W3C home > Mailing lists > Public > www-font@w3.org > July to September 2009

Re: Questions re web-fonts

From: John Hudson <tiro@tiro.com>
Date: Wed, 15 Jul 2009 09:39:57 -0700
Message-ID: <4A5E065D.1000205@tiro.com>
To: "Tab Atkins Jr." <jackalmage@gmail.com>
CC: www-font@w3.org
Tab Atkins Jr. wrote:

> Please don't dismiss the benefits of a quick solution, though - EOT
> (or a compatible derivative) can be 'adopted' *retroactively* so that
> users with obsolete browsers (in the IE line, which is the majority of
> internet users currently) can benefit.  Any new format, on the other
> hand, even if decided on today, will require at least half a decade
> before it's truly usable.

> When speaking practically, a 'quick' solution like this carries some
> very large benefits with it that can balance it against other
> solutions which may be more technically desirable.

That is all very reasonable, and I agree that retroactive compatibility 
is a significant benefit and one that needs to be weighed as a factor. 
What I don't like is the sense I have of being pushed towards such a 
solution, as if this factor automatically overrides others. I am also 
not convinced by talk of 'at least half a decade' to make a new format 
'truly usable' (and what does 'truly' imply in this context?). It seems 
to me that Opera et al have made naked TTF/OTF linking usable in 
comparatively short order. What will take so long?

There seem to be two time aspects to supporting any format: development 
time to do the coding, testing, etc., and roll-out and upgrade time 
during which one can expect some users to stick with older browser 
versions or to upgrade more slowly. Let's look at both of these.

One of the great things about Tal and Erik -- one of the reasons they 
are so highly respected and valued by their type design colleagues -- is 
that they not only come up with bright ideas but they also build things. 
This means that they tend to have a pretty good idea of the amount of 
work involved in implementing their ideas. So when Erik tells me that 
the code he needs to unpack and get a binary OT font out of his wrapped 
and compressed .webfont is less than one screen of Python, I figure he's 
right that this isn't going to take years to develop.

All the browsers are doing a much better job of convincing users to 
upgrade than used to be the case. Indeed, this is the case for software 
in general, and I think it would be a mistake to estimate roll-out and 
upgrade periods based on past experience. I also suspect that much of 
the user base sitting on older versions for longer represents, on the 
one hand, major site installations with conservative IT policies, and on 
the other hand people like my mother who are only going to upgrade 
software when someone else comes and does it for them. Neither of these 
user types strikes me as a major target for websites with fancy new 
typography: they're people who either want to maintain a known system or 
who are content with what they've got. Since any web font implementation 
is going to include fallbacks to exactly the same level of typographic 
(un)sophistication that these users find satisfactory, I fail to see 
what isn't 'truly usable' about any format that one can get into the 
code and out the door.

John Hudson
Received on Wednesday, 15 July 2009 16:40:43 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Saturday, 11 June 2011 00:14:02 GMT