...
Attendees
Agenda
1) Use Case reminder
...
SKOS
RDF/S
6) For next week.
Proceedings:
| View file | ||||
|---|---|---|---|---|
|
20170613 FIBO CFTC PoC
Shrini wants to do the Prototype internally before going to SDRs. Can we expose dummy SDR data as SPARQL end points to show to senior management. DN Blockchain was developed just for this purpose. Doing this internally first is prudent.
JN: Blockchian would not give us all of the data. Now, we either download the SDR data or they send it as CSV files. With FIBO, how will we get the data? SPARQL end points.
DA: We have found that SPARQL end points are often politically unacceptable.
JN: If SPARQL end points are not a good idea, what is? DA: If they give us the CSV's we can do the RDF work. Or, they can just put the data somewhere.
JN: We can also say that in the future, the SDRs can publish their data as RDF.
DN: Yes, that should be the future. DN: Shrini says it is too early to talk to SDR's? JN: Yes.
Do internally first to show the value to senior leadership. End game is how are we going to get data to us in the future?
DN: There are many ways that we do this in the future.
JN: Map IRS, e.g. the 90 fields that I sent earlier. All different names, Different labels. This is what is on the Jim Corley data base now. Show the advantage of a single triple store, and QA, QC and inferences. Don't need to do all 90. 40 would be wonderful.
DN: FIBO has many now such as SDR Id's and more. Should not be difficult. We should think of this as a phase with two small steps.
JN: Even if we link to all 90, there is still other data we get. If we want real data, we need to figure out how to get that or someone tells us how. In the meantime, I will have an SoW and run this up the chain to procurement.
DN: Is there precedent for one SDR is more important than another. JN: Yes, but we need to show how we harmonize data from 3 SDR's. Show some type of swap: Float float, fixed floats, etc.
DN: Then when JN is done we can see which fields are actually in FIBO. JN: Yes, I am filling out the descriptions now. But, all SDRs don't respond the same to a single field. Some say Y, some say Yes, Some say True, and each means the same thing.
DN: We can handle this. JN: We are just trying to get to yes from John Rogers and can tackle harder issues later.
DN: Ball is in your court at the moment. Few minor tweaks on our side. And then, we look at your new spreadsheet and see what we can do now, and then later. Bite size chunks.
JN: Sounds good. Have approval that James Corley is now working on this. But, can't make this meeting. Want to change the time. How about Wednesdays 1300-1430 Eastern time?
ACTION: DN and DWiz to check on this time.
Decisions:
Action items
- David Newman Dennis Wisnosky check on Wednesdays 1300-1430 Eastern time for a meeting that can include James.