Re: [css3-fonts] font-variant-* Minor Comments I

fantasai wrote:

> Are there other reasons the UA can vary the default, or
> are they only allowed
>    a) on by default
>    b) off by default
>    c) depends on text script

It varies depending upon whatever heuristics the user agent
cares to use.  This boils down to the implementation details
and often the application or not will be determined based on
whether "fast path" rendering can be used or not.  That's not
a great design for pages in general but it's what user agents
do and why we have 'auto' in the first place. ;)

I tweaked the wording a bit to discuss this.

>    # Authors should use it in cases where subscripts or
>    # superscripts will only contain the narrow range of
>    # characters supported by the fonts specified.
> Given that we are synthesizing characters not in that range, why are
> we recommending this? Isn't it better to synthesize (if necessary)
> via font-variant-position than to have the author manually
> synthesize it via 'vertical-align' and sizing?

No, I don't think we should encourage authors to use this property for
magic fallback behavior like this.  Whether rendering of the synthetic
variants is "better" or not will depend on the data in the font, it's
not guaranteed to be better, although with a well-designed font it
probably will be.  I think it's better to encourage caution in
author's minds.

>    # As a last resort, unscaled uppercase letter glyphs
>    # in a normal font may replace glyphs in a small-caps
>    # font so that the text appears in all uppercase letters.
> What has to fail for us to get to this point? I'm missing
> that context.

Er, dunno.  This is ancient and honorable wording from the ye olde
days of CSS.  Looks like it was updated at some point in CSS 2.1 to
something equally silly:

Delete?  Guessing this may have something to do with bitmap fonts,
which in some cases can't be scaled?  *shrug*

>    # swsh <feature-index>, cswh <feature-index>
> Are both enabled simultaneously? One or the other depending on some
> particular situation? It's unclear to me what this means.

Fonts generally use one or the other of these features, so yes, both
features are enabled to avoid having to define two separate
properties. There probably only should have been a single feature
registered for this, the fact that there are two is just a legacy


John Daggett

Received on Friday, 31 May 2013 04:35:42 UTC