2019-10-31 Meeting Notes

Meeting minutes status: Draft (pending approval)

Table of Contents

Date/Time

Thursday, 31 October 2019 @ 10:00 EST / 14:00 BST

Attendees

Name

Organisation

Github ID (optional)

Name

Organisation

Github ID (optional)

Riko Eksteen

Adaptive

rikoe

Mark Valeiras

ChartIQ

 

Leslie Spiro

Glue42

 

James McLeod

FINOS

 

Malcom Chedzoy

JPM

 

Nick Kolba

OpenFin

 

Evgeny Kovalyov

Refinitiv

 

Christi Reid

Adaptive

 

Colin Eberhardt

ScottLogic

 

Time Dinsdale

OpenFin

 

Sanjay Nagpal

Credit Suisse

 

Phil Broadbent

OpenFin

 

Vinod Mehta

FactSet

 

Outstanding Action Items

Agenda

Time

Item

Who

Notes from the Meeting

Time

Item

Who

Notes from the Meeting

5 min

Convene & roll call

Riko Eksteen



5 min

Approve Past Meeting Minutes

Everyone

Approved

5 min

Review action items from previous meetings

Everyone



10 mins

Working group member onboarding:

  • GitHub Teams

  • Confluence Members

@James McLeod (Unlicensed)

Riko will validate with FINOS Confluence team members, and look at creating an FDC3 API working group team on GitHub.

20 mins

Discussion topic: FDC3 entry point

OpenFin’s FDC3 service exposes a window.fdc3 object as entry point to the FDC3 API interface, and others may be doing the same. Should we formalise the way FDC3 is accessed across desktop agents to ensure compatibility.

What is the implication for so-called multi-agent environments (if that is something we should be considering)?

Also see: https://fdc3.finos.org/docs/1.0/api/DesktopAgent

Everyone

Global or module?

Could be an unnecessary complication/is it a real problem?

Riko will create a GitHub issue and we will socialise that way.

We need to collect requirements and decide priority.

20 mins

Discussion topic: Do we need a wider scope, i.e. an FDC3 protocol

This is a catch-all topic for multiple related issues which came up during our previous meeting:

  • APIs define an interface, but does not describe behaviour

  • Should we provide .NET API support (for e.g. WPF interop)

  • Cross-machine and client-server workflows

  • Exposure of FDC3-like operations via REST or other mechanisms (e.g. URL handlers)

  • npm & other package support

Everyone

Is it correct to assume it is a walled garden? This would require explicit bridges between them.

Does it reduce the value of FDC3 if applications operate within a specific implementation?

Agreeing on context is important for transference of data.

The group decided that people are fine with FDC3 operating within the context of a particular desktop agent for now and a wider protocol is not required. Building bridges should be cheap if required.

20 mins

API GitHub Issue Review (see GitHub issue list):

Everyone

Issues will be reviewed at the next meeting

5 min

AOB & adjourn



5 December will be the next meeting

1.1 spec release would be desirable

Decisions Made

  • The next API Working Group meeting will be on 5 December.

  • FDC3 is not trying to be a general protocol for completely open desktop interoperability, participants are fine with the idea that it operates within the context of a particular desktop agent and the communication mechanisms it defines, which are implementation details of that desktop agent.

  • A 1.1 release incorporating new channels API is desirable and will be proposed to the FDC3 PMC.

Action Items

Capture any actions that were identified here, and make sure they are also captured on the group's task page during, or immediately after, the meeting

@Riko Eksteen to send email to FDC3 API mailing list about missing FDC3 API Working Group charter
@Riko Eksteen to validate Confluence members against API WG participants and contact help@finos.org to add them if needed
@Riko Eksteen to create an FDC3 API Working Group team on GitHub
@Riko Eksteen to create a GitHub issue about FDC3 entry point and distribute to mailing list

 

Need help? Email help@finos.org we'll get back to you.

Content on this page is licensed under the CC BY 4.0 license.
Code on this page is licensed under the Apache 2.0 license.