Philip Jägenstedt

Results 1157 comments of Philip Jägenstedt

Root problem IMHO is that showing that many runs isn't useful in the first place. It comes from the fact that there's a "grace period" so that APIs have to...

Maybe as a safety, also check that the equivalent camelCase thing exists, and otherwise keep it?

I think this is probably caused by the interface object being found in Chrome via `navigator.geolocation.constructor`, and for example `navigator.geolocation.constructor.prototype.clearWatch` can be found via that object. In Firefox, `navigator.geolocation.constructor ==...

https://github.com/GoogleChromeLabs/confluence/issues/200 is related to this, but the usual manifestation of `[NoInterfaceObject]` is that if we can't find an instance we just don't know about its support. For `Geolocation`, an instance...

Oh my, this is because of https://github.com/mdittmer/web-apis/pull/98. I didn't check beyond the API Catalog and API count metric, and indeed *all* the other metrics turn into nonsense with new Edge...

I have pushed a new release that just deleted the new Edge data: https://web-confluence.appspot.com/ I didn't delete it from https://github.com/mdittmer/web-apis however, so I'll leave this issue open as a reminder...

This list of differences between Safari 13 and 14 might be interesting: https://web-confluence.appspot.com/#!/catalog?releases=%5B%22Safari_13.0.2_OSX_10.15%22,%22Safari_14.0_OSX_10.15.6%22%5D&q=%22count:1%22 Because we don't have Safari 13.1 data I don't know how much is new in Safari 14...

I've made a staging release with all Edge results deleted entirely, and the metrics actually seem to work, at a glance at least: https://confluence-b2397ab-dot-web-confluence.uc.r.appspot.com/ Releasing this is tricky though, because...

That looks like the right place, and I tried it in https://github.com/GoogleChromeLabs/confluence/pull/663, but the result isn't the same as when just deleting all of the Edge data. Edge is still...