The MSE spec unconditionally invokes the "If the media data is corrupted" section of the HTML spec. Before readyState has reached HAVE_METADATA, it would be more in line with how e.g. HTTP resources work to invoke "If the media data can be fetched but is found by inspection to be in an unsupported format, or can otherwise not be rendered at all."
At least in Opera, network errors are treated differently depending on readyState, so jumping straight into "If the media data is corrupted" when readyState is HAVE_METADATA would requiring special-casing MSE.
Oops, I mixed things up, resolving as invalid and trying again...
Actually, I am doubly confused, there's a problem with both "decode" and "network". The first comment is correct, except for the typo in the second paragraph -- I mean "decoding errors" and not "network errors."
Like in Bug 18624, the behavior should depend on readyState, for consistency with HTTP and other resources.
Looks good, but there are some missing words in "If [the] HTMLMediaElement.readyState attribute [is] greater than HAVE_NOTHING" and similar phrasing.
(In reply to comment #4)
> Looks good, but there are some missing words in "If [the]
> HTMLMediaElement.readyState attribute [is] greater than HAVE_NOTHING" and
> similar phrasing.
Oops. These fixes will be in the next set of updates.