Skip to content

Conversation

martinthomson
Copy link
Member

There was a lot of text here that was simply wrong, probably because it was outdated.

I've removed or reworked that part. I haven't cited the IETF work (yet), but there is a placeholder there.

Closes #56.

@martinthomson
Copy link
Member Author

Oh yeah, I forgot, this doesn't update the big sequence diagram. I might get to that, but without the original source, I'd have to start over.

This was referenced Mar 18, 2015
@beverloo
Copy link
Member

I agree that we need to make a decision about this, while I'm still erring towards the separated properties side myself. The problem is that no one feels strongly about it - there's arguments both for keeping it and for removing it.

Let's wait for @mvano's return and then find a way to drive this to a solution.

index.html Outdated
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think you lost some meaningful information in this heading. How about something like "Push service use" as what we really care about here is pointing to the IETF Web Push Protocol.

@mvano
Copy link
Contributor

mvano commented Apr 16, 2015

Thanks for the simplifications Martin! Now that the IETF work is progressing and has support from Mozilla, Microsoft, Google, and interest from others as well, this generally looks good to me. If you could address some of my comments, this pull request should get merged ASAP.

index.html Outdated
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I've added the sequence diagram source in vdx format. I made it in Lucidchart, and editing it should work well there.

@martinthomson
Copy link
Member Author

OK, I just had a shot at this. I made some fairly big changes to the diagram, so check those over. I don't know how you want to do comments; maybe I can share it in Lucidchart.

@paul-em
Copy link
Contributor

paul-em commented Apr 16, 2015

I think it is way better to understand.

The only minor things I found: Shouldn't clear push subscription happen after unsubscription, so that you can wait for it to resolve successfully before you delete it?
And I find "set push subscription" not very informative, why not take the old "store push subscription" or "save push subscription", or was there another thought behind?

@martinthomson
Copy link
Member Author

@paul-em, you shouldn't remove a subscription before you tell everyone that is using it to stop using it. Otherwise, they will expect to have messages delivered to you and (probably) get errors.

set vs save vs store matters not to me (other than the time it will now take to regenerate all the files). I thought that set and clear were neatly symmetrical, but didn't really think that hard about it.

@paul-em
Copy link
Contributor

paul-em commented Apr 17, 2015

Oh ok makes sense to remove before then.

@mvano
Copy link
Contributor

mvano commented Apr 17, 2015

Martin, the diagram changes generally look good to me. Just a few nits:

  1. the clear push subscription should have a PushSubscription as an argument, like the set action
  2. If you can crop the image and drop the blank bits on the edges, less screen space will be wasted. We can always add padding in CSS if needed.

@mvano
Copy link
Contributor

mvano commented Apr 17, 2015

Hmm, maybe you can address those nits later, let's just get this merged for now.

mvano added a commit that referenced this pull request Apr 17, 2015
Use just endpoint, drop subscription id, diagram cleanup, refer to draft IETF web push protocol
@mvano mvano merged commit d55040e into w3c:gh-pages Apr 17, 2015
@martinthomson martinthomson deleted the single_endpoint branch August 5, 2015 16:00
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

Remove registrationId
4 participants