On Fri, Sep 05, 2008 at 11:28:17AM -0400, C. Scott Ananian wrote: > general antagonistic tenor of the thread
Sorry if I caused/contributed any part. > * Quantitative measurements. "Switching performance is down to x ms > rather than x ms". Perhaps rl's tests could be run? I don't know of good ways to measure the actualy performance, since in some noddy tests I last did I would see map and expose events in code far enough ahead of the actual rendering to dissuade me from believing they actually allowed useful measurement of the user experience. > * Thorough testing. It is possible to launch an activity and close it > without a screenshot ever being taken. What happens in that case? I did that test with erikg's first patch, and a "no preview available" thumbnail is shown. My testing was not exhaustive but gave me plenty of confidence that at least the obvious/normal case was gravefully handled. > * More moderate solutions. We're currently taking a lot of > screenshots. Can we quantify what the benefits of taking "fewer" > screenshots without making this into an all-or-nothing discussion? I think erikg's latest patch is pretty moderate - it takes screenshots when the user saves the state. If someone can come up with an even more moderate solution, of course, that'd be great. > [other suggestions: > [throttling the screenshot rate] > [improving screenshot-taking performance] > [API for activities to request a screenshot be taken] The disadvantages of the other suggestions - though certainly more moderate - include 1) a bit to a lot more complex; and 2) no patch exists to implement/use them. That being said, the API for activities to take a screenshot of themselves already exists[1], and when combined with your activity updater it'd be pretty trivial for an activity version to be released to take a screenshot of itself more frequently. > UI-centric approach It is a shame that this discussion has taken place when Eben isn't around, as I'm sure it could benefit. As lots of the Sugar guys are busy as well, I'm sure some more opinions will come around soon... > --scott Martin 1. http://dev.laptop.org/git?p=sugar-toolkit;a=blob;f=src/sugar/activity/activity.py;h=73eeea7499dcd4dcd28cea5c55648b8f414654e9;hb=HEAD
pgpzkNcFfOjAT.pgp
Description: PGP signature
_______________________________________________ Sugar mailing list [email protected] http://lists.laptop.org/listinfo/sugar
