On Sat, Mar 29, 2014 at 3:05 PM, Chris Peterson <[email protected]>wrote:
> On 3/29/14, 11:38 AM, Till Schneidereit wrote: > >> Just stumbled upon this: >> https://groups.google.com/forum/#!topic/mozilla.dev.gaia/Il3DiRpIZ0U >> >> In this particular case, the right conclusions was arrived at, but I >> wonder >> if we should try to be more proactive about getting people to not use any >> features we want to remove eventually. Or, at least, be free to deoptimize >> as needed. >> > > Till, by "be more proactive", do you just telling people (on the lists you > mentioned) or adding warnings (to jshint or console)? > > At the JS work week, we discussed adding telemetry for JS 1.8 (and other > non-standard <script> type values). Is that still interesting? > We should do all three, I think. I filed bug 988386[1] on the last. Note that that's for web content, though, not platform/Firefox/Gaja code. The second is the most difficult to get right, though: we don't want warning fatigue to completely undermine any effect. We could try collecting info on how much the features we want to eventually remove are being used, respectively. Based on that, we could start adding warnings about the least-used ones. In a way, that's backwards, as those are less of a problem, but it at least dampens fatigue. [1]: https://bugzilla.mozilla.org/show_bug.cgi?id=988386 _______________________________________________ dev-tech-js-engine-internals mailing list [email protected] https://lists.mozilla.org/listinfo/dev-tech-js-engine-internals
