Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Adding terms from ESIP harmonization cryo activity #190

Merged
merged 8 commits into from Jun 8, 2020

Conversation

charlesvardeman
Copy link
Collaborator

Adding back progress from ESIP harmonization cluster efforts on creating definitions for the cryosphere terms. Crossrefs to from SWEET to ENVO terms added during the harmonization captured using skos:closeMatch and a prov:wasDerived from in the definition. Additional provenance added by rdfs:seeAlso to GitHub issue where the discussion was documented. Additional discussion is needed to create a graph fragment for a prov:Activity in some namespace that corresponds to the ESIP harmonization cluster effort to capture the provenance of additions of definitions to SWEET. There were no axiom changes to realmCryo.ttl, but axiom changes were needed for phenCryo.ttl to reconcile the definition with existing SWEET axioms. Periglacial was changed to be a subclass of PhysicalProcess since it is not a Glacial Process. Periglaciation needs more discussion since the term does not occur in any community of practice references. It was changed to match Periglaciation axiomatization, ENVO created it as a synonym of Periglacial.

@rduerr rduerr requested a review from lewismc April 28, 2020 20:50
lewismc
lewismc previously approved these changes Apr 29, 2020
Copy link
Member

@lewismc lewismc left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

+1, I've reviewed the additions for syntax and sampled many IRI's.
I like the way you have structured the skos:definition and will back back and reflect this in my Scala logic.

@lewismc lewismc added this to the 3.5.0 milestone Apr 29, 2020
@lewismc lewismc added Cryosphere Issues related to cryospheric terminology enhancement phenomena (macroscale) labels Apr 29, 2020
Copy link
Collaborator

@smrgeoinfo smrgeoinfo left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Various problems with definitions that are not definitions, are not clear, or don't seem correct.

src/phenCryo.ttl Show resolved Hide resolved
src/phenCryo.ttl Show resolved Hide resolved
src/phenCryo.ttl Show resolved Hide resolved
src/phenCryo.ttl Show resolved Hide resolved
src/phenCryo.ttl Show resolved Hide resolved
src/realmCryo.ttl Show resolved Hide resolved
src/realmCryo.ttl Show resolved Hide resolved
src/realmCryo.ttl Show resolved Hide resolved
src/realmCryo.ttl Show resolved Hide resolved
src/realmCryo.ttl Show resolved Hide resolved
@lewismc
Copy link
Member

lewismc commented Apr 30, 2020

Hi @smrgeoinfo thank you so much for the review this is exactly the kind of thing we need. My review was based on syntax and IRI resolution... not particularly on the quality of the annotations.

@charlesvardeman
Copy link
Collaborator Author

So, these are the definitions the @rdurr had collected and @pbuttigieg entered into ENVO from which they have been extracted (hence the prov:wasDerivedFrom triple). Would @smrgeoinfo be able to join the next Semantic Harmonization Cluster Hackathon on May 13th to help resolve some of these issues? Alternatively, I can remove the definition annotations leaving the SKOS crosswalk to ENVO and let the definitions live in ENVO?

@smrgeoinfo
Copy link
Collaborator

smrgeoinfo commented Apr 30, 2020

@rduerr -- are you actually OK with the definitions I flagged? As I went through the review, you'll note there are many definitions I thought were useful (as in no comment), and I have to confess that when I read those I thought "this sounds like Ruth..."

@smrgeoinfo
Copy link
Collaborator

Yes, I'd be interested in the hackathon on May 13, please send details; I have some experience with glaciers (two field seasons in Antarctica). To me the big problem is construction of logically coherent Aristotilean definitions. My experience over the years is that its difficult to get geoscientists to logically analyze their assertions in a way that's useful for automated computation. Humans are so much better at listening to what someone means, as opposed to what they say, that they find it hard to really consider the actual words they use. The actual words are all the computer system has to work with....

@rduerr
Copy link
Contributor

rduerr commented Apr 30, 2020 via email

Changing Englacial to EnglacialProcess to be explicit

Co-authored-by: rduerr <ruth.duerr3@gmail.com>
@charlesvardeman charlesvardeman dismissed stale reviews from lewismc via f773476 May 1, 2020 18:54
Change Englacial to EnglacialProcess and added deprecated
@charlesvardeman
Copy link
Collaborator Author

@smrgeoinfo here is a link to the ESIP wiki for the Harmonization Cluster. The meetings are on the ESIP Calendar

@rduerr
Copy link
Contributor

rduerr commented May 2, 2020 via email

Copy link
Collaborator

@smrgeoinfo smrgeoinfo left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Guess it depends on how you define geology.

One way or the other, the AGI glossary is a useful reference for Earth Science definitions and includes terms like glacial and glaciation with useful definitions, commonly more that one, reflecting the polysemous use of words in natural language and the science literature. If SWEET is intended to be an 'ontology' useful for precise semantic annotation, then each class in SWEET must map to one definition/usage, and there should be separate classes for the different meanings, with glosses on the refs:label to distinguish them for people. Opaque identifiers are an obvious benefit in this situation.... If that's not the intention, save trouble and point people to an existing glossary or glossaries for definitions, and call SWEET a glossary, not an ontology.

src/phenCryo.ttl Show resolved Hide resolved
src/phenCryo.ttl Show resolved Hide resolved
src/phenCryo.ttl Show resolved Hide resolved
src/phenCryo.ttl Show resolved Hide resolved
@rduerr
Copy link
Contributor

rduerr commented May 3, 2020

OK, actually now that you have the full citation, I can say that the AGI glossary is one of the glossaries in the list I was using. Too bad most cryosphere folks have never heard of it (it isn't in the compilation of terms for WMO's Global Cryosphere Watch). I guess that is just the way different disciplines work! Before any of these are resolved, we need to decide whether:

  1. Correcting all of the SWEET problems both hierarchical and axiomatic at once is really a requirement (since that would take years to release any updates),
  2. Correcting all of the problems in a single SWEET file is required before an update can be made (which also would take a long time to get to the first pull request that passed),
  3. Commenting on terms and axioms changes that are not addressed by a particular pull request is OK (i.e., its OK to review things that are not in scope of a pull request, even when there are other issues to address these);

list-serv update
@lewismc
Copy link
Member

lewismc commented May 13, 2020

Hi @rduerr I kinda feel like the three issue that you bring up are important... but I would rather take those discussions outside of this pull request.
This pull request is quite large as it is... and it turns out that some of the definitions needs to be updated inline with some of the peer review from @smrgeoinfo.

Can someone please outline what the blockers are for this pull request? I'm not able to parse that out from above. Thank you

pbuttigieg added a commit to EnvironmentOntology/envo that referenced this pull request May 13, 2020
@rduerr
Copy link
Contributor

rduerr commented May 13, 2020

Hi Lewis,

I agree with your assessment above and actually that is the conclusion from today's meeting. And we even have proposed answers to my three questions from above:

  1. Correcting all of the SWEET problems both hierarchical and axiomatic at once is really a requirement (since that would take years to release any updates),

We agreed that semantic harmonization is NOT correcting all of the SWEET problems, just adding definitions and pointers to the relevant ENVO terms.

  1. Correcting all of the problems in a single SWEET file is required before an update can be made (which also would take a long time to get to the first pull request that passed),

We agreed that this was also NOT the mission of the Harmonization work. These two decisions mean that @smrgeoinfo will be writing a series of issues about the non-Harmonization issues he brought up.

  1. Commenting on terms and axioms changes that are not addressed by a particular pull request is OK (i.e., its OK to review things that are not in scope of a pull request, even when there are other issues to address these);

We didn't explicitly agree on this, but given the other two and the request to have @smrgeoinfo turn his comments into issues kind of says that we agreed that a review should only be commenting on changes made in the request and that other issues observed should be documented as issues.

That said there are a few outstanding items:

  1. We didn't actually get to review all of the comments from @smrgeoinfo today (there are a few left) but plan to do that at the meeting next Wednesday.
  2. There are come procedural/policy questions that need to be resolved and documented in the SWEET how-to's. The one I remember off the top of my head is how to deprecate a term; also how to update a term (given that the current term is not representative of its subclasses, etc.).

Any pointer's to existing relevant policy/procedures would be helpful here or if no one objects to the decisions made and content of the request, those could form the core of additional editorial guidance.

Copy link
Collaborator

@smrgeoinfo smrgeoinfo left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I went through my comments, most are related to SWEET definitions, and don't directly affect harmonization with ENVO. Problems with the imported ENVO definitions have been fixed. Several issues with labels and definitions added in Esipfed/SWEET issue tracker. Broader problem with logical incoherence and lack of definitions in SWEET too big to solve in one PR!

Copy link
Contributor

@rduerr rduerr left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Looks good to me!

Copy link
Member

@lewismc lewismc left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

+1

@lewismc lewismc merged commit 025c2f7 into ESIPFed:master Jun 8, 2020
@lewismc
Copy link
Member

lewismc commented Jun 8, 2020

Excellent work folks.

@lewismc lewismc linked an issue Jul 17, 2020 that may be closed by this pull request
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Cryosphere Issues related to cryospheric terminology enhancement phenomena (macroscale)
Projects
None yet
Development

Successfully merging this pull request may close these issues.

Ice field definition
7 participants