This makes me very happy, as backwards compat for performance is a huge bit of complexity.
We should have a blanket message or something for when we're attempting to connect to an unsupported platform -- that the platform is unsupported and that they should use Firefox X host. Any ideas here? On Wed, Jan 27, 2016 at 9:11 AM, Dave Camp <[email protected]> wrote: > For the Firefox TVs in market I think you should identify the latest ESR > release that has support those televisions and recommend that your > developers use those. > > -dave > > On Wed, Jan 27, 2016 at 7:07 AM, Shih-Chiang Chien <[email protected]> > wrote: > > > Hi Ryan, > > > > There are still Firefox OS TV (2015/2016) in market that supports WebIDE > > debugging. Glad to hear that you keep the backwards capability in devtool > > architecture. Breaking the existing functionality will make the end user > > really upset and it'll hurt Mozilla's reputation. > > > > Best Regards, > > Shih-Chiang Chien > > Mozilla Taiwan > > > > On Wed, Jan 27, 2016 at 9:29 PM, J. Ryan Stinnett <[email protected]> > > wrote: > > > >> Keep in mind that we still want to support Firefox for Android back to > >> release channel, so we still want to keep traits in general. That's a > >> much smaller time window than we supported for Firefox OS, so it > >> should allow for some simplification. > >> > >> Also, a number of our traits are used by Valence as flags to tell the > >> client "that's not implemented yet", so those are important to keep to > >> avoid breaking this use case. > >> > >> So, I think the type of compatibility that could now be removed are > >> things where the protocol semantics changed in some Gecko older than > >> release channel and the trait is not useful for saying "I don't > >> support feature X". > >> > >> I have updated our backward compatibility page with this information: > >> > >> https://wiki.mozilla.org/DevTools/Backwards_Compatibility > >> > >> - Ryan > >> > >> > >> On Wed, Jan 27, 2016 at 6:38 AM, Joe Walker <[email protected]> > wrote: > >> > > >> > TL;DR: Yes, if that helps. > >> > > >> > Support for any fxos debugees is always possible using the 'right' > >> debugger > >> > version. > >> > Maintaining strict backwards compatibility just made it easier for > >> people > >> > that didn't want to find the right debugger version. > >> > > >> > So I think we should feel free to remove the maintenance burden of > >> > supporting older fxos debugees. That doesn't mean we need to rip it > all > >> out > >> > tomorrow, but we shouldn't put any significant effort into backwards > >> compat > >> > for fxos from now. > >> > > >> > There are likely to be nuances to this which we can discuss, but in > >> > principle the answer your question is "Yes, if that helps". > >> > > >> > Joe. > >> > > >> > On Tue, Jan 26, 2016 at 10:20 PM J. Ryan Stinnett <[email protected]> > >> wrote: > >> >> > >> >> On Mon, Jan 25, 2016 at 2:31 PM, Nick Fitzgerald > >> >> <[email protected]> wrote: > >> >> > Perhaps we can now remove remote debugger protocol compatibility > >> support > >> >> > for legacy fxos debuggees? > >> >> > >> >> I'll bring up this topic at next week's DevTools meeting. > >> >> > >> >> - Ryan > >> >> _______________________________________________ > >> >> dev-developer-tools mailing list > >> >> [email protected] > >> >> https://lists.mozilla.org/listinfo/dev-developer-tools > >> _______________________________________________ > >> dev-fxos mailing list > >> [email protected] > >> https://lists.mozilla.org/listinfo/dev-fxos > >> > > > > > _______________________________________________ > dev-developer-tools mailing list > [email protected] > https://lists.mozilla.org/listinfo/dev-developer-tools >
_______________________________________________ dev-fxos mailing list [email protected] https://lists.mozilla.org/listinfo/dev-fxos
