W3C home > Mailing lists > Public > public-houdini@w3.org > February 2015

Re: Font Metrics concerns

From: Tab Atkins Jr. <jackalmage@gmail.com>
Date: Sun, 8 Feb 2015 13:55:38 +1100
Message-ID: <CAAWBYDBs107yyjZwzE+W5j14chHL+--vj5u6ZCrwAaqAzApZDg@mail.gmail.com>
To: Alan Stearns <stearns@adobe.com>
Cc: "public-houdini@w3.org" <public-houdini@w3.org>, John Daggett <jdaggett@mozilla.com>, Bram Stein <stein@adobe.com>
On Sun, Feb 8, 2015 at 1:45 PM, Alan Stearns <stearns@adobe.com> wrote:
> So we’ve resolved to create a spec for a font interface that contains font
> metrics information necessary [1] for script to produce interesting
> typographic effects (drop caps, baseline grids, etc.) I have some initial
> thoughts about how this information should be exposed, and some concern
> about how we should rationalize what’s currently specified for canvas
> TextMetrics [2].
> There are quite a few metrics we could expose. I don’t know that it’s
> necessary to include every possible metric in the first cut of this
> interface. It seems to me that we could start with a reasonable subset and
> add more information as needed. I think a reasonable subset would include:
> alphabetic baseline
> hanging baseline
> ideographic baseline
> ascent
> descent
> x height
> cap height
> Steve warned that there are several different meanings for ‘ascent’ and
> ‘descent’ in different fonts. So I assume this interface should pick one
> and describe how to translate known differences to what the interface
> provides.
> The TextMetrics interface has attributes for some of these metrics in CSS
> pixel units. This makes sense for the canvas measurements being made by
> that interface, but I’m thinking the font interface we’re designing should
> be expressed in more abstract units. One possibility is to add a
> unitsPerEm attribute and express all the other metrics in those
> font-specific units. That’s probably the most natural thing to do for
> typographers. But I think it could also work to rationalize on a single
> unitsPerEm value or use an em percentage.

An em percentage is fine with me.  No need to indirect through some
variable amount of units.

Received on Sunday, 8 February 2015 02:56:25 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 11 February 2015 15:47:05 UTC