Skip to end of metadata
Go to start of metadata

You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 9 Next »

Date

Attendees

 

Dennis Wisnosky

 

Decisions

No meeting July 4th 2015

July 27th is preliminary draft spec to OMG

Proceedings 20150629 FIBO.docx

Meeting Presentation – FIBO Securities Content Team Meeting 20150629.pptx

 

20150629 FIBO-FBC FCT

 

Gareth and others have posted revisions to FBC since June 1 meeting . Also conversations with Steve Creek have been productive.  Main coverage gap is examples, that is equivalent, in the EU, of concepts that have been defined from a specifically US perspective.  Elisa still working on aligning regulators to regulations.  Everything we see today will be in Pink GitHub later today.

 

What's the rationale for us doing KYC?  This is in the context of counterparty identification.  FBC is starting to include concepts that are required for KYC compliance alongside BE. These include registration authorities and the like.

 

Elisa has added a new ontology for US­specific jurisdiction regulatory agencies.  Needs an equivalent new ontology for EU.  Discuss whether these are part of the standard, or example individuals. Elisa recommends they are part of the standard.  Goal is to publish for Cambridge, MA OMG meeting.  Additions to Foundations: classifications schemes; Quantities (new module), with new name.  FND Accounting Module has new ontologies.

 

Recommendation from AB to move language, countries and currencies instants, to a separate specification (however, we recommend to keep Currencies in FIBO).

 

Discovered some missing things in regulatory agencies, e.g. the notion of a license (noun).  This is distinct from a Charter.  Being added in FBC not Foundations.  Characterized as a kind of Agreement (against MB advice in recent correspondence).

 

Action:  MB to revisit his recommendations and suggest a better parent ­ by default  this would be moved to Legal document, which sounds fine.

 

Regulation added over the weekend ­ defined in law, has an identifier, is "issued" by some regulatory agency.

There is also a class called "Regulation" in the legacy FIBO.  These are the same concept ­ we now have more restrictions defining membership of this class.  Also needed for Affiliate definitions, as well as Regs, 144­A and so on in reference data.

 

Is this Financial Regulation only?  No. The above is ALL regulations, as it should be, and we will add a child class called Financial Regulation.  Added the notion of Legislature, as a kind of Legal Entity.  Has defining characteristics, e.g. what it represents, capacities and so on. This corresponds to the individuals we got from Steve Creek, which were things like Acts of Congress.  See email correspondence on branches of government.  See also Reg Z.  We would now be able to more fully characterize each of the kinds of regulation we are interested in.

 

Looking at the model in Protege.  Tests include nearly 2000 individuals including language and country codes, and test individuals.  See inferences drawn about Licenses, based on them being Agreements. Allows for identification of "counterparties" presumably as Licensees, licensors.  Will have instances of regulation by the end of today. Not there now.

 

Functional business entities are included e.g. Bank, Financial Service Provider. There are instances of these in the test data.

 

The Bank of England has some kinds of classification of kinds of financial intermediary.  Some of this is on the Wiki already, for various UK jurisdictions.  Suggestion from David Newman that if we are not able to provide axioms for a class of financial service provider, that these should be removed from the ontology.  Note that we have the means of distinguishing many of these by way of describing the kinds of services they provide.  Needs a high level list of services to do this. Steve has provided some of these, Specific to Citi. There are more in the wiki.

 

Differentiating these classes of service provider therefore depends on adding restrictions that refer to individuals, where these are individual Acts and so on.

 

Action: Need a list of definitive, high level services we can use to differentiate these kinds of entities and who regulates them.  So e.g. if someone is regulated by a given regulation, then it means they are a bank holding company.  Need international and EU equivalents to some of these EU examples.  Need to understand the nature of the regulator that is regulating them, not just the name.  EU is more complex, as everyone creates their own set of concepts and then "passport" these across to the equivalents in their jurisdictions within the EU.

 

Mike suggests (a) that we should not remove things that don't have axioms, since there will be cases where we have distinctions that are well understood but hard to model. However that's not really an issue

for kinds of "Services" since services are a well understood concepts.  Mike notes (b) that reasoning over specific regulations, is a means to detect, not a means to define, a given kind of thing. Can we not define what the characteristics are, e.g. in terms of service and also in terms of who the services is provided to.

 

Elisa is doing this bottom­up, by identifying what kinds of individuals (as individual Acts), as a means to eventually identify the high level kinds of service etc. that institutions provide.  BIAN provides a full list of kinds f service that financial institutions provide.  Elisa feels we are not ready to yet model all of the kinds of service that exist in financial services.  The end goal is still to provide definitions of kinds of thing based on some defining service.  The present approach of listing individual regulations is a means to that end.  This will be the final piece that FBC needs.  Elisa and Steve are working together to identify what kinds of services there are.

 

Slide 13:  Now we need to look at the Counterparty Exposure use case and see if we have everything we need for that now. Counterparty exposure is the use case. Counterparty identification is a requirement along the way to that.

 

Regulated Actions: BoE has a new application form in which every organization intending to become a Financial Institution, ha to identify the services they intend to provide.

 

The OMG Spec question regarding Country codes, currency codes, language codes instances.  An OMG AB member suggested that the language currency and country codes were so broadly applicable across other concerns, that they should be managed by the Analysis and Design TF as another spec.  There was discussion on the AB side, whereby most AB members agreed with this proposal. Elisa pointed out this has been in FIBO for some time. Asked whether we are being asked to create a separate spec.  The answer was that if we propose this t the AB, they may tell us to move it to a separate spec. Some AB members offered to assist in the work for this.  With later conversation with Dennis, given that language and country codes are in TC37 they could be considered outside the FIBO remit, whereas Currency is managed by TC68, which is the same space as FIBO.

 

Elisa proposes to take the AB recommendation so as not to delay FBC by another meeting cycle.  Will be working with Unisys, Fujitsu and perhaps others, who have offered to carry out the work that this change would require.

 

Given our dependency on those things, do we run the risk of the AB saying that our submission is dependent on another spec that has not yet been through the process?  Elisa asked them this at the time, and they said they would help us to avoid that pitfall.  The main risk is capacity to get this done.  But, not a procedural roadblock.  We will be unhappy if these TC37 related specs do not work their way through the process in good time.  The language representation ontology and the country (codes) representation ontology that Elisa has developed, is proposed to move to the other AB driven spec. These classes did already exist in legacy FIBO.  The proposed new ontologies will not use the abstractions that are provided in FIBO  e.g. as kinds of code, kinds of scheme and so on, as Information Constructs.  From Elisa's perspective, the ability to disambiguate between concepts e.g. information constructs, is superfluous.  Foundations will continue to look after the higher level disambiguation between otherwise similar looking concepts, outside of what FBC needs to develop in the current iteration.

Action items

  • July 27th is preliminary draft spec to OMG
    Error rendering macro 'jira' : Unable to locate Jira server for this macro. It may be due to Application Link configuration.
  • MB to revisit his recommendations and suggest a better parent ­ by default  this would be moved to Legal document, which sounds fine.
  • Need a list of definitive, high level services we can use to differentiate these kinds of entities and who regulates them.  So e.g. if someone is regulated by a given regulation, then it means they are a bank holding company.
  • No labels