Re: Downloadable fonts and image replacement

The DRM part of the discussion could, and even may, be turning into the 
standard discussion between those that believe everything should be open 
(and preferably free) and those that believe that people should be 
compensated for their work. Clearly, in-expensive is desirable as is paying 
people for their work. That, however, is not what is important in this 
discussion. This discussion should be focused on providing the mechanisms 
for both kinds of people to exist in this world and should not be focused 
on which policy to implement or even which policy is right. CSS should be 
providing mechanisms, not policies.

So enough of the philosophy, more comments are below.

At 02:09 AM 4/26/2006, Laurens Holst wrote:
>David Woolley schreef:
>>That's the effect of putting a TTF font on the web!  Anyone can download
>>it and use it in any web site.
>
>Well, I think that's no different for PDF, what's stopping you from 
>extracting the original fonts from a PDF file (except that there may not 
>be tools available at the moment to do that)?

First, not all PDF files have embedded fonts and, second, those fonts that 
are embedded may be subset. That, of course, only makes it less useful to 
extract the font; it does not make it impossible.

What makes the system work is that most people/users are desirous of being 
honest and following the rules of use provided that they know what those 
rules are. So, it is important to allow the rules to be expressed.

>>>bogus anyway, as for restricted use it doesn=E2=80=99t provide informatio=
>>>n about=20
>>>where it can be used, and the field can easily be changed as well. Also,=20
>>EOT does provide information on where it can be used; effectively the 
>>proposal was to allow unrestricted fonts to be used without this 
>>mechanism - you would still need to use something like EOT for level 2 or 3.
>
>I suppose. But it's easy to circumvent that protection, if you want.
>
>Using a font that you didn't pay for is illegal, DRM or no.

This is not exactly true. The whole point of the DRM bits in the font is to 
identify which uses are allowed to users that did not pay for the font. For 
example, if the font allows rendering to a screen or other media, then this 
can be done by someone who has not paid for the font. But, they may not use 
that same font to make changes to a document even if the font were embedded 
in that document.

>  So I really don't see the need for DRM, it doesn't actually help font 
> authors to find illegal applications of their fonts more easily, and I 
> sincerely doubt whether in a lawsuit they will have *more* right for 
> compensation of unauthorised use of their work when DRM is present, than 
> without.

Again, the environment that we live in is, generally, one in which people 
do follow laws when they are aware of the laws. (And, yes I know that you 
can point to many examples where people break laws that they do not agree 
with.)
Therefore, a role of DRM information is to indicate what the legal uses of 
a given font are (and what uses are, therefore, illegal and to be avoided). 
The DRM bits do not, as you say, make it easier to enforce; they make it 
easier for users to voluntarily comply with the rules for the use of the font.

>It's just that I don't see a point in all kinds of protection measures 
>when they can be broken easily by someone intent on using them illegally. 
>For the purpose of making things more complicated, I'd say having them 
>referenced in CSS files should be sufficient to scare away the layman 
>user. For the purpose of someone who really wants to use them illegally, 
>he is breaking the law by stealing, and the legal system is solid enough 
>to take care of that.

The DRM bits are not "protection measures"; they are usage information. It 
tis the User Agents that have an obligation to implement correct usage so 
that the user of the UA does not have to concern himself with the rules and 
can count on his UA to do the right thing for him.

>So that's why I think all kinds of semi-protections like using restricting 
>site URLs and distribution levels is really useless and only making things 
>needlessly difficult.

I agree that whatever is done with fonts should be as easy to use as is 
possible. But, the UA can and should hide most of the complexity of the 
process.


>What would be useful is making a clear note in the specification text that 
>most fonts (even the ones delivered with the user's OS) have copyrights on 
>them and may not be redistributed without permission. That should be 
>sufficient warning.

Most users of files styled by CSS do not read the CSS specification so I do 
not think that the specification is the right place in which to make such a 
statement.


>WRT claiming ignorance, so I suppose I can also claim ignorance about the 
>copyright laws when I copy that game from my neighbour? I don't think that 
>such an argument would hold. Violations of copyright by non-professionals 
>is not where the money is anyway (contrary to e.g. distributing music, 
>where there are some valid arguments for using DRM).

Assuming I agree with you assertion that it is not the non-professionals 
that would be buying fonts (I think this is what you meant and I have no 
knowledge for or against this proposition) then I think that you are 
arguing that we should have DRM information so that the professional users 
can do the right thing.


>In any case, I really think this is something that the legal system should 
>take care of and not the user agents, whose goal is to show the fonts, not 
>protect the font authors.

I think that you are taking a too narrow view of the problem. There are 
multiple goals:

1. we want to enable well styled text (using whatever fonts are legally 
available)

2. we want to encourage font vendors to make as many of their fonts 
available so that they can be used to produce well styled text.

Providing DRM capabilities helps meet the second part of the goals.

Having the CSS specification require that UA's implement detection of and 
correct respect for the DRM information is a proper and correct function of 
the specification. It helps the user be putting the emphasis on his tools 
doing the work for him. It helps the font vendor feel more comfortable 
about making his fonts available for use in CSS styling. It does not seem 
to hurt anyone (except perhaps the programmer that has to write the code in 
the UA). Why is this so big an issue.


         Steve
=====================================
Steve Zilles
115 Lansberry Court,
Los Gatos, CA 95032-4710
steve@zilles.org