w3c / mediacapture-handle

https://w3c.github.io/mediacapture-handle/
Other
14 stars 10 forks source link

Integrate the Actions compromise presented at January WebRTC interim #6

Closed jan-ivar closed 2 years ago

jan-ivar commented 2 years ago

We agreed to proceed with the API presented at the WebRTC interim in January, which adds the important Actions component, which through standardization won’t limit this critical cross-app functionality to web properties owned by the same corporation.

The Actions presented are the template rudimentary “next", “previous", “first", and “last” ones needed to advance slides from a VC (that is: to drive a tab-captured generic web presentation from another generic VC tab), as an addition to the spec's “identification” section (which lets properties partner to provide the same and more through tight integration across specific products).

This is important for scope, and a blocker for CfA.

eladalon1983 commented 2 years ago

It is important to accurately portray agreements. What we agreed to do was propose to the Working Group something along the general lines of Actions, and that I did - I came up with a concrete proposal, I presented it to the Working Group, and I debated its merits at length with @youennf. The debate started here, in the comment from Jan 14. Please contrast my engagement and Youenn's with that of other participants in the debate. I think my commitment Actions has been very high.

Where do we stand now?

It serves nobody to block Identity on something entirely distinct and without consensus.

But you're welcome to send a PR, debate it with Youenn, and get it merged. If the current Capture Handle document is adopted by the W3C, we can follow the official process in debating PRs for adding Actions.

alvestrand commented 2 years ago

As I said in January, I am in favor of having Capture Handle and Actions as two separate documents. In addition to what Elad says - I believe that Actions, if it finds interest with Web developers, has a lot wider applicability than Capture Handle, and should not be tied to it. (I also think that Actions are out of scope for WebRTC, and should be pursued in WICG or WebApps, but we can debate the merits of that argument somewhat separately from the debate about integrating it in the Capture Handle document.)

jan-ivar commented 2 years ago

It is disappointing to see us backing away from compromises reached at the January meeting.

I think it's valid concern that the final shape of actions are not clear, and would be satisfied by including a statement that includes their consideration, since they're a mitigation to concerns raised with the Identity part of this document. This should allow us to keep refining the API shape and discuss whether this should be one document or two. Anything less is not a compromise I fear.

alvestrand commented 2 years ago

Note: Personal position.

I did not agree with the inclusion of the two functionalities in the same document in January, and don't do so now. I do not agree that the concern raised (if I understand it - it's an attempt to mandate support for loosely coupled applications when all the use cases we have been presented so far by Web developers are for the abilitiy to support tightly coupled applications).

I support the development of vocabularies of interaction for loosely coupled applications, such as that proposed in the "actions" idea, but note that such attempts have failed before (WebIntents). Attempting to block the ability to write tightly integrated applications based on the desire to provide functionality that has previously failed makes no sense to me.

eladalon1983 commented 2 years ago

It is disappointing to see us backing away from compromises reached at the January meeting.

I have done no such thing. Clearly, we see things very differently. At any rate, you're welcome to submit PRs to that effect and resume my previous efforts to bring Youenn on board. You'll have my full support in the matter. Good luck!

I think it's valid concern that the final shape of actions are not clear, and would be satisfied by including a statement that includes their consideration

Are you saying that you'd be satisfied if we include a simple clarification, saying that the WebRTC Working Group deems desirable a future API along the lines of Actions? Am I reading you correctly, @jan-ivar? If so, we have a path forward. Please suggest this clarification via PR and let's close this topic.

jan-ivar commented 2 years ago

... it's an attempt to mandate support for loosely coupled applications when all the use cases we have been presented so far by Web developers are for the abilitiy to support tightly coupled applications).

@alvestrand Here's a web developer from the public-webrtc mailing list last June:

"Hi Elad,

I find this API really interesting and I can understand the value for google and other service providers. However, it is unclear what is the benefit for the rest of the community. Let me explain my concerns.

Given that the method are opt-in, I foresee that only the web sites interested in being captured will ever use this API, and given that the the web sites can set the domains that will be allowed to receive that information, it is not unreasonable to think that they will only allow for the same company VC products.

So my worries are that we will end up having an API that will be only enabled in google docs to be able to expose information to google meet, and in microsoft 360 to expose information to microsoft teams, and they will be able to provide a much better presentation experience than the rest of VC services. I am not saying that these are Google or Microsoft intentions, but that is a more than feasible possibility.

I understand the value of an API like that, but I think it should be a benefit for all, not just for the ones that control both the content and the conferencing services. I really hope that the API can be modified so this can happen.

Best regards Sergio"

jan-ivar commented 2 years ago

At any rate, you're welcome to submit PRs to that effect and resume my previous efforts to bring Youenn on board. You'll have my full support in the matter. Good luck!

Thanks, here's the PR you requested https://github.com/WICG/capture-handle/pull/15. PTAL. I've cc'ed @youennf.

My understanding from the January meeting was that in the interest of progress, everyone was willing to adopt and iterate from what was presented at the January meeting.

eladalon1983 commented 2 years ago

Thanks, here's the PR you requested w3c/mediacapture-screen-share#15. PTAL. I've cc'ed @youennf.

Thanks. I will indeed take a look.

My understanding from the January meeting was that in the interest of progress, everyone was willing to adopt and iterate from what was presented at the January meeting.

I do not recall @youennf agreeing to that. Can he either comment here saying as much, or can you point out his agreement elsewhere?

By the way, let's remember what I agreed to. Quoting myself:

There are two core issues addressed in this thread, which I'll call "identity" (original proposal) and "actions" (additional mechanisms). I also prefer splitting the work, but I tentatively propose addressing both issues together as an attempt to reach a compromise with @jan-ivar. To avoid risking misrepresenting his position, I'd like to ask @jan-ivar to explain why he thinks the two should be combined.

This "tentative proposal" was intended to expedite things and help build consensus. It's usually much harder to reach a compromise on multiple topics than on one. The thread with Youenn starting January 13 proves that Actions is not an exception to this rule. I no longer think the trade-off presented by such a proposal is favorable. (Edit: I will reconsider if Youenn throws his support behind a combined proposal. But I think you should reconsider yourself whether we'd be investing our time wisely in meticulously crafting Actions, when none of us has concrete plans to implement it.)

eladalon1983 commented 2 years ago

@jan-ivar, IIANM this is all now integrated, so I'm closing. If I'm mistaken, do please reopen.