/
2018-03-13 Meeting notes

2018-03-13 Meeting notes

Owned by Dennis Wisnosky
Last updated: May 25, 2020 by Elisa Kendall

Date

Attendees



Agenda

1) Use Case reminder

2) Where we are on our road map. 

3) Open Action Items

4) JIRA Issues Review - https://jira.edmcouncil.org/projects/DER/issues/DER-10?filter=allopenissues

5) Todays content discussion.

SMIF OWL-UML

SKOS

RDF/S

6) For next week.

Proceedings:

Elisa shows many many resolved DER issues in the WIKI.  DA what about LEI capable entities.  EK Done.

DA Saw discrepancy from various sources. EK all removed, that was 2 years ago. 

DN what is now visible in Legal Entities? Remove some classes and restrictions still not done.  legally capablePerson still in.  What to replace it with.  EK Need more context.  DN Surprised that limitedliabilityCompany not included. DA Maybe the details are in a Git branch. DA will look. DA can't fins Resolution S. DN Right now dead in the water because WF had blocked Github access for a while. DN SourceTree not working. DA to DN, it is your branch. DER Resolution S in Davids fork. DN can't use SourceTree. EK needs to look to see what is done and not done. Some maybe OBE's. EK needs to look in GitHub.

DA recalls that some of the WF lawyers hedged much on what exists and not exists. DN to EK R U looking in Release?  YES, much of it is. DN R U removing legally incorporatedPartnership?  EK Yes. Some things were removed and came back. DA came back because they appeared elsewhere in MBs model.  EK CCM duplicates classes twice.  This causes problems. DN wants to eliminate Affiliates.  Will try to get to this before March deliverable, but may not. DA is this in DEV.  DA, Yes. EK crept in when Values was changed. Could be a property that went from a string value to a class. EK wants to kill anything associated with branding. DN R these, and others like them, from MB early work. EK Yes. EK to figure out if useful, need to do forensics across multiple ontologies.  If nothing is referenced, it adds zero value and should be eliminated. DN There abstractions that might have some use someday, but have no value now.  Therefore can be wiped out. EK can't make changes in FND unless MB says so.

DN Have we done a universal about file. DA Yes, we have both ta about and a catalog file. DA if a person is not contributing to FIBo, just using, it is a zip file downloadable.  If it is a FIBO Developer, then the these need to be generated as new.  EK When things are being killed, like now then need to regenerate every time. EK had this problem with Joe Turso.  Needed to regenerate.  DA in the current pub there has been both about and catalog files since Q3 of last year. It is clear how to do this in the Developer's guide. 

ACTION DWIZ: Take the broken image out of the Developers Guide.  

https://wiki.edmcouncil.org/display/INFRA/FIBO+Developer%27s+Guide Look for "Local Testing tools"

DA Need to add a 6th step to Local Testing tools. Make this part of the Git Hook is better.  Make the Git Hook run all of the 6 steps automatically.

EK to DN Is it ok if I just kill stuff?  YES. When we come back to this, we will certainly have other ideas. EK Will be working ton this till the end of this week. Do as much as possible. Raise One issue and then do it.  DN Yes, I trust your judgement.

DN When will DA have this done. DA can do by hand now and then put into Git.  Will do this now.  

Pete Taking the GLEIF dump and putting it in FIBO.  Took the Mexico LOU.  Takes 30 Minutes for the L1 and 1 min for the L2 and then load into Protege. Need to make scalable and operational.  The LEI Level 1 dump is the LEI records. The LE raw data then becomes a FIBO version. Working out details with EK. Some compromises are adding an ontology prefix to guarantee a unique URI for each LEI. Not the same as the EK approach because the EK approach is not unique. Pete takes the LE name and adds LE to that. Append LE to the LEI from a trade. Also take the country information and adding LEs to these.  L2 stuff is the relationship information published by Gleif. Added info so that L2 can be linked to L1.  Four relationship types. Ignoring Period info. FIBO needs more work to represent ownership info.  Working with EK to do this.  All works in Protege. Can assert direct relationships and inferred relationships. Uses ownership relationships.  DN Impressed. 

DA FIBO does not have direct ownership, how did u do that? Can this compute the ownership tree.  Does the property chain reify the direct ownership.  Pete Yes. DA How does this relate ultimate owner and global ultimate parent? Pete GLEIF has the actual GLEIF data. PR It would be interesting to compare the inferred with what people say is the actual. DA did this for Pitney Bowles and it was the same.

PR Canin fer subsidiary because it is an equivalent class showing parent relationship. Does not need the 5 class pattern. Infers the relationship.  Pete need reification because it makes relationships totally clear.  DA but this confuses other issues and makes work arounds like Pete has done necessary and useful. Don't need the 5 class pattern for ownership. Firmly against the 5 object pattern.  EK FIBO has both and it is not clear.  

Pete to JN What do you think?  Fantastic information!

EK changing Percentage control in BE from a class to a property. Pete taking this as declared from the L2 data. A bug n the LEI is that sometimes says 1.00 not 100. 

EK haspercentageControl is the FIBO BE property. More work to do in BE. May need to add another property. Pete has a number of relationship types already.  Jeff whether shares or control should be able to be calculated.  Happy to work with Pete to show the direction that the GLEIF is going. Actual ownership types will not be done this year. Pete if people are reporting # of shares owned, is the total # of shares reported? Jeff No! Sometime in the future. Depends on the regulatory board. Should triangulate with Open Corporates.  Pete Yes have been.  Open Corporates does not have the LEI information even tho they run Open LEI. Jeff, they do internally, but they have a revenue model that keeps them separate externally. Pete Makes it difficult to work with. Jeff uses a statistical frequency approach that mostly works. 


Decisions:

Action items

  • Dennis Wisnosky  Take the broken image out of the Developers Guide.
  • Dean Allemang Create a JIRA issue, do the coding and initial a Pull Request to do the steps discussed automatically.
  • Dennis Wisnosky Add this to the next DER meeting agenda.
  •