Apple Releases Safari Technology Preview 54 With Bug Fixes and Feature Improvements

really, why don't they just fix the thing to get a perfect score or is this another case where they don't agree with the standard and refuse to follow it? smh

Scores are based on what Google want to put in Chrome. Safari therefore loses a lot of points for not supporting Ogg Theora / Opus, WebM VP8 / WebM VP9. The only real user of any of these is YouTube for VP9.

Basically not all points are equal. The missing input types are a problem. Not supporting WebM with Vorbis support isn't.

482 out of 555.

Scores are based on what Google want to put in Chrome. Safari therefore loses a lot of points for not supporting Ogg Theora / Opus, WebM VP8 / WebM VP9. The only real user of any of these is YouTube for VP9.

Basically not all points are equal. The missing input types are a problem. Not supporting WebM with Vorbis support isn't.

I agree with the spirit of what you're saying. But in this particular case, I don't think html5test.com actually awards any points to Ogg Vorbis / Opus / WebM. They are on the list but they're listed as "Optional" when you click on them (with no points assigned).
[doublepost=1524108358][/doublepost]

safari might be ok for home users, but is practically non-existent in pro area. can't imagine e.g. a solution for telco officially supported on safari. probably mainly a US thing. use whatewer u like, but safari will always be a minor browser. i've tried it couple of times but never really liked it. better alternatives exit - practically the same as with apple maps

set date = 2001 ; s/safari/firefox/g

I think you're probably implicitly referring to Google Chrome as being the browser for the "pro area," as it is the marketshare leader. I give you that Google is pushing for web standards, but I don't know that I agree that Chrome is "better" in all senses of the word. I regard Google Chrome as unmitigated spyware: It installs a persistent background app that can in turn download, install, and run code whenever it wants. When you launch Chrome, it makes connections to a wide collection of Google domains even if you have all Google services unchecked or try to set it for maximum privacy. If you block one of these connections it tries another domain. Google may argue that this behavior is necessary to traverse locked-down networks, detect man-in-the-middle, but I just see it as malware-like behavior.

I'd actually use Chrome if it made zero (0) connections to any Google property, ever, unless I explicitly hit a button that read "Check for Updates." And then I wouldn't press that button; I'd periodically download new versions of Chrome from Google itself. Because I don't trust Google not to leak some GUID back to the Googleplex during a simple update call.

Google holds the market lead for now. IE did back in 200X, then lost it because it was complacent with web standards. Google may lose it because of privacy.