The caching of compose windows was implemented a long time ago as a result of the large amount of xbl taking a long time to start up the compose window. We've had anecdotal evidence of non-cached windows not taking massive amounts time to open up. We should add telemetry hooks to measure non-cached versus cached times. These should primarily be on compose windows that aren't also opening emails (e.g. reply/forward), though getting those stats separately might also be interesting. This should give us a better picture of what is happening in real life, and hence, if we need to keep the cached compose window or not.
Mark did you get anywhere with this one ?
We'll just remove the recycling in bug 777732.
Status: NEW → RESOLVED
Last Resolved: 2 years ago
Resolution: --- → WONTFIX
You need to log in before you can comment on or make changes to this bug.