Co-author of the article here.

I agree with dpark - an unstated goal of this proposal was for developers to be able to easily move legacy sites with legacy endpoints to long sessions without needing to update those endpoints or pages.

The service worker approach can effectively intercept requests to any endpoint and perform the re-generation of a short cookie if needed, without needing to change every page on the site or the legacy endpoints.

Co-author of the article here.

An unstated goal (we probably should have stated it, looking back) was for sites to be able to easily move legacy sites with legacy endpoints to long sessions.

The service worker approach can effectively intercept requests to any endpoint and perform the re-generation of a short cookie if needed, without needing to change every page on the site or the legacy endpoints.

One additional benefit is that it minimizes the transmission of the long term token, which is generally good if you're worried about it somehow getting intercepted. You may or may not be too concerned about that risk though.

(Disclaimer: I work on these features on Chrome)

We're exploring that. It's not 100% clear either way, but I personally hope we can do it.

Feel free to star this issue if you want to keep track of progress: https://bugs.chromium.org/p/chromium/issues/detail?id=516147

I built Electron's native Windows notification integration and I'd be suuuuper happy about seeing it become a standard Chromium feature.

According to that thread the reason for not doing Win10 notifications is: It would create a weird state where Chrome behaves differently on Win 10 than on Win 7/8 and developers of extensions/websites wouldn't know what they design for.

Which is pretty weak, IMO. They argue that Win10 adoption is less. But I think the people using Win10 are already more than those using any form of OSX, ever.

Mods - title should read "Native Mac Notifications Behind A Flag in Chrome Stable"

Looks great! Out of interest, does this take advantage of any native features via node or could it also be a website?

I'm curious if the motivation of making a desktop app was just for the "installed app user experience" or because it was needed to access some API.

FWIW Alex Russell wrote a great article about why our current approaches don't work and why this is interesting to explore: https://infrequently.org/2015/08/doing-science-on-the-web/

[Full disclosure: I work on this project]

But it's opt out vs opt in which is a huge difference.

Really? Huge difference? I would say about 20% (being generous with that number) of the good apps that I install try to misuse the notification. I don't find it that hard to disable the notifications for them. The benefits of notifications outweigh the spamming by a lot.

iOS is Opt-In

Sites can definitely work offline using the new Service Worker system: http://www.html5rocks.com/en/tutorials/service-worker/introd.... Try this example: https://voice-memos.appspot.com.

That feature was designed but never built, so his point still stands. In my ideal world sites can do that stuff, they just have to ask, and users can always say no.

So the way iOS apps already do this...