Closed jmccrossan-fox closed 4 years ago
@jmccrossan-fox Thank you for raising this issue!
To date, we have not made stylistic recommendations in TTML2; nor have we singled out treatment of content from specific languages. Rather, we have defined features that are useful across a range of languages usage. Finally, suggesting the use of just tts:shear
is problematic, since it doesn't apply to span
, which is what tts:fontStyle
applies to. So overall, I am opposed to this suggestion, both in general and in its particulars.
I should add that it is the responsibility of the content author to select fonts that correspond to the styles they employ, and that, if they want to specify an "italic" form of a CJK font, then they are free to do so. They can always embed the font directly or reference it as an external resource using the font
element.
@cconcolato Any input on italic vs oblique vs shear?
I would not object to moving this issue to IMSC, which does have precedent for making author recommendations. Also, as a point of note, CAP2TT [1] maps inline shear to oblique style when generating imsc11 output from Lambda CAP (a popular Japanese subtitle format) since imsc11 does not support tts:fontShear
.
Also, as a point of note, CAP2TT [1] maps inline shear to oblique style when generating imsc11 output from Lambda CAP (a popular Japanese subtitle format) since imsc11 does not support tts:fontShear.
Presumably tts:shear
could be used if "oblique" is applied to all text within a paragraph, including ruby text.
@palemieux yes, in Lambda CAP, one can apply 斜め (i.e., obliquing) with a specified angle to either (1) a "screen" (= paragraph in TTML) or (2) inline text (= span in TTML); CAP2TT translates the former to tts:shear
and the latter to tts:style='oblique'
when generating imsc11 output; for ttml2 output, it translates both to tts:fontShear
;
I'm fine with adding the note to IMSC.
@palemieux would you be ok moving this issue to IMSC?
Do we have consensus to move it? If so, then one of the editors can do the mechanics of the move.
@skynavga I think we have consensus, but I haven't tested that yet. Will add to agenda for this week, and we can action it after then if need be.
Since it is unclear looking at the thread, I'm noting that this issue arrived in this repository after being transferred from ttml2, with TTWG consensus.
Per CSS
Many Japanese fonts do not have an "italic" version.
Consider adding an informative note in the definition of tts:fontStyle highlighting the above and recommending that, to avoid indeterministic results, the content author specify tts:shear on Japanese text instead of tts:fontStyle="italic".