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

Re: .webfont Proposal

From: Thomas Lord <lord@emf.net>
Date: Thu, 09 Jul 2009 11:28:57 -0700
To: Erik van Blokland <erik@letterror.com>
Cc: Tal Leming <tal@typesupply.com>, www-font <www-font@w3.org>
Message-Id: <1247164137.6773.37.camel@dell-desktop.example.com>
On Thu, 2009-07-09 at 09:26 +0200, Erik van Blokland wrote:
> On Jul 9, 2009, at 12:08 AM, Thomas Lord wrote:

> > Suppose that all of that is fine with you
> > and Erik and the font vendors.  You can separate
> > representation from logical structure and have a more binary
> > friendly format such as MIME.  You can work with,
> > perhaps extending, the ccREL vocabulary.

> I'll have a look, thanks.


Thanks.  Just as an anecdote I should mention
that I came across ccREL *after* I thought of the
basic idea of my wrapper proposal.  It was quite an
eye opener of the form: "Ah.  I see.  Those guys
are actually several *years* ahead of me.  Well,
I can pretend to console myself with the old 'great
minds think alike' notion." :-)


> Two concerns though.

> At first glance it seems ccREL offers licensing info on elements that  
> end up on a page, which implies some sort of UI behaviour. A font  
> would not have a straightforward location for such data, as it is  
> everywhere and nowhere.


The notion of my wrapper proposal is to create a
"container" format.   There's one media file, the 
container.  In it is a regular font file in OTF,
perhaps EOT-lite, or whatever.   Also in the same
container is a *page about the font*.    On that page
there can be RDFa content which conveys licensing 
information in a machine-processable form.  Does
that make sense or am I being too terse?



> Working with ccREL would mean trying to convince the Creative Commons  
> about adding support for non-CC proprietary stuff?

You can use *and extend* ccREL without needing
any permission or agreement from the Creative Commons
folks.   Where they have already defined abstract
vocabulary that fits the bill - use what they already
defined.  Where something new is needed, declare a new
XML namespace and define the new vocabulary you need.

Now, to be sure, there's probably no harm in talking
with the Creative Commons guys and saying to them 
"your vocabulary is incomplete for our needs, please
consider additions like ______" but nothing, as far as I
can see, hinges on how they respond to such advice.


>  While these are all  
> sympathetic folks, it would just be another uphill battle. 

Again: the technical structure of ccREL and RDFa is such
that you need absolutely 0 buy-in or cooperation from
the Creative Commons guys.   You can ignore them or engage
them as see fit.   Should you engage them, it's all upside
insofar as if they say "no" or "go away" or whatever then
just ignore them.    I would guess they won't feel any
urgency to say a lot more than "thanks for letting us know
what you're doing;  we'll consider it".  But, again, it
doesn't matter: you do not need to get any kind of agreement
from them at all.   You don't even, formally speaking, 
need to talk with them.


> I think we  
> can take ideas from proposals like ccREL, it certainly helps to be  
> able to point to precedence.

More than just precedence.

You can see from the ccREL spec and the RDFa 
tutorial that a key element of those is encouraging
browser makers to add new features to make this meta-data
apparent to users in a friendly way.

You, I think, ought to want to leverage that.  "Us too,
We want that as well!"  Then browser makers have 
double the incentive to satisfy both "you guys" and
the "CC guys".

It's really a virtue of what the CC guys have done
so far that you have a chance here to buy into their
work without being obliged to cooperate with them - 
just the kind of thing a good standards process is 
supposed to yield.



> >  I would suggest that
> > the meta-data should be HTML.

> In which context would this be visible?


As an example, my opinion is that when I look
at a page and it has linked resources (images, 
fonts, audio or video files, etc.) that a toolbar-type
thingie in my browser should offer up "about" links
for those.   

Additionally, suppose my browser has (in addition
to a "save this page" option) a "save the font used
here" option.   I look at a page, it has a nice font,
I click "save this font".....

I think a reasonable Recommendation would say that 
the UA SHOULD, if "save this font" is chosen, display
a summary of the ccREL data extracted from the RDFa 
markup of the HTML page, with an option to view the
entire HTML page.    In other words, I click "save font"
and a dialog box pops up that says:

    Copyright (c) xxxx...........
    License information:  [some link]
    Prohibited: copying, distribution, derivative works
    [button: view full meta-data for this font]

    [options: Proceed to save font?  [yes] [cancel]]

-t
Received on Thursday, 9 July 2009 18:29:37 GMT

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