search

w3c / imsc-vnext-reqs

IMSC Next version Requirements
https://w3c.github.io/imsc-vnext-reqs/
5 stars 8 forks source link

Consider adding tts:fontShear #43

Open cconcolato opened 4 years ago

cconcolato commented 4 years ago

IMSC1.1 has support for tts:shear which provides sub-optimal results if some cases. For example, shearing a left-justified, multiline paragraph does not preserve the left-justification. Another example is with regards to ruby. Shearing a paragraph with ruby shifts ruby text with regards to their base character, slightly modifying the alignment with the base character. CSS is defining a new value for its font-style propertie (https://drafts.csswg.org/css-fonts-4/#valdef-font-style-oblique-angle) which could be used to map tts:fontShear, and that is intended to cover all complex cases (vertical text, ruby, tts:textCombine).

palemieux commented 4 years ago

slightly modifying the alignment with the base character.

I thought this was in fact the desired outcome: https://github.com/w3c/ttml2/issues/784#issuecomment-390858956

Has the thinking changed?

cconcolato commented 4 years ago

Has the thinking changed?

On my side, I would slightly modify what I expressed here: https://github.com/w3c/ttml2/issues/784#issuecomment-392140264 For tatechuyoko, no change. The combined set of characters has to be sheared as a whole. I think that is meant to be the case with "font-style: oblique ". For ruby alignment, it depends on the value of rubyAlign and on the length of the ruby text. Some people think keeping the original alignment is better for readability, while some others think it is better to shift the ruby.

nigelmegitt commented 4 years ago

Some people think keeping the original alignment is better for readability, while some others think it is better to shift the ruby.

Can these different people be classified into different constituencies with different concerns? In other words, is there (for example) one view held mainly by print typography folk and another for electronic display folk?

nigelmegitt commented 4 years ago

During yesterday's call the TTWG discussed our approach to future requirements in IMSC including this issue and decided to consider this for the next version of IMSC after IMSC 1.2.