Meetings/Telecon2014.10.21

Event details

Agenda

Resolutions

* Minutes of 29th September approved

Minutes

Chair
Martin Pain
Scribe
Steve Speicher
Attendees
Martin Pain (IBM), Steve Speicher (IBM), Tim Friessinger (IBM)
Regrets
None

Original Chat transcript from room: oslc-automation

Topic: review of minutes from 9/23

https://wiki.oasis-open.org/oslc-automation/Meetings/Telecon2014.09.23

No objections, minutes approved

Topic: Automation 3.0 Plans

Steve Speicher (IBM): Martin: reviewing how to make things easier to read and therefore easier to implement

Steve Speicher (IBM): Steve: suggest that we organize/motivate spec to help minimize the onramp to implement to support some key scenarios, like making query optional or arranging the spec to make it easier to read

Steve Speicher (IBM): Martin: reviewing how to break down the spec into smaller documents, consumable parts

Steve Speicher (IBM): ...options around creating requests, basic/core auto capability, unattended automation, discovery of plans/actions, ...

Steve Speicher (IBM): Steve: be good if a subset of parts was needed to support scenarios, so you don't have to read more than you need (easier to read then).

Steve Speicher (IBM): Martin: the basic part would have to be done for all impls but we could try to accomplish this as we go along, and see how it looks

Steve Speicher (IBM): Tim: what is the difference between basic spec and how to create requests spec, wouldn't you need to know how to create requests in basic?

Steve Speicher (IBM): Martin: the basic will cover a "basic" case but the additional spec will show all the various permutations

Steve Speicher (IBM): Tim: it would be good to have a single simple prefered method to create requests in basic spec

Steve Speicher (IBM): Martin: probably 2 ways, one attended (dialog) or another unattended (POST to creation factory)

Martin Pain (IBM): Martin: As long as we can keep it concise

Martin Pain (IBM): Martin: Could put non-normative description in basic document and refer out to other document for normative document, if we can't keep normative description concise

Steve Speicher (IBM): Martin: looking for editors/authors for specs

Steve Speicher (IBM): Steve: perhaps after LDP and core 3.0 stuff moves along

Topic: Integration Patterns

Steve Speicher (IBM): See https://wiki.oasis-open.org/oslc-automation/IntegrationPatterns

Steve Speicher (IBM): Steve: How does a consumer know when a provider will execute the request or up to the consumer/agent?

Steve Speicher (IBM): Martin: it is out of the spec, impl dependent

Steve Speicher (IBM): Related work at open-services.net (although for a higher-level definition of "Integration patterns"): http://open-services.net/wiki/communications/IntegrationPatterns/

Steve Speicher (IBM): Steve: gave a summary of the "to be done" higher level integration patterns work through the Comm UG, where one might be a "Automation a DevOps Pipeline" pattern, which could point off to these

Steve Speicher (IBM): ...think the current Sequences sound a bit like variations on scenarios, think we should reference and align with the higher level IP effort

Steve Speicher (IBM): Steve: will draft up high-level auto integration pattern to contribute to common efforts

Topic: Systems Engineering Update

No progress on Systems Engineering work.

Topic: High Availability Update

Steve Speicher (IBM): Tim: currently updating the draft to meet template and respec

Other business

Steve Speicher (IBM): Martin: Sam Padgett sent a proposal for optimization to the comments list linking the results to the request

Steve Speicher (IBM): ...using a Link relation header

Meeting adjourned until 2 weeks' time.

Membership changes

At the end of this meeting:

Next meeting:

(See OASIS TC Process - Membership and participation - Voting member)

Meetings/Telecon2014.10.21 (last edited 2014-10-21 16:02:16 by martinpain)