System Architecture diagram / document explaining the context for events

[ultrasaurus]

We would like to write up some info so that folks can arrive at this repo and understand what kind of event system we're talking about.

Starting docs:

• presentation -- slide 23 (below) shows the difference between general messaging, and "Eventing" which is what CloudEvents is focused on. Events are statements of fact.
  

• https://github.com/cncf/wg-serverless/tree/master/proposals/cloudevents

• Cloud-Native Event Mapping (CNEM)

• Common Event Use Cases

• OpenWhisk has a nice illustration of the kind of system we're talking about CloudEvents:
  

Assigned to: @ultrasaurus @yaronha

[duglin]

@ultrasaurus you ok with owning this one?

[duglin]

Perhaps including the current "Reference" section of the spec (where we show what some providers do today).

[ultrasaurus]

@duglin yes, happy to help out here. Also, I think you can assign multiple owners. Yaron volunteered to help, but I don't know his github!

[duglin]

Took a sec but I managed to hunt him down @yaronha :-)

[ultrasaurus]

It turned out to be a little challenging to align all of the concepts and terminology (since some people and platforms use the same word for different, but related things). I would like to address this in stages -- if we can align on the really basic concepts (just the ones that are required for the current draft spec) then it will be easier to cover the next set of concepts.

Here's some draft slides that I would like to review at tomorrow's meeting:
https://docs.google.com/presentation/d/1xippFyX9fXqlTWUh3waZcanVMHF12z0AbZ8EK3fk3ys/edit#slide=id.p

If we can agree on slides 1-5, then I could turn them into a markdown document with images interspersed and make a PR -- or export PDF and check it in -- or whatever format people like.

[duglin]

@ultrasaurus I like what you have so far, and I like that it appears you used a lot of the same text we have in the spec itself. That consistency will help newbies.

I don't think there will be much disagreement, or discussion, around slide 1-4 since those appear to be directly from the spec - so any disagreement on those should probably be addressed via PRs to the spec. Slide 5 (definition of 'action') is new and while I suspect is out of scope for the spec (since we're just defining the shape of events, not the processing model), I do think that talking about actions in a system architecture doc will probably help people understand the complete picture.

w.r.t. format of the output of this work, I'd prefer an md file, once we have a baseline doc, so its easy for people to issue PRs against it. But to get to the point of having a baseline/initial-draft, I think a google doc might be easiest.

[duglin]

@ultrasaurus are you still interested in working on this one?

[duglin]

I'm going to suggest that we close this one since I think our primer doc covers our design/architectural thinking for how to use the spec and why we made the decisions we did. Plus we'll continue to add more text over time.

[deissnerk]

@duglin I agree. By now we are also in a working mode that people create specific PRs, if they feel that something is missing.

[duglin]

Agreed to close on the 6/13 call.