I guess the issue here though is that there's no way for me to tell what's going on (because the log file couldn't be created). D11351 solves this for me.

Yeah, this is by design, at least in theory. Reopen if you get more evidence that it's not working or whatever?

Oh I see, we attempt to restart it if it dies.... I'm not quite sure whether this was happening (my guess is that it would've died immediately).

Obsoleted by T6559.

I implemented (very simple) reconnects in JX.WebSocket and they seem to work pretty well.

This doesn't seem to happen anymore after T6559: Change notification system implementation not to require Flash.

This doesn't seem to occur any more after T6559: Change notification system implementation not to require Flash.

Yeah, this is me misunderstanding npm (or maybe having an old understanding of it, IIRC -g didn't exist a couple of years ago). I'll get rid of the -g stuff.

(the other link was the 2nd hit on Google, but this is the official documentation: https://docs.npmjs.com/getting-started/installing-npm-packages-globally)

Here's an implementation of a web sockets server in PHP: https://github.com/hach-que/Omni/blob/master/src/pipe/WebSocketEndpoint.php

In T6559#90099, @epriestley wrote:

Overall, I'm satisfied that WebSockets aren't a crazy mess anymore, and I think we can move forward to WebSockets only and drop Flash.

Looks like @epriestley has beaten me.

Diff that very hackily implements desktop notifications on top of standard notifications: https://secure.phabricator.com/differential/diff/27013/

How large an implementation are we looking at to get WebSockets + localStorage synchronization? Is it more like 100 lines, 1K lines, 10K lines, or what?

In T6559#90012, @epriestley wrote:

For example, do websocket connections show up in browser debugging tools?

Flash can also elect a master tab, so we can play sounds (like chat notifications) once, and publish desktop notifications once, pretty easily. We need to be able to do this. Can we reasonably do it with localStorage?

Flash can play sounds pretty reasonably. Can we do that in a nice, cross-browser way without it?

T6559 may obsolete this.

Might be obsoleted by the outcome of T6559.

I also think it would be fine to drop realtime support for IE8 and IE9. Other HTML5 features (like drag-and-drop uploads, I believe) don't work there, and I think it's reasonable to require users to upgrade to get fancy bells and whistles.

Per D11143, there are kind of two separate issues here:

My intention is to keep the actual message structure the same, so using Flash as a fallback should be possible.

I doubt we're likely to add an external just for that. If there is a way to keep flash as fallback (unsupported) while only adding features to the Websocket version, that might be the best course.

We could use SockJS or similar to emulate the behaviour for IE8/IE9?

What about IE8 and IE9?

I've been pondering a WebSockets implementation and am going to start mucking around with a prototype. I don't think this task is too difficult. In fact, WebSockets probably gives us a few things for free that we had otherwose hacked into the Flash implementation of Aphlict.

I could name another reason: interoperability between Firefox and Flash on Windows has been abismal the past couple of years, crashing or hanging plugin-container.exe causing a lot of us to preemptively disable the Flash plugin. Using web sockets (where available) would be a great improvement.

Feel free to rejigger priority. That said, I think if this was "high" priority it should DEFINITELY be a Support Impact thing too.

What prompted me to look for this feature is that I thought there was an issue with my installation that prevented notifications from being highlighted then I thought it was Phabricator issue till I knew it was by design.

Top level I'd have to ask what problem are we solving. What prompted you to come and ask for this feature, maybe something already exists in Phabricator that could solve that? If you feel you're missing information, I'd rather start there.

Oh, I see some discussion in T1403. We can let this sit and see if there's more interest in it, but I think we're unlikely to ever pursue this.

I don't really think this is worth implementing -- we haven't seen any other requests for it, and I don't think GitHub's behavior is very good (I end up with a lot of duplicate unread notifications). Do you think it's worth pursuing, @chad?

Having the favicon also change would be very cool. I usually have my phabricator tab pinned, which means that I don't see the title.

Closed by commit rP568b0e06e7d5.

◀ Merged tasks: T6026.