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

Re: Font license vs. conversion between font formats

From: Christopher Fynn <cfynn@gmx.net>
Date: Thu, 09 Jul 2009 21:10:31 +0600
Message-ID: <4A560867.4020707@gmx.net>
To: www-font <www-font@w3.org>
CC: Erik van Blokland <erik@letterror.com>
Erik van Blokland wrote:

> On 9 jul 2009, at 13:42, Mikko Rantalainen wrote:

>> However, I'm trying to say that you can put simple roadblocks but unless
>> you've real security, you should not be surprised to find that those
>> roadblocks have been circumvented in no time. After that those
>> roadblocks are obstacles only for honest authors, users and browser
>> vendors. They do not prevent even casual copying (because copying tools
>> will be advanced enough).

> Ah, nihilism. I think you'll find most honest authors actually 
> interested in licensing and willing to take the required steps to do the 
> right thing. Given that any font ever released is one torrent search 
> away, foundries still sell fonts. To who? honest authors. These are not 
> people who too stoopid to find the torrent, they actually *want* to buy 
> our fonts and support our work.

Yes and @font-face linking obviously extends the places you can license 
these paying customers to use your fonts.

> Now then, as you concur the technical part of this webfont proposal is 
> so easy it can be implemented as fast as it can be circumvented, I 
> really don't see why you're having a hissy fit about it. The domains 
> part is optional (read the proposal). Wrapping a font should be trivial 
> for you.

> The bottom line is. the folks who prefer to find the torrent, lift the 
> font from the wrapper, or rewrite the domains are not honest authors.

> A raw ttf / otf, sent unasked for to the honest reader, ends up 
> ready-for-action on the reader's desktop. If the font is not accompanied 
> by any declaration of its origin and under which conditions it was 
> licensed to the honest author, what conclusion is the honest reader to 
> reach? The reader didn't agree on any license. The author didn't list 
> any conditions or stipulations. If the reader uses the font, it is not a 
> dishonest step, there is no promise broken, no intent ignored.

> Taking a font out of a wrapper requires a step. However trivial, it is a 
> step. An unlocked door that says "keep out". Honest folks will not 
> enter, because the sign states intent, it requires effort to consider 
> the potential risks and get in anyway.
> Given the sheer number of people who will get these fonts (how many page 
> hits a day, global, all people, all sites, day in day out?), and the 
> stupendous number of honest readers, such a barrier will have a lot of 
> effect.

> If my trust in mankind is so far off center that this scheme would not 
> work, there is hardly any reason to discuss licensing raw fonts either.

 > Erik

The main drawback I see in the near term is that this won't work on any 
browser today; While it could be implemented quickly, realistically I 
don't think it would be supported in IE until at least IE v9 - and, for 
practical use, web authors would need to wait until the majority of IE 
users upgraded to that as yet to be created version; if the proposal is 
accepted, I'd expect support in other browsers might appear somewhat 
sooner than in IE.

Meanwhile, authors who want to use @font-face are going to have to serve 
EOT fonts to IE users and OTF/TTF fonts to the rest - and limit 
themselves to fonts with licenses that permit this.

Once support for this proposed format starts to appear in browsers, you 
then have to serve your fonts up three ways for a while - only when 
support for this new format gets near ubiquitous are you be able to use 
a single web-font format. In other words, there would be a considerable 
period of transition (? years) before this format alone was very useful.

If that is the intention of the exercise - you can only remove support 
for OTF/TTF & EOT in browsers when the majority of websites using 
@font-face with those formats have been updated to use the new. This 
could easily require an additional period of several years.

I'm not quite sure why one couldn't simply add all this meta information 
to the font files themselves and continue to use existing formats ~ Then 
all we have to do is persuade OS vendors or browser vendors to display 
this information as an option whenever a font is installed on a system 
- similar to the way most software install programs display a licence.
Since the GPL suggests that free software displays a licence, I don't 
think anyone would object to this. Anyway even with your proposal you 
still have to get browser vendors to implement some means of displaying 
the meta information, so what's the difference?

Compression can be handled at the server level without requiring support 
for a new or additional font file format or wrapper.

----

Playing devil's advocate:- Since your proposal is so obviously 
extensible why can't someone add root string or DRM extensions to it 
later, then some commercial font licenses require these be used, and at 
least one browser vendor with a significant market share be persuaded to 
support such extensions? Perhaps more importantly, what do you think the 
likelihood of someone trying to make something like this actually happen 
would be?

- Chris
Received on Thursday, 9 July 2009 15:11:20 GMT

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