
Brian rightly mentioned the further thinking we have done over CSMD for the metadata design in NFFA project.
One thing we realized doing that work was the need in clearly defined semantics behind metadata elements (the problem highlighted by Ray), so we iteratively discussed and produced a Common Vocabulary – which IMHO should be a recommended practice for every newly created or updated metadata model.
Another thing we realized was the need of having an “operational” perspective when you design metadata and when you actually apply metadata model in practice. Producing “how to use” guidelines should be considered another recommended practice for metadata designers.
These two practices AND all communication around them are essential for the success of metadata model, as metadata – even cleverly designed – cannot work “by itself”.
Having said all this as a sort of a proof that I am not an alien to metadata design :-), I should express a mild scepticism over any substantial metadata design effort. What is below is both an expression of my dissatisfaction and a suggestion of what we could do differently, including in RDA.
Here it goes. I can’t help but think that research information/data managers, as well as researchers themselves could be better off with having a relatively small “core” metadata model with all the tweaks (that are required indeed to fit different cases of metadata use) done through annotations. An annotation can be as an example about data quality or data provenance, so there is no need to include the whole sections for data quality or data provenance or even placeholders for them in a common metadata model – as the mature recommendations for them may well already exist (e.g. W3C PROV bunch of recommendations for provenance).
A core metadata model with an “interface” for supplying/linking annotations may be enough. The annotation itself can “explain” what it is about (provenance, quality, …) – the task of metadata designer is just make the core model “linkable” to annotations, and the task of data/metadata curator (who BTW can be a machine agent) is to actually discover and link the annotation, or a few, to the core description.
This view on metadata design can be named “opportunistic” and relies on collaborative (but not necessarily strongly coordinated) effort of various participants of the research lifecycle who can produce building blocks for suitable metadata descriptions “on demand” – depending on the actual (and different) use cases for metadata. This view is opposite to the traditional “predesigned” approach that relies more on a sophisticated, possibly multi-level metadata model that helps to discover data, to see its relation to different organizational or intellectual entities, and finally interpret it in a meaningful way, so that different actors involved in the research lifecycle just fill in the placeholders left in the large and universal model designed for them by a dedicated group of “metadata gurus”.
In fact, what I suggest is let us treat each metadata record as a “sentence”. It can be a pretty complex sentence – like an extensive multi-element and multi-level description according to an elaborated metadata model, or it can be a simple one (“DublinCore”-like or “DataCite”-like, or whatever). What is important is the ability (of information/data curators and researchers) to combine granular “sentences” in longer ones in order to express what is needed in a particular use case (or a particular range of use cases) for metadata.
To promote this “sentence-centric” view on metadata, I did some small research and suggested a so-called “Core Research Activity Model” http://purl.org/net/epubs/work/10938342 . I am far from seeing it as “The” model: it just shows a way of alternative thinking about metadata and metadata design. This model is also fresh in a way that it is not obsessed with data which is, in the model’s view, just an artefact of a research activity. If you are intrigued, take a look into this 6-pager which includes even a possible RDF representation – which is of course just a representation and should not be confused with the model itself.
With kind regards,
Vasily Bunakov
STFC Scientific Computing
From: raymond.plante=***@***.***-groups.org [mailto:***@***.***-groups.org] On Behalf Of RayPlante
Sent: 02 March 2016 09:34
To: ***@***.***-groups.org
Cc: Bunakov, Vasily (STFC,RAL,SC)
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
Hey Steve -- thanks for the Google docs version--very helpful.
Hi Brian--nice model for the CSMD. Borrowing some of that detail would useful for the more general model.
________________________________
From: brian.matthews=***@***.***-groups.org <***@***.***-groups.org> on behalf of brianmatthews <***@***.***>
Sent: Tuesday, March 1, 2016 8:33 PM
To: ***@***.***; ***@***.***-groups.org
Cc: ***@***.***
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
What we have done identifies a number of base concepts which are either:
- Part of the experimental context – e.g. people (in various roles) facility, instrument, project, sample (a key entity in our world), measurements or
- Describe “Data Assets” and their management – e.g. data sets, data files, archives
These conceptual entities then can have attributes such as titles, identifiers (universally!), timestamps, formats etc.
The core list in the spreadsheet mixes up the “conceptual elements” and the “attribute elements” – understandably as this is a list.
One of Keith’s diagrams does distinguish these.
When I get a moment I’ll comment on the spreadsheet.
I’ll try to arrange a summary of our draft specification to distribute.
One thing that we have done also is begin to give an “abstract workflow for experimental science”.
I don’t believe that you can clearly define (contextual) metadata without a model of what you are describing.
Thanks
Brian
- Show quoted text -From: steve.richard=***@***.***-groups.org [mailto:***@***.***-groups.org] On Behalf Of srichardUSGIN
Sent: 01 March 2016 23:45
To: Matthews, Brian (STFC,RAL,SC); ***@***.***-groups.org
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
Brian—Excellent! How does your analysis compare with what’s in the spreadsheet?
From: ***@***.*** [mailto:***@***.***]
Sent: Wednesday, March 02, 2016 8:14 AM
To: Steve Richard <***@***.***>; ***@***.***; ***@***.***-groups.org
Cc: ***@***.***
Subject: RE: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
We (that is in STFC) have also been working on similar models in the context of facilities and nano-materials data.
We have just completed a deliverable for the project NFFA-Europe (www.nffa.eu) where we have
defined a core set of concepts and fields for experiments. The document is not quite released as yet –
I’ll send a link as soon as it is, but I’ll be talking about it at the joint Materials session this afternoon.
It’s based on our earlier work on Core Scientific MetaData for facilities data.
http://purl.org/net/csmd
thanks
Brian
From: steve.richard=***@***.***-groups.org [mailto:***@***.***-groups.org] On Behalf Of srichardUSGIN
Sent: 01 March 2016 22:52
To: RayPlante; Metadata IG
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
Ray—thanks for compiling this. I wasn’t at the session yesterday, but looking at the slides I had the same reaction. I’ve taken the liberty of making a google doc spreadsheet from your Excel to make it easier to collect comments and discussion. The resource definition is on the first tab, metadata element definitions on the second tab. At this point if people want, I’d suggest adding a new column for each reviewer. Comments can also be added using the google comments.
the spreadsheet is at
https://docs.google.com/spreadsheets/d/18Wi2GNM10d3e_iDWgDS-w4M4Zeaei1wO...
permissions are set up so anyone with the link can edit.
the doc is also published at
https://docs.google.com/spreadsheets/d/18Wi2GNM10d3e_iDWgDS-w4M4Zeaei1wO...
with that link you can only view, so you can share for read-only access.
steve
From: raymond.plante=***@***.***-groups.org [mailto:***@***.***-groups.org] On Behalf Of RayPlante
Sent: Wednesday, March 02, 2016 6:45 AM
To: Metadata IG <***@***.***-groups.org>
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
Dear MIG,
At yesterday's joint metadata session, I felt that a major barrier to effectively discussing the proposed elements--i.e. evaluating as to whether the elements were "the right set"--was the lack of definitions for concepts they are to represent. In an effort improve the discussion, I offer a strawman set of definitions for each of the terms in slide 9 ("Open Data: Purposing the elements") in a spreadsheet posted at https://rd-alliance.org/metadata-p7-tokyo.html. This was a useful exercise at least for me as I realized that I really didn't understand the intended concept behind a few of the elements.
Another thing that I think is important to clarify is the overall model for the things we are trying to describe with these elements. This is covered in part by block diagram in the slides that were presented, but I suspect more should be spelled out. As context for my posted defiitions, I assumed that these elements are (complex) properties describing a thing I refer to as a Resource; a more precise definition is given within that spreadsheet. I will also note that my own thinking comes from work I'm currently doing as part of the Materials Science Interest Group (and the proposed WG on resource registries for Material Science); a conceptual model is describe here: http://wiki.nationaldataservice.org/ResourceConceptDataModel.
hope this is helpful,
Ray
--
Full post: https://rd-alliance.org/group/metadata-ig/post/slides-and-notes-joint-me...
Manage my subscriptions: https://rd-alliance.org/mailinglist
Stop emails for this post: https://rd-alliance.org/mailinglist/unsubscribe/51502
Hey Vasily,
Very insightful post; +1 on your 2 realizations when approaching metadata design.
I was intrigued by your discussion of the annotation approach (I'll spend some time on the plane trying to digest your paper). I was wondering you might share a few sentences toward "annotations for dummies"--not so much that I'm completely unfamiliar with the term as much as I've encountered several variations on a theme bearing this name. You might comment, for instance, if this is an approach that allows people other than a provider of some resource to make statements about it and have those characterizations be part of the globally accessible description of it. Does it sit firmly in the ontological approach to metadata development? I gather from your discussion of "sentences" (akin to RDF statements), the answer is yes.
In as much as I understand your perspective, the annotation approach sounds like something that can exist and be effective as a complement to some "core" description provided by the resource provider, particularly if, as you suggest, that the core schema is not so expansive.
So I bring my own skepticism to metadata development efforts, despite the fact I'm engaged that very activity for Materials Science. I worry about whether a common metadata schema can simultaneously address the needs of high-level discovery and lower-level analysis. My interest in non-disruptive metadata extensibility is an attempt to address this concern: let's start off small and extend as needed by real use cases.
And though it appeals to my info-nerd side, I also worry about the practicality of extracting information (let alone "knowledge") from sentences (is this like RDF triple statements?) that obscures any structure that a mere scientist could wrap their head around. Does getting anything useful out of a big graph of statements require a) learning a still-obscure query language, b) understanding of a vast or numerous ontologies, or c) a really smart tool (that I've yet to see)?
That said, the sentence approach is probably more flexible to restructuring for addressing both high-level discovery and low-level analysis. (Did that make sense?)
Anyway, I hope you can pardon the post-conference dinner thoughts. FWIW, it helps to understand where we think we're going (and what kind of work it will take to get there).
cheers,
Ray
________________________________
From: ***@***.*** <***@***.***>
Sent: Wednesday, March 2, 2016 6:12 AM
To: Plante, Raymond L. (Fed); ***@***.***-groups.org
Cc: ***@***.***
Subject: RE: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
Brian rightly mentioned the further thinking we have done over CSMD for the metadata design in NFFA project.
One thing we realized doing that work was the need in clearly defined semantics behind metadata elements (the problem highlighted by Ray), so we iteratively discussed and produced a Common Vocabulary – which IMHO should be a recommended practice for every newly created or updated metadata model.
Another thing we realized was the need of having an “operational” perspective when you design metadata and when you actually apply metadata model in practice. Producing “how to use” guidelines should be considered another recommended practice for metadata designers.
These two practices AND all communication around them are essential for the success of metadata model, as metadata – even cleverly designed – cannot work “by itself”.
Having said all this as a sort of a proof that I am not an alien to metadata design :-), I should express a mild scepticism over any substantial metadata design effort. What is below is both an expression of my dissatisfaction and a suggestion of what we could do differently, including in RDA.
Here it goes. I can’t help but think that research information/data managers, as well as researchers themselves could be better off with having a relatively small “core” metadata model with all the tweaks (that are required indeed to fit different cases of metadata use) done through annotations. An annotation can be as an example about data quality or data provenance, so there is no need to include the whole sections for data quality or data provenance or even placeholders for them in a common metadata model – as the mature recommendations for them may well already exist (e.g. W3C PROV bunch of recommendations for provenance).
A core metadata model with an “interface” for supplying/linking annotations may be enough. The annotation itself can “explain” what it is about (provenance, quality, …) – the task of metadata designer is just make the core model “linkable” to annotations, and the task of data/metadata curator (who BTW can be a machine agent) is to actually discover and link the annotation, or a few, to the core description.
This view on metadata design can be named “opportunistic” and relies on collaborative (but not necessarily strongly coordinated) effort of various participants of the research lifecycle who can produce building blocks for suitable metadata descriptions “on demand” – depending on the actual (and different) use cases for metadata. This view is opposite to the traditional “predesigned” approach that relies more on a sophisticated, possibly multi-level metadata model that helps to discover data, to see its relation to different organizational or intellectual entities, and finally interpret it in a meaningful way, so that different actors involved in the research lifecycle just fill in the placeholders left in the large and universal model designed for them by a dedicated group of “metadata gurus”.
In fact, what I suggest is let us treat each metadata record as a “sentence”. It can be a pretty complex sentence – like an extensive multi-element and multi-level description according to an elaborated metadata model, or it can be a simple one (“DublinCore”-like or “DataCite”-like, or whatever). What is important is the ability (of information/data curators and researchers) to combine granular “sentences” in longer ones in order to express what is needed in a particular use case (or a particular range of use cases) for metadata.
To promote this “sentence-centric” view on metadata, I did some small research and suggested a so-called “Core Research Activity Model” http://purl.org/net/epubs/work/10938342 . I am far from seeing it as “The” model: it just shows a way of alternative thinking about metadata and metadata design. This model is also fresh in a way that it is not obsessed with data which is, in the model’s view, just an artefact of a research activity. If you are intrigued, take a look into this 6-pager which includes even a possible RDF representation – which is of course just a representation and should not be confused with the model itself.
With kind regards,
Vasily Bunakov
STFC Scientific Computing
From: raymond.plante=***@***.***-groups.org [mailto:***@***.***-groups.org] On Behalf Of RayPlante
Sent: 02 March 2016 09:34
To: ***@***.***-groups.org
Cc: Bunakov, Vasily (STFC,RAL,SC)
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
Hey Steve -- thanks for the Google docs version--very helpful.
Hi Brian--nice model for the CSMD. Borrowing some of that detail would useful for the more general model.
________________________________
From: brian.matthews=***@***.***-groups.org <***@***.***-groups.org> on behalf of brianmatthews <***@***.***>
Sent: Tuesday, March 1, 2016 8:33 PM
To: ***@***.***; ***@***.***-groups.org
Cc: ***@***.***
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
What we have done identifies a number of base concepts which are either:
- Part of the experimental context – e.g. people (in various roles) facility, instrument, project, sample (a key entity in our world), measurements or
- Describe “Data Assets” and their management – e.g. data sets, data files, archives
These conceptual entities then can have attributes such as titles, identifiers (universally!), timestamps, formats etc.
The core list in the spreadsheet mixes up the “conceptual elements” and the “attribute elements” – understandably as this is a list.
One of Keith’s diagrams does distinguish these.
When I get a moment I’ll comment on the spreadsheet.
I’ll try to arrange a summary of our draft specification to distribute.
One thing that we have done also is begin to give an “abstract workflow for experimental science”.
I don’t believe that you can clearly define (contextual) metadata without a model of what you are describing.
Thanks
Brian
From: steve.richard=***@***.***-groups.org [mailto:***@***.***-groups.org] On Behalf Of srichardUSGIN
Sent: 01 March 2016 23:45
To: Matthews, Brian (STFC,RAL,SC); ***@***.***-groups.org
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
Brian—Excellent! How does your analysis compare with what’s in the spreadsheet?
From: ***@***.*** [mailto:***@***.***]
Sent: Wednesday, March 02, 2016 8:14 AM
To: Steve Richard <***@***.***>; ***@***.***; ***@***.***-groups.org
Cc: ***@***.***
Subject: RE: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
We (that is in STFC) have also been working on similar models in the context of facilities and nano-materials data.
We have just completed a deliverable for the project NFFA-Europe (www.nffa.eu) where we have
defined a core set of concepts and fields for experiments. The document is not quite released as yet –
I’ll send a link as soon as it is, but I’ll be talking about it at the joint Materials session this afternoon.
It’s based on our earlier work on Core Scientific MetaData for facilities data.
http://purl.org/net/csmd
thanks
Brian
- Show quoted text -From: steve.richard=***@***.***-groups.org [mailto:***@***.***-groups.org] On Behalf Of srichardUSGIN
Sent: 01 March 2016 22:52
To: RayPlante; Metadata IG
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
Ray—thanks for compiling this. I wasn’t at the session yesterday, but looking at the slides I had the same reaction. I’ve taken the liberty of making a google doc spreadsheet from your Excel to make it easier to collect comments and discussion. The resource definition is on the first tab, metadata element definitions on the second tab. At this point if people want, I’d suggest adding a new column for each reviewer. Comments can also be added using the google comments.
the spreadsheet is at
https://docs.google.com/spreadsheets/d/18Wi2GNM10d3e_iDWgDS-w4M4Zeaei1wO...
permissions are set up so anyone with the link can edit.
the doc is also published at
https://docs.google.com/spreadsheets/d/18Wi2GNM10d3e_iDWgDS-w4M4Zeaei1wO...
with that link you can only view, so you can share for read-only access.
steve
From: raymond.plante=***@***.***-groups.org [mailto:***@***.***-groups.org] On Behalf Of RayPlante
Sent: Wednesday, March 02, 2016 6:45 AM
To: Metadata IG <***@***.***-groups.org>
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
Dear MIG,
At yesterday's joint metadata session, I felt that a major barrier to effectively discussing the proposed elements--i.e. evaluating as to whether the elements were "the right set"--was the lack of definitions for concepts they are to represent. In an effort improve the discussion, I offer a strawman set of definitions for each of the terms in slide 9 ("Open Data: Purposing the elements") in a spreadsheet posted at https://rd-alliance.org/metadata-p7-tokyo.html. This was a useful exercise at least for me as I realized that I really didn't understand the intended concept behind a few of the elements.
Another thing that I think is important to clarify is the overall model for the things we are trying to describe with these elements. This is covered in part by block diagram in the slides that were presented, but I suspect more should be spelled out. As context for my posted defiitions, I assumed that these elements are (complex) properties describing a thing I refer to as a Resource; a more precise definition is given within that spreadsheet. I will also note that my own thinking comes from work I'm currently doing as part of the Materials Science Interest Group (and the proposed WG on resource registries for Material Science); a conceptual model is describe here: http://wiki.nationaldataservice.org/ResourceConceptDataModel.
hope this is helpful,
Ray
--
Full post: https://rd-alliance.org/group/metadata-ig/post/slides-and-notes-joint-me...
Manage my subscriptions: https://rd-alliance.org/mailinglist
Stop emails for this post: https://rd-alliance.org/mailinglist/unsubscribe/51502
Hey Vasily,
Very insightful post; +1 on your 2 realizations when approaching metadata design.
I was intrigued by your discussion of the annotation approach (I'll spend some time on the plane trying to digest your paper). I was wondering you might share a few sentences toward "annotations for dummies"--not so much that I'm completely unfamiliar with the term as much as I've encountered several variations on a theme bearing this name. You might comment, for instance, if this is an approach that allows people other than a provider of some resource to make statements about it and have those characterizations be part of the globally accessible description of it. Does it sit firmly in the ontological approach to metadata development? I gather from your discussion of "sentences" (akin to RDF statements), the answer is yes.
In as much as I understand your perspective, the annotation approach sounds like something that can exist and be effective as a complement to some "core" description provided by the resource provider, particularly if, as you suggest, that the core schema is not so expansive.
So I bring my own skepticism to metadata development efforts, despite the fact I'm engaged that very activity for Materials Science. I worry about whether a common metadata schema can simultaneously address the needs of high-level discovery and lower-level analysis. My interest in non-disruptive metadata extensibility is an attempt to address this concern: let's start off small and extend as needed by real use cases.
And though it appeals to my info-nerd side, I also worry about the practicality of extracting information (let alone "knowledge") from sentences (is this like RDF triple statements?) that obscures any structure that a mere scientist could wrap their head around. Does getting anything useful out of a big graph of statements require a) learning a still-obscure query language, b) understanding of a vast or numerous ontologies, or c) a really smart tool (that I've yet to see)?
That said, the sentence approach is probably more flexible to restructuring for addressing both high-level discovery and low-level analysis. (Did that make sense?)
Anyway, I hope you can pardon the post-conference dinner thoughts. FWIW, it helps to understand where we think we're going (and what kind of work it will take to get there).
cheers,
Ray
________________________________
From: ***@***.*** <***@***.***>
Sent: Wednesday, March 2, 2016 6:12 AM
To: Plante, Raymond L. (Fed); ***@***.***-groups.org
Cc: ***@***.***
Subject: RE: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
Brian rightly mentioned the further thinking we have done over CSMD for the metadata design in NFFA project.
One thing we realized doing that work was the need in clearly defined semantics behind metadata elements (the problem highlighted by Ray), so we iteratively discussed and produced a Common Vocabulary – which IMHO should be a recommended practice for every newly created or updated metadata model.
Another thing we realized was the need of having an “operational” perspective when you design metadata and when you actually apply metadata model in practice. Producing “how to use” guidelines should be considered another recommended practice for metadata designers.
These two practices AND all communication around them are essential for the success of metadata model, as metadata – even cleverly designed – cannot work “by itself”.
Having said all this as a sort of a proof that I am not an alien to metadata design :-), I should express a mild scepticism over any substantial metadata design effort. What is below is both an expression of my dissatisfaction and a suggestion of what we could do differently, including in RDA.
Here it goes. I can’t help but think that research information/data managers, as well as researchers themselves could be better off with having a relatively small “core” metadata model with all the tweaks (that are required indeed to fit different cases of metadata use) done through annotations. An annotation can be as an example about data quality or data provenance, so there is no need to include the whole sections for data quality or data provenance or even placeholders for them in a common metadata model – as the mature recommendations for them may well already exist (e.g. W3C PROV bunch of recommendations for provenance).
A core metadata model with an “interface” for supplying/linking annotations may be enough. The annotation itself can “explain” what it is about (provenance, quality, …) – the task of metadata designer is just make the core model “linkable” to annotations, and the task of data/metadata curator (who BTW can be a machine agent) is to actually discover and link the annotation, or a few, to the core description.
This view on metadata design can be named “opportunistic” and relies on collaborative (but not necessarily strongly coordinated) effort of various participants of the research lifecycle who can produce building blocks for suitable metadata descriptions “on demand” – depending on the actual (and different) use cases for metadata. This view is opposite to the traditional “predesigned” approach that relies more on a sophisticated, possibly multi-level metadata model that helps to discover data, to see its relation to different organizational or intellectual entities, and finally interpret it in a meaningful way, so that different actors involved in the research lifecycle just fill in the placeholders left in the large and universal model designed for them by a dedicated group of “metadata gurus”.
In fact, what I suggest is let us treat each metadata record as a “sentence”. It can be a pretty complex sentence – like an extensive multi-element and multi-level description according to an elaborated metadata model, or it can be a simple one (“DublinCore”-like or “DataCite”-like, or whatever). What is important is the ability (of information/data curators and researchers) to combine granular “sentences” in longer ones in order to express what is needed in a particular use case (or a particular range of use cases) for metadata.
To promote this “sentence-centric” view on metadata, I did some small research and suggested a so-called “Core Research Activity Model” http://purl.org/net/epubs/work/10938342 . I am far from seeing it as “The” model: it just shows a way of alternative thinking about metadata and metadata design. This model is also fresh in a way that it is not obsessed with data which is, in the model’s view, just an artefact of a research activity. If you are intrigued, take a look into this 6-pager which includes even a possible RDF representation – which is of course just a representation and should not be confused with the model itself.
With kind regards,
Vasily Bunakov
STFC Scientific Computing
From: raymond.plante=***@***.***-groups.org [mailto:***@***.***-groups.org] On Behalf Of RayPlante
Sent: 02 March 2016 09:34
To: ***@***.***-groups.org
Cc: Bunakov, Vasily (STFC,RAL,SC)
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
Hey Steve -- thanks for the Google docs version--very helpful.
Hi Brian--nice model for the CSMD. Borrowing some of that detail would useful for the more general model.
________________________________
From: brian.matthews=***@***.***-groups.org <***@***.***-groups.org> on behalf of brianmatthews <***@***.***>
Sent: Tuesday, March 1, 2016 8:33 PM
To: ***@***.***; ***@***.***-groups.org
Cc: ***@***.***
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
What we have done identifies a number of base concepts which are either:
- Part of the experimental context – e.g. people (in various roles) facility, instrument, project, sample (a key entity in our world), measurements or
- Describe “Data Assets” and their management – e.g. data sets, data files, archives
These conceptual entities then can have attributes such as titles, identifiers (universally!), timestamps, formats etc.
The core list in the spreadsheet mixes up the “conceptual elements” and the “attribute elements” – understandably as this is a list.
One of Keith’s diagrams does distinguish these.
When I get a moment I’ll comment on the spreadsheet.
I’ll try to arrange a summary of our draft specification to distribute.
One thing that we have done also is begin to give an “abstract workflow for experimental science”.
I don’t believe that you can clearly define (contextual) metadata without a model of what you are describing.
Thanks
Brian
From: steve.richard=***@***.***-groups.org [mailto:***@***.***-groups.org] On Behalf Of srichardUSGIN
Sent: 01 March 2016 23:45
To: Matthews, Brian (STFC,RAL,SC); ***@***.***-groups.org
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
Brian—Excellent! How does your analysis compare with what’s in the spreadsheet?
From: ***@***.*** [mailto:***@***.***]
Sent: Wednesday, March 02, 2016 8:14 AM
To: Steve Richard <***@***.***>; ***@***.***; ***@***.***-groups.org
Cc: ***@***.***
Subject: RE: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
We (that is in STFC) have also been working on similar models in the context of facilities and nano-materials data.
We have just completed a deliverable for the project NFFA-Europe (www.nffa.eu) where we have
defined a core set of concepts and fields for experiments. The document is not quite released as yet –
I’ll send a link as soon as it is, but I’ll be talking about it at the joint Materials session this afternoon.
It’s based on our earlier work on Core Scientific MetaData for facilities data.
http://purl.org/net/csmd
thanks
Brian
From: steve.richard=***@***.***-groups.org [mailto:***@***.***-groups.org] On Behalf Of srichardUSGIN
Sent: 01 March 2016 22:52
To: RayPlante; Metadata IG
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
Ray—thanks for compiling this. I wasn’t at the session yesterday, but looking at the slides I had the same reaction. I’ve taken the liberty of making a google doc spreadsheet from your Excel to make it easier to collect comments and discussion. The resource definition is on the first tab, metadata element definitions on the second tab. At this point if people want, I’d suggest adding a new column for each reviewer. Comments can also be added using the google comments.
the spreadsheet is at
https://docs.google.com/spreadsheets/d/18Wi2GNM10d3e_iDWgDS-w4M4Zeaei1wO...
permissions are set up so anyone with the link can edit.
the doc is also published at
https://docs.google.com/spreadsheets/d/18Wi2GNM10d3e_iDWgDS-w4M4Zeaei1wO...
with that link you can only view, so you can share for read-only access.
steve
- Show quoted text -From: raymond.plante=***@***.***-groups.org [mailto:***@***.***-groups.org] On Behalf Of RayPlante
Sent: Wednesday, March 02, 2016 6:45 AM
To: Metadata IG <***@***.***-groups.org>
Subject: Re: [rda-metadata-ig] Slides and Notes from Joint Meeting at Tokyo P7
Dear MIG,
At yesterday's joint metadata session, I felt that a major barrier to effectively discussing the proposed elements--i.e. evaluating as to whether the elements were "the right set"--was the lack of definitions for concepts they are to represent. In an effort improve the discussion, I offer a strawman set of definitions for each of the terms in slide 9 ("Open Data: Purposing the elements") in a spreadsheet posted at https://rd-alliance.org/metadata-p7-tokyo.html. This was a useful exercise at least for me as I realized that I really didn't understand the intended concept behind a few of the elements.
Another thing that I think is important to clarify is the overall model for the things we are trying to describe with these elements. This is covered in part by block diagram in the slides that were presented, but I suspect more should be spelled out. As context for my posted defiitions, I assumed that these elements are (complex) properties describing a thing I refer to as a Resource; a more precise definition is given within that spreadsheet. I will also note that my own thinking comes from work I'm currently doing as part of the Materials Science Interest Group (and the proposed WG on resource registries for Material Science); a conceptual model is describe here: http://wiki.nationaldataservice.org/ResourceConceptDataModel.
hope this is helpful,
Ray
--
Full post: https://rd-alliance.org/group/metadata-ig/post/slides-and-notes-joint-me...
Manage my subscriptions: https://rd-alliance.org/mailinglist
Stop emails for this post: https://rd-alliance.org/mailinglist/unsubscribe/51502