- From: François Daoust via GitHub <sysbot+gh@w3.org>
- Date: Wed, 26 Aug 2015 13:09:30 +0000
- To: public-webtiming@w3.org
The following commits were just pushed by tidoust to https://github.com/webtiming/timingobject: * Adjusted the prose on the TimingStateVector constructor The text suggested that the "update" method takes a TimingStateVector as parameter, which is slightly incorrect, since the method rather takes an object that follows the TimingStateVectorUpdate dictionary. by François Daoust https://github.com/webtiming/timingobject/commit/b9d455b23372bbb69b2aec2d7d5f78fc666d1144 * Separated the performance clock, the internal clock and the logical clock The updates were made to clarify what is what. The "performance clock" is the one exposed by Performance.now(). I also made HR-TIME a normative dependency (instead of an informative one), since we now require its support. A timing object has an "internal clock" which is either the "performance clock" or a "logical clock" that follows the skew updates reported by the timing resource provider. I also updated the "query()" procedure of timing object which was still referring to the now gone "now()" method. by François Daoust https://github.com/webtiming/timingobject/commit/373d5fbce6bba4dc05a46dc1fbcedc2fb416b65a * Dropped the note on objects running in different threads The use of observable properties has nothing to do with threads, it is merely a way to simply things for the timing provider. by François Daoust https://github.com/webtiming/timingobject/commit/7243da8282da0b75a6504e779f98ff42a5e488b5
Received on Wednesday, 26 August 2015 13:09:35 UTC