Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!usc!elroy.jpl.nasa.gov!ncar!ico!rcd
From: rcd@ico.isc.com (Dick Dunn)
Newsgroups: comp.lang.postscript
Subject: Re: How to tell a pirated PostScript font?
Message-ID: <1991May9.214741.3352@ico.isc.com>
Date: 9 May 91 21:47:41 GMT
References: <28986@spudge.UUCP> <1991May6.225858.5945@ico.isc.com> <29033@spudge.UUCP>
Organization: Interactive Systems Corporation, Boulder, CO
Lines: 54

johnm@spudge.UUCP (John Munsch) writes:
> rcd@ico.isc.com (Dick Dunn) writes:
> >[ramble about comparisons]
> >...You might start by checking the unique-id values between your
> >fonts and the licensed ones (e.g., from Adobe) - a match would be awfully
> >suspicious.

What I had intended here--but neglected to say (I seem to have assumed my
posting had a telepathic sidebar!)--is that those of us who have licenses
to some of the Adobe versions might be able to offer bits of info to help
John.

> This is not a flame against Dick Dunn but against all the people who didn't
> bother to read what I said in my original posting.

> I distinctly said in my original posting that comparing to the actual fonts
> from Adobe was out of the question.  If I had the real fonts (all $1900
> worth according to my latest Adobe catalog) I wouldn't be looking at PD
> fonts... would I!  And yet, it has been suggested to me that I compare the
> file sizes, compare font IDs, and even print out samples of each at 72 pt.
> and compare the outlines!

You want to know whether a program you have is equivalent to a program you
don't have, and you want to make the comparison without any information
about the program you don't have.  That doesn't leave much--the only hope
is that you can find something in the program you have which could only
have been put there by (say) Adobe [or the creator of the original font].

> No kidding, I think I could have discerned that all by myself.  What I want
> to know is if there is a way to check WITHOUT owning the typefaces from 
> Adobe to compare.

I assumed you understood the nature of Type 1 fonts--that they're just pro-
grams, and (in theory) anyone can write one.  Therefore, suppose that a
thief has a font program F.  Suppose there is some unique characteristic of
that font program which would identify it as having been created by Adobe--
the thief just creates F' which is the same as F with the distinguishing
characteristic removed.  Any distinguishing characteristic which becomes
known can probably be removed without much trouble.  If Adobe puts some
sort of hidden marker inside a font program, they're unlikely to tell the
world about it, since that just makes it easy to find and remove it.

Am I missing something here?  If we have two programs, A and B, and we want
to know whether A and B are "essentially the same" without examining
either B or its output, we're stuck looking for some marker in A that could
only have been put there by the author of B.  The marker is either inci-
dental (has no effect on output), in which case it can be removed, or it's
essential (affects output), in which case something else can be substituted
to produce nearly-identical output.  A smart thief won't leave a marker in
the code unless he doesn't know it's there.  If he doesn't know it's there,
how are we to know?
-- 
Dick Dunn     rcd@ico.isc.com -or- ico!rcd       Boulder, CO   (303)449-2870
   ...If you plant ice, you're gonna harvest wind.