FC: Agfa DMCA debate: Charles Platt vs. Matt Deatherage

[Declan McCullagh <declan () well com>]

Both Charles and Matt are longtime Politechnicals (and also are both 
technical folks, both Mac users, and both writers). Read on for their exchange.

Previous Politech messages:
http://www.politechbot.com/cgi-bin/politech.cgi?name=agfa

-Declan

---

Date: Wed, 8 May 2002 18:02:20 -0400 (EDT)
From: Charles Platt <cp () panix com>
To: Declan McCullagh <declan () well com>
Cc: politech () politechbot com, <tom7 () cs cmu edu>,
        <embeddinginfo () agfamonotype com>, <mattd () macjournals com>,
        <cp () panix com>
Subject: Re: FC: Matt Deatherage on Agfa, DMCA, and font embedding problems

I used to design fonts and am interested in this debate. I think Matt is
being at least as "disingenuous" as Agfa. Matt could easily rewrite his
software to satisfy Agfa while filling the needs that he claims his
software was intended to fill. Agfa, on the other hand, seems to be
worrying about an issue that is no longer relevant.

Matt's argument seems to be that "many" (how many, really?) font designers
want to give their fonts away, and were surprised to find that Distiller 4
won't allow this, because Distiller 4 won't embed fonts that have an
invalid permission bit. Is this correct? If so, his legal recourse is very
easy: Rewrite his software so that it will reset ONLY a permission bit
that has an invalid value, or will allow the user to set a more
restrictive permission (but not a less restrictive permission) on a bit
that has a valid value.

This would satisfy the need that he claims his software was written to
fulfill, and should also be acceptable to Agfa.

But what about the type designer who wants (for some reason) to downgrade
a valid but restrictive permission on his own type, to make it more
embeddable? I infer from Matt's argument that such a designer would have
deliberately set the permission to one of the four valid values
previously--and therefore, I infer that such a designer already has
software that sets bits, and can reset bits. Therefore the designer would
have no need for Matt's software.

So much for the legal solution to the problem. From a practical
perspective, software doesn't die. Therefore, presumably at this point
everyone who wants Matt's software already has it, or can get it from a
friend; and therefore he would lose nothing but his "principled stand" by
withdrawing his program. Actually I tend to suspect that the "principled
stand" may be rooted partly in ego and the stubbornness of an individual
who doesn't like to have a large corporation telling him what to do. I can
understand this, but it doesn't form the basis for an ethical or legal
argument.

>From a broader copyright perspective, this situation is interesting
because it foreshadows the possible eventual dilemma of text in general.
Through a quirk in history, fonts have not been subject to US copyright
law. Perhaps coincidentally, some type houses went out of business. Will
this be the fate of text publishers if text, in effect, loses copyright
protection online simply because the protection is so hard to enforce?

Maybe; but on the other hand, Adobe is still busily creating new
typefaces, and is a type house that entered the field around the same time
that fonts became more widely pirated than ever before. Apparently they
find font production worthwhile because their business clients pay, while
individuals tend to steal the fonts without a care in the world--or, more
likely, they get "1,000 fonts for $9.95" from Office Max. Surely Agfa is
in a similar position to Adobe, in which case it should recognize the
realities of this situation. Its customers are corporate users, not
individuals who can't see the difference between Times Roman and its many
freely available imitations.

I note also that Adobe markets very highly priced graphic arts software
that is serial-numbered but not copy protected. Indeed, I used copies of
Illustrator and Photoshop that were of, um, uncertain ancestry, before I
bit the bullet and paid about $1000 to Adobe for my current, registered,
legal copies. I know many designers who went through a similar sequence.
Some (including me) believe in fact that the only way to make individuals
buy expensive software is by allowing them, with a nod and a wink, to use
it illegally for a while. If the software is really, really good, many
people eventually buy it. The ones that don't probably can't afford it
anyway--but will continue to promote it by sharing their illegal copies
with friends. (None of this is relevant to businesses, which almost always
purchase the legal right to use software on multiple workstations.)

In this sense, for individuals, Photoshop is a form of implicit shareware.
Adobe certainly doesn't seem interested in prosecuting the various naive
folks who post messages to the Photoshop usenet news group asking for, or
offering, serial numbers.

Agfa might consider the pointlessness of copyright in this kind of
situation.

--Charles Platt

---

Date: Wed,  8 May 2002 17:53:09 -0500
From: Matt Deatherage <mattd () macjournals com>
Subject: Re: FC: Matt Deatherage on Agfa, DMCA, and font embedding problems
To: Charles Platt <cp () panix com>
cc: Declan McCullagh <declan () well com>, politech () politechbot com,
        tom7 () cs cmu edu, embeddinginfo () agfamonotype com, cp () panix com

On 5/8/02 at 6:02 PM, Charles Platt <cp () panix com> wrote:

> I used to design fonts and am interested in this debate. I think Matt
> is being at least as "disingenuous" as Agfa. Matt could easily rewrite
> his software to satisfy Agfa

In the "credit where it's not due" department, I haven't written any
software.  "Tom 7," copied on your letter, wrote the software in question
and got the letters from Agfa.

> Matt's argument seems to be that "many" (how many, really?) font
> designers want to give their fonts away, and were surprised to find
> that Distiller 4 won't allow this, because Distiller 4 won't embed
> fonts that have an invalid permission bit. Is this correct? If so, his
> legal recourse is very easy: Rewrite his software so that it will reset
> ONLY a permission bit that has an invalid value, or will allow the user
> to set a more restrictive permission (but not a less restrictive
> permission) on a bit that has a valid value.

Allowing a font to be embedded for viewing or printing (but not editing)
is a common choice for font designers.  This is hardly "giving the fonts
away."

> But what about the type designer who wants (for some reason) to
> downgrade a valid but restrictive permission on his own type, to make
> it more embeddable? I infer from Matt's argument that such a designer
> would have deliberately set the permission to one of the four valid
> values previously--and therefore, I infer that such a designer already
> has software that sets bits, and can reset bits. Therefore the designer
> would have no need for Matt's software.

By Agfa's argument, no software anywhere should allow this, because it
would be "circumventing" the embedding bits.  Agfa would only allow
regenerating the font from its source files, but DMCA does not so limit
artists.  If anything, DMCA _protects_ artists.

And the entire point is that Fontographer, an extremely popular font
design tool, does not allow setting the bits properly.  The argument then
boils down to allowing a tool to reset Fontographer's bogus bits but that
does not work on any other font.  Again, DMCA does not limit font
designers to such choices.

I agree that a program that only reset invalid bits wouldn't trigger any
DMCA complaints, but I also say that font designers have the right to set
those bits any way they choose, with any tool they choose.  Agfa does not
have the right to keep competing font designers from using tools of their
choosing (or authorship) because other people _might_ use them to clear
bits that are _not_ a DMCA protection.

Granted, it's often easier to regenerate copyrighted files with accurate
information.  It is not _always_ easier, as in this case.  DMCA may outlaw
programs designed to circumvent copy protection, but embedding bits are
not copy protection, font designers have the right to set them however
they want.  Tom 7 could rewrite his tool to make it less onerous to Agfa
if he wanted, but he has the right to set his bits how he wants using
whatever tool he wants.  He designed the fonts.

--Matt

Matt Deatherage                                  <mattd () macjournals com>
GCSF, Incorporated                          <http://www.macjournals.com>

---

Date: Wed, 8 May 2002 20:48:50 -0400 (EDT)
From: Charles Platt <cp () panix com>
To: Matt Deatherage <mattd () macjournals com>
Cc: Declan McCullagh <declan () well com>, <politech () politechbot com>,
        <tom7 () cs cmu edu>, <embeddinginfo () agfamonotype com>, <cp () panix com>
Subject: Re: FC: Matt Deatherage on Agfa, DMCA, and font embedding problems

Sorry I missed the earlier reference to Tom 7 as the person at the center
of this discussion. You write:

> I agree that a program that only reset invalid bits wouldn't trigger any
> DMCA complaints, but I also say that font designers have the right to
> set those bits any way they choose, with any tool they choose.  Agfa
> does not have the right to keep competing font designers from using
> tools of their choosing (or authorship) because other people _might_ use
> them to clear bits that are _not_ a DMCA protection.

Well, this is the core issue, isn't it? And I would say that the DMCA
leaves it open to debate, which presumably is why we are having the
debate. Personally I think the DMCA is a hopeless attempt to control the
uncontrollable, and the problem is exacerbated because the encoding of
fonts was never designed to be secure. I think Agfa should give up and
follow the Adobe model: Sell to companies that buy their fonts, and forget
Main Street USA (or Font Designers USA, for that matter). The battle has
already been lost on that front. This is the pragmatic argument as opposed
to a principled argument.

On the other hand, the Agfa letter was unusually eloquent as a corporate
statement, and roused some sympathy in me. I don't think the situation is
quite as simple as you suggest.

How would I feel, as a former font designer, if I were still trying to
make money in that field, and a piece of software made it easier for other
people to reset my permission bit so that my fonts could be not only
viewed but effortlessly, automatically copied? Not so good. Surely you can
share this point of view. If my creative work was protected by just a
couple of bits, and someone made it possible to reset those bits--that's
not going to make me happy. It will certainly discourage me from designing
new fonts.

I feel that anyone who writes software that performs this kind of task
should weigh these issues rather than just shrug them off.

--CP

---




-------------------------------------------------------------------------
POLITECH -- Declan McCullagh's politics and technology mailing list
You may redistribute this message freely if you include this notice.
To subscribe to Politech: http://www.politechbot.com/info/subscribe.html
This message is archived at http://www.politechbot.com/
Declan McCullagh's photographs are at http://www.mccullagh.org/
-------------------------------------------------------------------------
Sign this pro-therapeutic cloning petition: http://www.franklinsociety.org
-------------------------------------------------------------------------