What exception should speechSynthesis.speak() throw for reused SpeechSynthesis?

[foolip]

https://w3c.github.io/speech-api/webspeechapi.html#dfn-ttsspeak says "The SpeechSynthesis object takes exclusive ownership of the SpeechSynthesisUtterance object. Passing it as a speak() argument to another SpeechSynthesis object should throw an exception."

It doesn't say what exception to throw. It should.

Spun off from #7 and a TODO in web-platform-tests/wpt#7510

[foolip]

As it stands, this doesn't throw an exception in any browser:
https://wpt.fyi/results/speech-api/SpeechSynthesis-speak-ownership.html

@jdsmith3000 @andrenatal, do you believe this should throw an exception? In Edge and Firefox, is this case handled at all? In Chromium, it looks like there's just no concept of ownership of SpeechSynthesisUtterance objects, if it's possible to pass them to speak() then they'll be added to an internal queue.

[stephenmcgruer]

Note also from web-platform-tests/wpt#23097, it was pointed out:

The language used is should throw an exception, not MUST throw an exception.

So technically not throwing an exception is spec-compliant, but that also sounds like a spec bug. SHOULD is reserved for cases where there is a good reason to disobey the behavior in specific circumstances (and it is usually best to spell out those circumstances).

There are some other good points in that issue; worth a read.

[marcoscaceres]

So technically not throwing an exception is spec-compliant, but that also sounds like a spec bug.

Totally.