- From: Erik van Blokland <evb@knoware.nl>
- Date: Fri, 9 Aug 96 01:59:14 +0200
- To: "Chris Lilley" <Chris.Lilley@sophia.inria.fr>, "w3" <www-font@w3.org>, "Paul Haeberli" <paul@sgi.com>, "Jonathan Hoefler" <hoefler@typography.com>
Chris Lilley:
>> Actually, the pixelfont document has been extended a bit.
> Last Modified: Thursday, 08-Aug-96 01:13:12 GMT
Erik van Blokland:
[generated on Thu Aug 8 19:13:35 1996]
Perhaps it got stuck in a cache somewhere? Anyway, it's there now:
http://www.letterror.com/pixels.html
>The main substance of the argument seems to be not that
>pixelfonts are
>good, but that outline fonts on the Web are bad.
Well observed! Putting outline fonts, for instance the typefaces that I
make and try to sell (still a legal practice in the western world), into
webpages to make sure the recipient gets to see the page the right way on
a 600 dpi screen also happens to broadcast the font across the world.
The music industry has years of experience with this, backed by hefty
copyright laws. By a strange twist of fate, US copyright law does not
offer the same protection to typefaces, nor is there a mechanism to
ensure remuneration for use of typefaces on the web. Even if such a
mechanism would be started and new legislation would protect typefaces,
it will take years for this to work. In the mean time, schemes to enclose
my fonts (and everybody elses') in every single webpage are beyond beta,
and ready to ship. What am I supposed to say? take them please? take my
fonts for progress? There are alternatives and I want to make sure they
are investigated.
>You seem to take the
>position that outline fonts should only be used by
>phototypesetting
>bureaux, everyone else should use bitmaps to ensure that they
>don't copy outline fonts.
Nope. When fontformats are sufficiently smart to protect their contents,
as for instance java fonts might be able to do, when fonts have become
active bits of code that produce type instead of passive data containers
that rely on the integrity of the operating system it happens to end up
on, only then would it make sense for type manufacturers to allow
embedding of their fonts in webpages. That means a whole lot more needs
to be developed than some 2 bit protection that says "please don't embed
me, mister operating system, sir?" Remember how long the type 1
encryption survived? And Adobe really tried then. The next time around
encrypted time will last half as long.
I agree that pixelfonts won't be the answer to type in the online world
forever. But neither are the current generations of outline formats, nor
will coming generations be much better for a while. In the mean time lets
use something that suffices, has some very cool applications, and can be
implemented with existing technology. Support for pixelfonts from HTML
won't be that complex, and by the time better formats for safe embedding
have been developed, HTML will have gone through 4 or 5 generations
anyway.
There is a big need for good type on the net, also stated in the piece,
but giving away outlines will have a worse long-term effect than the
temporary use of pixelfonts until we can figure out how to deal with
popular distribution of valuable bits of data, without enabling worldwide
piracy. There are some promising schemes that would allow the use of
fonts online, but also keep track of the risk of exposure, and assure
payment for the use of type (and not necessarily by the user!) Current
embedding proposals do nothing like it. They just ship the fonts and hope
for the best. I don't share that hope.
Many of the smaller typefoundries do not allow their fonts to be embedded
in webpages. That includes all fonts from FontShop and Emigre and many
others. But when embedding of outlines fonts becomes daily practice, it
means __all__ fonts are exposed to millions of people, by millions of
people.
I can sell a font to a client and tell him not to put the thing on the
web, but in practice there is not much I can do if he does. It does not
take much imagination, nor technical savvy for someone to write an
utility to 'unembed' the fonts in the documents.
Perhaps people at W3 are not that familiar with the value of typefaces
and perhaps unconciously the idea that "all software should be free"
leads to the conclusion that outline fonts should then also become part
of the public domain and it doesn't matter to give them away?
Well, it does not work that way. Typefoundries will have to respond with
tougher licenses and higher prices. Some fonts might actually be banned
from online use altogether. This is not good, neither for the foundries
nor for the users, and it is very much avoidable. Let's please try.
>So you don't much care to define what pixelfonts
>might mean or how they might be used, just that people shouldn't
>use outline fonts on the Web.
Wow. Please try to read the rest of the article again.
did you read the articles Paul Haeberli wrote on building pixelfonts?
What more do you need?
Not defining how pixelfonts could be used? Use netscape 2.0 (or any other
browser that supports animated gifs) and look at the page again. Then
think about what pixelfonts could do, and what you would have to do to
make outlines do that.
While you're at it, more trained pixels on:
http://www.letterror.com/movables.html
Pardon my emotional response when I'm trying to protect my discpline from
a very real threat. Typedesign might be a perhaps more obscure enterprise
than SGML programming, it is certainly not without merit, and perhaps
longer lasting.
erik van blokland, LettError
home of the randomfonts
letterror http://www.letterror.com
typelab http://www.dol.com/TypeLab/
Received on Thursday, 8 August 1996 19:59:05 UTC