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

format changes as false protection

From: Thomas Lord <lord@emf.net>
Date: Wed, 29 Jul 2009 12:50:31 -0700
To: www-font@w3.org
Cc: John Hudson <tiro@tiro.com>, Dirk Pranke <dpranke@chromium.org>, "Tab Atkins Jr." <jackalmage@gmail.com>
Message-Id: <1248897031.5922.71.camel@dell-desktop.example.com>
There is an idea floating around that we
should settle on a new font format in order
to put a (weak) barrier between "the web" and
"the desktop".

I offer the observation that that will almost
certainly fail to work.  Here is why:

Thesis: As web fonts come into popular
use, publicly licensed (libre) fonts will be
the most widely used fonts, *by far*.

Reason: The majority of users will not want
to pay license fees or enter contracts and
they will not want to get in legal trouble.
They will use libre fonts.

Thesis: Downloading fonts from web to desktop
will gain in popularity.

Reason: There being nothing illegal about it,
users will download libre fonts to use in their
word processor documents, the titelbars of their
windows, and so forth.  This form of font sharing
will become relatively popular and beneficial.

Thesis: Support for such downloading will be 
made highly streamlined, if not automatic.

Reason: People will build software to make it
easy to download and install a web font as a system
font.  Such software will be trivial to write.
It will have a perfectly lawful purpose.  As 
users observe others downloading fonts and making
interesting use of them, they'll want to do the
same and the very simple software needed to make it
easy will become common -- like software for 
playing MP3s, for example.


That means that Dirk is right when he says that
a new format with no greater rationale than to
be a new format is an objectionable idea for 
a Recommendation.

I'd observe that that leaves us about where
we are, though:

EOT-lite is not a new format for its own sake,
it is a new format with a potential for retrospective
compatibility.

.webfont is not a new format for its own sake,
it conveys a new form of meta-data which UAs 
SHOULD process and present according to certain
rules.

The mime wrapper proposal shares a rationale
with .webfont but is more careful in describing
what that new meta-data can contain and how it 
should be presented.  It also proposes a format
more in line with web architecture generally
and with greater opportunity for generalization to
other media types.


-t




On Wed, 2009-07-29 at 11:59 -0700, John Hudson wrote:
> Dirk Pranke wrote:
> 
> > Conversely, there are certainly people that will point out that this
> > does not provide any value at all for legitimate users, only hurdles
> > that they have to jump over. I.e., why should browser vendors
> > implement features that make users' lives harder, when even the font
> > makers admit that this provides no real security or restrictions? Why
> > aren't we willing to trust users to do the right thing?
> 
> In my case because I've been in this business for fifteen years and I've 
> seen an awful lot of users do the wrong thing.
> 
> But I disagree that these protections offer no benefit to users. If I am 
> at risk of breaking the law, e.g. by trespassing on private property, a 
> fence or a sign is helpful to me. I say this as someone who lives in a 
> rural area where ancient trails run across what are now subdivided lots, 
> and I like to know if I'm at risk of walking onto someone else's property.
> 
> Also, what are the 'hurdles'? What about a web font format makes users' 
> lives harder? All but a few existing TTF and OTF fonts are not licensed 
> for use on the web. The hurdle that users -- users in this case being 
> web designers/authors, not web readers -- are going to need to jump over 
> is obtaining a new license that permit upload to web servers. [Whether 
> that license is something that costs extra to the regular font license 
> is a business decision for the individual foundries.] If the new license 
> comes with a new version of the font that is packaged for web serving, 
> is compressed ready for web serving and hence does not require users to 
> involve themselves in server-side compression for font data, and 
> contains serialised license data making absolutely clear that this font 
> is licensed for web use, all the user has to do is upload the font and 
> put the appropriate code in his or her CSS. How is this more onerous 
> than obtaining a new license to upload an uncompressed naked TTF or OTF 
> font and linking to that in CSS?
> 
> Also, as I've stated numerous times, font maker/owner and user are not 
> exclusive categories. In the case of companies with custom typefaces, 
> they are both the owners of those fonts with an interest in protecting 
> them and also the users who want to make use of these custom types in 
> their websites.
> 
> If an interoperable web font format is taken up by all the browsers, I 
> suspect the makers and providers of free fonts will be quick to make 
> downloads available in that format. [I have academic clients who make 
> the custom fonts I created for them freely available[1] for 
> non-commercial use, and I would provide them with new fonts in a web 
> font format to make available for free download.] Yes, there will be a 
> file format distinction between the desktop installable version of a 
> free font and the version served on the web, but since those files 
> reside in different places why shouldn't there be a format distinction? 
> Again, there doesn't seem to me anything onerous about this.
> 
> John Hudson
> 
> 
> [1] http://www.sbl-site.org/educational/biblicalfonts.aspx
> 
Received on Wednesday, 29 July 2009 19:51:14 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Saturday, 11 June 2011 00:14:03 GMT