[Pianod] WebSockets

[Michael R. Hines]

It will be at least a couple of years before websockets is widely
available.

I'd stick with the titanium option right now, as it is the most
deterministic, cross-platform solution at the moment.

- Michael

On Fri, 2013-02-01 at 23:27 -0800, Peter Li wrote:
> Sounds good guys.  I was thinking this weekend I might finally have some 
> time to work on the Titanium client a bit more.  Should I go ahead with 
> that or is this websockets stuff going to shake things up a bit?  Like 
> maybe worth trying PhoneGap using websockets, or just trying it from the 
> mobile browsers?  Would be nice not to proliferate clients more than 
> necessary.
> 
> Thanks,
> P
> 
> 
> On 01/14/2013 07:21 AM, Michael R. Hines wrote:
> > That's exactly right, in fact here's a comprehensive list of WebSocket
> > compatibility:
> >
> > http://caniuse.com/websockets
> >
> > The only thing stopping me from trying it was the lackluster mobile
> > device adoption.
> >
> > There are literally millions of old iOS and android devices out there
> > that will take a couple years to be replaced before they get WebSocket
> > support......
> >
> > On Mon, 2013-01-14 at 00:58 -0500, Perette Barella wrote:
> >> So I've been mulling on the web access thing, and it's pretty clear
> >> WebSocket support is the Right Way to implements web clients going
> >> forward, with socket.io a sort of workaround to improvise until
> >> browser support is ubiquitous.  Chrome and Firefox have support,
> >> Safari has some old broken rough spec version.
> >>
> >> So with Chrome or Firefox, using the tool
> >> here: http://www.websocket.org/echo.html
> >> Give it this location: ws://perette.homeunix.org:4446/pianod
> >>
> >> The code is rough and needs refinement before release, and FTR it's
> >> not actually part of pianod-- it's a websockets gateway I've built
> >> with the intent of solving this for everything at once.  It'll be at
> >> least a few more days before release.
> >>
> >> Putting the WebSocket in text mode, you just send the command (the
> >> gateway adds the CRLF) and it returns each line as a separate packet,
> >> CR/LF/CRLF (whichever it finds) stripped.
> >>
> >> Hopefully the rest of the browsers will catch up to RFC6455 soon so
> >> it'll be cross-browser.
> >>
> >> Perette
> _______________________________________________
> Pianod mailing list
> Pianod at lists.deviousfish.com
> http://lists.deviousfish.com/listinfo.cgi/pianod-deviousfish.com