Are you allowed to digitize an old font and sell it by its old font name?

Uli's picture

A Typophile member asked me by email the question, whether he is permitted to digitize an old font no longer sold by the original foundry and then sell this digitized font using the old font name.

Since this is a question which will interest other Typophile members too, I summarize here what I wrote to the member, who asked me the question. While the trademark laws vary from country to country, there are usually two principles applied in most countries and summarized here at the Wikipedia website http://en.wikipedia.org/wiki/Trademark as follows:

"Proprietary rights in relation to a trademark may be established through actual use in the marketplace, or through registration of the mark with the trademarks office of a particular jurisdiction."

The Typophile member asked me, whether he is permitted to digitize the old font "Erbar Medieval" and sell it by its original font name "Erbar Medieval".

Here is a sample of one of the bold cuts of this old font

www.sanskritweb.net/temporary/Erbar-II.jpg

drawn from a book by Georg Kandler

"Erinnerungen an den Bleisatz", Volume 1, 1995, page 26
("Reminiscences of the foundry type era")

Erbar Mediaeval is a typeface which was designed by the German Jakob Erbar, who died in 1935 in Cologne in Germany.

Even if fonts were copyrightable in Germany (which they are not), the "70-years-post-mortem" copyright protection period would have expired a couple of years ago (1935 + 70 = 2005).

However, fonts may be registered as designs in Germany for a period of 25 years starting with the first publication of the font. This period expired too.

Since Erbar died in 1935, all his copyrightable works (e.g. his books etc.), if any, and all his fonts registered as designs, if any, have already fallen into public domain.

Therefore today nobody can can claim any copyright to the works and to the fonts made by Jakob Erbar, because Erbar's death was more than 70 years ago.

Do not make the legal mistake of claiming copyright to digitizations of the public domain Erbar fonts, because nobody can claim copyright to something that is in public domain. Neither the heirs of Erbar nor anyone else can do that. Therefore, a digitization of Erbar Mediaeval will be public domain too, just like the original font made by Erbar is public domain.

Many people at the Typophile website erroneously think that they can become the copyright owner of a public domain font by digitizing this public domain font. This is legal nonsense.

As regards the name of a font, it may be registered forever, provided you pay the trademark registration fees forever. I did NOT research in depth whether anyone registered the name "Erbar Mediaeval" for a font. Such a research is cumbersome and time-consuming, because the name of a product may be registered in any country of the world.

As regards the website www.bauertypes.com, a certain "c. Dr. Ferran, nº 6-8 Entlo 2º A 08034 Barcelona Tel. +34 931127993 - Fax +34 932524806" seems to be responsible for this website. At the subsite

www.bauertypes.com/categorias/index/32/quienes-somos

we read:

"Trademarks: BauerTypes, como principal fabricante de tipografía en la era de plomo posee la patente o TradeMark de un sinfín de tipografías dentro del marco de los grandes clásicos. Entre las más destacadas se encuentran: ver listado"

However, the trademark list

http://www.bauertypes.com/arxiu/Pdfs/c/catalogo_nd.pdf

does not even mention fonts by Erbar as trademarks

The Neufville website www.neufville says here:

http://www.neufville.com/nd/uk/colof_uk.htm

"... Erbar Mediaeval ... are trademarks of Bauer Types SA"

"Neufville Digital makes no warranties as to the completeness or accuracy of this information."

The information that "Erbar Mediaeval" is a trademark of Bauer Types SA is inaccurate, because neither of the two requirements for a trademark protection are met:

Firstly, "Erbar Mediaeval" is not protected by registration
Secondly, "Erbar Mediaeval" is not protected by actual use

"Proprietary rights in relation to a trademark may be established through actual use in the marketplace, or through registration of the mark."

Since neither Bauer nor Neufville sell a font called "Erbar Mediaeval", this font name is no trademark by actual use. To mention a name of an old font in a list does not make you the owner of a trademark. You have to sell the font to achieve protection by actual use.

This means that the Typophile member is permitted to digitize the public domain font Erbar Mediaeval and to sell this font by its original font name Erbar Mediaeval.

However the Typophile member cannot claim copyright to his digitization of the public domain font. But he may sell the font against the payment of a font price (but not against the payment a copyright licence fee).

Remember: If you digitize an old public domain book, you do not become the copyright owner of this public domain book. The same holds true for old public domain fonts, irrespective of whether fonts are copyrightable or not.

Richard Fink's picture

Uli has a point here. There are several valid defenses to copyright infringement and equitable estoppel is one of them. Antitrust would be another. (And I would imagine *that* was considered a door best left unopened.)

And interestingly, there is also, in the US, something called Copyright Estoppel in which the copyright owner falsely represents portions of his or her work and then later asserts copyright in that material in order to sue a copyist. According to my reference material, this defense almost always succeeds where the copyright owner has made an express false representation.

That goes a long way in explaining why the first line in the Adobe Font Licensing FAQ is worded the way it is.

"Generally, font software is copyrightable."

Well, which font software do they mean? The fonts themselves as "font software"? Or Adobe Type Manager? Or what?
And then there's the weasel word, "Generally", implying that some font software might not be copyrightable. Well, which font software is that?

(I'm sorry, but stuff like this amuses me.)

Ya gotta love lawyers. The hardest thing to swallow about the legal system is that, as a system, it's arranged so that lawyers can earn a living. For some, a very handsome living indeed. Once you understand that, it all makes sense.

As Roy Cohn once said, "I don't care if the other side knows the law, as long as I know the judge."

Uli's picture

Judge Whyte's "summary judgment" has a Wikipedia entry:

http://en.wikipedia.org/wiki/Adobe_Systems,_Inc._v._Southern_Software,_Inc.

There we read:

"Since the font editing programs extracted control points from Adobe's fonts, use of those programs constituted copying of literal expression. Thus, King also infringed on Adobe's copyright through intermediate copying into RAM."

Was this Wiki entry sponsored by Judge Ronald M. Whyte?

- According to Judge Whyte, whosoever uses Fontlab or Fontographer, which extract the control points from fonts, commits copyright infringement by illegal copying.

- According to Judge Whyte, whosoever uses the Windows or Mac operating system, which extract the control points from fonts, commits copyright infringement.

- According to Judge Whyte, whosoever uses Adobe Acrobat and other PDF tools, which extract the control points from fonts, commits copyright infringement.

- According to Judge Whyte, whosoever uses Coreldraw or other drawing programs, which extract the control points from fonts, commits copyright infringement.

- According to Judge Whyte, whosoever uses Internet Explorer or any other browser, which extract the control points from fonts to display this Typophile thread you are reading at this very moment, commits copyright infringement.

Believe Judge Whyte, and you'll believe anything.

Uli's picture

13 years ago, Judge Whyte declared in his strange summary judgment:

"King also infringed when he loaded into his computer's random access memory copies of the Adobe program."

Now I prove step by step that the Whyte judgment is utter legal nonsense:

Step 1:

Load e.g. this PDF containing a Whyte judgment into your computer's memory:

www.greenisthenewred.com/blog/wp-content/Images/100712_aeta4_dismissed.pdf

Step 2:

Look at the blue running header:

"Case5:09-cr-00263-RMW Document177 Filed07/12/10 Page1 of 14"

This running header starts with the capital letter "C" of the word "Case".

This "C" is typeset in Helvetica by this "program" embedded into the PDF:

%ADOBeginSubsetFont: KQPJUO+Helvetica AddGlyphs
...
%%Copyright: Copyright 2011 Adobe System Incorporated. All rights reserved.
...
/C <1C60D8A8C9B6079F6D8209747A86332E73924F1B07F271169856A53C9157
71D7C530ADE68D8CA49F0A6B3478F9E8176949771F40D91B68E1A88C72CE81CC
0A91B3CFB642960E7EB2127A1637D49961A11F5DD40622F39379CDAB7E5B349D
F832E4955008C5E3208AB258B012A237635179513D820ACF76E0A17ED392C351
599687E8E7D0> |-

Step 3:

According to Judge Whyte, unless you have licensed the "program" of the letter "C" of Helvetica, you just committed the crime of copyright by loading the "program" of the letter "C" into your computer's random access memory.

This is utter legal nonsense.

According to Judge Whyte, millions of people, day by day, commit millions of crimes of copyright or millions of copyright infringements by downloading and viewing millions of PDF files from millions of Internet websites.

oldnick's picture

Was this Wiki entry sponsored by Judge Ronald M. Whyte?

All of the programs you mention UTILIZE the points, by way of the OS, to render the fonts; they do not EXTRACT the points and copy them over to a renamed font...apples and oranges.

Uli's picture

oldnick:

> they do not EXTRACT the points and copy them

In the example of glyph "C" shown above two extractions happened:

1) Acrobat Distiller extracted the glyph "C" from the font file.

2) Acrobat Reader extracted the glyph "C" from the PDF file.

According to Judge Whyte, these are two copyright infringements.

If Acrobat Distiller extracts a glyph from a font file and inserts it into the PDF file, this two-step process is done by copying the glyph from the font file into the "computer's random access memory" (Whyte) and then by copying the glyph from the "computer's random access memory" into the PDF file. Whyte said: "King also infringed when he loaded into his computer's random access memory copies of the Adobe program" and declared this copying process a copyright infringement.

If you view the Whyte PDF file

www.greenisthenewred.com/blog/wp-content/Images/100712_aeta4_dismissed.pdf

with Acrobat Reader, the opposite takes place. Acrobat Reader copies the glyph of "C" into the "computer's random access memory" (Whyte), and then this glyph of "C" is rendered (I skip the intermediate steps). Whyte declared this copying process a copyright infringement too, because for Whyte any loading of copies of glyphs into the computer's random access memory is a copyright infringement for Judge Whyte.

Richard Fink's picture

@oldnick

Whyte's statements are clear. Uli is correct.
If you are incredulous, it's an indication of just how badly copyright law fits with digital tech.
You are probably just having a hard time believing that a 'copying' into RAM - which we all know is necessary to view anything on a computer - counts as a 'copying' under copyright law.

Rich

oldnick's picture

My EULA--and, I suspect, that of many others--allows embedding in PDF files, so no copyright laws are broken if the "extraction" is done with my permission. Apples and oranges again. And, while Whyte's choice of words was unfortunate, his intent was clear: copying another's copyrighted work VERBATIM and claiming it as your own is plagiarism, and subject to prosecution.

Uli's picture

Richard Fink:

> Whyte's statements are clear. Uli is correct.

Thanks for telling the truth.

oldnick:

> My EULA--and, I suspect, that of many others--allows embedding in PDF files, so no copyright laws are broken if the "extraction" is done with my permission. Apples and oranges again.

You simply don't get it (or don't want to get it). If someone downloads one of the millions of PDF files from one of the millions of websites, this has nothing to do with your EULA, which is only a contract between you and a font buyer of yours.

According to Judge Whyte, if someone downloads a PDF file with a font embedded, he is committing copyright infringement. Judge Whyte was not clever enough to understand the consequences of his strange judgement. By his bizarre judgment, Judge Whyte makes millions of internet users and millions of PDF file users to copyright infringers. Therefore his judgment is legal nonsense.

Uli's picture

If Judge Ronald M. Whyte views this PDF of his own judgement:

www.sanskritweb.net/temporary/Whyte-Judgment.pdf

he commits copyright infringement according to his judgment.

After viewing this PDF file, Judge Ronald Whyte will state:

"I, Judge Ronald M. Whyte, also infringed when I loaded into my computer's random access memory copies of the Adobe program."

Typophile members should not view this Whyte-Judgment.pdf.

oldnick's picture

his judgment is legal nonsense

So, file a legal brief in a U.S. court challenging his decision; until you do, you're merely venting your prejudices to no avail, except the continued annoyance of others.

Uli's picture

oldnick:

>So, file a legal brief in a U.S. court challenging his decision

As a European, I could not do that. Besides, US judgments are irrelevant for Europe.

But even if I were an American and offered the VeracitySSK ripoff clone for sale, Adobe would not dare to sue me on account of the preposterous Whyte judgment.

Google for VeracitySSK, and you will see that dozens of outfits offer this clone, but none of them was ever sued by Adobe, since the Whyte judgment is legal nonsense.

Nick Shinn's picture

There is also the court of professional opinion, which counts for something if you want to make a career rather than just a buck.
In this court, the larger your business, and the longer-established your reputation, the more you can get away with.

Uli's picture

Considering that Judge Ronald M. Whyte only blamed Paul King for selling copies of Utopia as Veracity, but never blamed all the many other "criminal offenders" (§ 506 US Copyright Law) for selling copies of Utopia as Veracity, I wonder whether Judge Ronald M. Whyte heeds the constitutional principle "All men are created equal".

Richard Fink's picture

@all

I don't know what "legal nonsense" means, all I can know is how courts have ruled. And upon reflection, if I'm taking issue with someone who has the bona fides of Thomas Phinney, maybe I ought to back up what I say a little.

First, it has to be noted that all typefaces are unprotectable and in the public domain in the US. It doesn't matter if the typeface was created yesterday or 400 years ago. The title of this thread carries the implication that it could be otherwise - that one typeface could be "more" public domain than another. Hah! Ain't so.

In case law, there are four cases that appear to have special significance in light of the Adobe v SSI ruling:
L. Batlin And Son v Snyder decided by the US Court Of Appeals, Second Circuit in 1976.
Bridgeman Art Library v Corel Corp decided by a US District Court in 1999.
Meshworks v Toyota Motor Corp decided by the US Court Of Appeals, Tenth Circuit in 2008.
Lastly, there's the Supreme Court opinion from 1991, Feist v Rural.

When you read these opinions along with their cross-citations in total, it's hard see Whyte's decision in Adobe v SSI as anything other than an abberration.
(And BTW - in Whyte's decision there is a quotation from an Announcement in the Federal Registry by the US Copyright Office titled "Registrability Of Computer Programs That Generate Typefaces" that is completely misinterpreted. If you read the whole thing (2 pages) it's perfectly clear that the "computer programs" it refers to are what we would normally refer to as rasterizers, not TrueType or PostScript fonts. And it is carefully worded to make no claims as to the validity or level of copyright protection that may be obtained by registration, no matter how you interpret what kind of software it references.)
This is just one of several glaring errors in the SSI decision. IMHO.

My guess is that King was not a sympathetic defendant. And Whyte was happy to go along with a cherry-picking of the facts and precedent in order to find against him. It happens, and maybe King had it coming and the good guys won. Who knows.

Anyway, it's impossible to reconcile SSI with these other decisions.
Adobe lucked out, that's all. Speaking only for myself - I would have to have especially opaque blinders on not to consider fonts, all fonts - certainly, absolutely, the outlines - as public domain. Period.
Meshworks especially, being an appellate decision, has a far greater weight of authority than SSI. And it would, I imagine, be controlling in any "fonts as copyrightable software" case since the facts are nearly identical.

Before the advent of computers all typefaces were public domain. With the advent of computers, a claim was made that, as digital creations, typefaces could be copyrighted as software.
The courts, on the whole, say this ain't so. It's pretty clear, frankly.

And this might partially explain why Uli's dog hasn't barked, that is, why there's been no infringement suits in the US for thirteen years. (But legal costs and the overall effetiveness of licensing can explain it, too.)

Note that licensing is contract law and another matter entirely.

rich

Uli's picture

Richard Fink:

You made a high-class contribution to this thread.

Many thanks. Having read the Meshworks judgment,
it hits the nail on the head. It's a must-read.

William Berkson's picture

Maybe I don't get it, but I don't see how the Meshworks undermines Adobe vs SSI. The point was the Meshworks took points from a Toyota design, and so the software didn't, in the view of the court, represent anything new and copyrightable. That's not the case for Adobe's fonts.

Thomas Phinney's picture

Two things:

1) Richard, you are utterly incorrect in your interpretation of the Copyright Office decision. I think this part of the decision is fairly telling: "The creation of scalable font output programs to produce harmonious fonts consisting of hundreds of characters typically involves many decisions in drafting the instructions that drive the printer. The expression of these decisions is neither limited by the unprotectable shape of the characters nor functionally mandated. This expression, provided it meets the usual standards of authorship, is thus registrable as a computer program."

If you think this refers to rasterizer software rather than the font program itself, you're nearly alone in that belief. Most notably, the copyright office itself clearly disagrees with you, which is why they started accepting registrations of copyright on fonts-as-programs immediately after this. there have been thousands of such registrations.

In light of the above quote (which is admittedly more vague than one might like), I would argue Judge Whyte was simply clarifying the law by saying that the placement of points on bezier vector outlines was not functionally mandated and hence protected by copyright, and by declaring so as part of a summary judgment he said it was the case as a simple matter of law.

2) In considering other court decisions, you'll have to explain how and why you see them as contradictory to Adobe v SSI. For example, a phone directory and a font outline (possibly with hinting code) seem to me to be clearly not the same thing.

Cheers,

T

Richard Fink's picture

Thomas,

I would prefer just to leave it alone at this point. Summarizing the holdings from each case and connecting the dots is just too much work. And I'm under no obligation, thank heavens, to educate anybody on this issue except myself. It won't win me any points on anybody's scorecard. But I stand by my conclusion, though.

There's been too much FUD for too many years and that will continue. One self-interested person will say Tomato and another self-interested person will say Tomahto.
And without the concurrence of some leading legal people, it would have little to no impact, if even then.

That's why I just cited the cases so that anyone who wishes can make of them what they will. If I do write up a full analysis - and seek input from the legal community - you'll be the first to know. (I do intend to discuss it with my IP attorney, however.)

With regard to your interpretation of the Announcement in the Federal Registry my answer is this, to anybody who's interested and stumbles upon this thread:

1) Read the 1986 policy and the 1992 revision and come to your own conclusions.

I don't know enough about how the Copyright Office works to comment on why they might or might not issue a registration. The Announcement was about a procedural issue -about the need for an applicant to file a disclaimer, a procedure adopted with its previous policy decision in 1986 - which they discontinued. Other than that, I don't know anything other than what it says and it's hard to square what it says with the notion that they're talking about something akin to an OpenType font unless you cherry pick paragraphs. And there is the added complication that protectable material could certainly be included in an OpenType file. But not a typeface AS typeface.

See this scholarly essay here, for another take on this.

2) Being able to register a font is proof of nothing. As the court answered in Bridgewater v Corel. [Emphasis mine]

"plaintiff argues also that the fact that the Register of Copyright issued a certificate of registration for one of plaintiff's transparencies demonstrates that its photographs are copyrightable under U.S. law. The argument is misguided."

"No one disputes that most photographs are copyrightable. In consequence, the issuance of a certificate of registration for a photograph proves nothing. And while the certificate is prima facie evidence of the validity of the copyright, including the originality of the work, the presumption is not irrebuttable. Here, the facts pertinent to the issue of originality are undisputed. The Court has held as a matter of law, and reiterates, that plaintiff's works are not original under either British or United States law."

In other words, in Bridgewater, the court rendered the registration certificate fit for wrapping fish.

BTW - In fishing up the links I kept discovering more and more legal or legal-like stuff. This looked interesting by a lawyer named Blake Fry.
Why Typefaces Proliferate Without Copyright Protection

Anyway, bottom line, as the decision in Bridgewater says, registrability proves nothing.

And now I'm outta heah.

Rich

Uli's picture

As regards Blake Fry's treatise, read this on page 439:

"Some argue (scilicet Devroye), however, that the code that describes a modern outlined computer font is not software at all, but a set of data points. If that’s accurate, protecting a computer font would be like protecting metal type because of the type-maker’s decisions about how to best hold his chisel when carving it."

I wish Thomas Phinney were listening to what mathematicians such as Luc Devroye have to say about this matter instead of listening to those who cheat dimwits by selling data points as computer programs.

Thomas Phinney's picture

Luc DeVroye promotes piracy and frequently misrepresents people and situations. I stopped paying attention to anything he wrote a long time ago. Moreover, Uli is now trying to change the discussion to what the status quo SHOULD be, but that's not the question at hand. If somebody thinks the copyright office is wrongheaded, that's fine, but that has no bearing on the question in the title of this thread.

Richard, perhaps you should note the sentence right before the one you emboldened. Read the FIRST sentence this time: "No one disputes that most photographs are copyrightable. In consequence, the issuance of a certificate of registration for a photograph proves nothing."

In other words, a certificate of registration for a *particular* font doesn't tell you anything about the legitimacy of that particular font registration. I could rip off the code of a digital font and rename it and register the copyright, and if it ever got contested, I'd lose. But the fact that the copyright office routinely issues copyrights for fonts as software IS telling, and your quote doesn't say otherwise.

I'm not claiming that typefaces are protected by copyright in the USA. But it seems reasonably clear that digital fonts are, as software. And by the way Blake Fry thinks they likely are as well (whether or not he thinks it reasonable that they be protected in that way). He has a whole section of that treatise entitled "Computer Fonts are (Probably) Protected as Software."

Cheers,

T

Té Rowan's picture

Them's libellous words, T-man!

Uli's picture

1) Luc DeVroye promotes piracy
2) Thomas Phinney promotes fraud

Professor Luc Devroye claims that data points are not computer programs. This is not the same as promoting piracy.

Thomas Phinney claims that data points of glyphs are computer programs. This is not the same as promoting fraud.

BUT WATCH OUT:

1) If data points ARE computer programs, then anyone viewing this PDF

www.sanskritweb.net/forgers/veracity.pdf

would commit the crime of copyright infringement. Do you know, why?

2) If data points ARE NOT computer programs, then any font company licensing fonts as computer programs would commit the crime of fraud. Do you know, why?

William Berkson's picture

Uli, if I remember correctly, all computer programs are, in machine language, sets of numbers. Are the numbers that indicate glyphs, spacing and kerning part of an instruction? Well, in conjunction with a rasterizer program, the numbers in the font direct the computer or printer on what to put on the screen or put on the page. The rasterizer by itself can't do anything. It needs the input to be operational. Hence one could argue that the "rasterizer + font information" is intended to be a "program" for producing letters. Or you could argue that since the lines that are obviously instructions, except for open type programming, are not in the font, the font is not a program.

So are the points part of a program or not? It seems to me you could argue either way, and the situation is rather ambiguous. The judge in Adobe vs SSI ruled "program." You want to make it obvious that it is one way or the other, when it is not. It is a borderline issue. And in such cases, judicial decisions create the law.

I'm not saying that this is the best way to deal with IP protection for fonts, but it is what we've got now in the US.

You keep raging about the situation being black-and-white, and everyone else being fools, but that's you oversimplifying the situation, and not situation itself, which is messy and ambiguous.

Thomas Phinney's picture

Libel has to be untrue. Luc DeVroye has provided links to pirated font downloads.

Uli is again clouding the issue, in two ways:

1) I was not representing my personal opinion, but commenting on what is generally accepted as being the state of US copyright law. I'm saying “this is how the law is treating it.” Trying to express that as my personal opinion and contrasting it with Luc's personal opinions as to what the law ought to be is irrelevant to the current discussion. I have not been engaging in a discussion of whether copyright “ought” to cover digital fonts and how it ought to do so. Just because I make a statement that a situation is XYZ does not mean that I agree with that situation, nor even if I agree with the effect of that situation that I agree with how that outcome was reached.

2) I do not claim that data points of glyphs are computer programs, nor has anybody else I cited. I claim that the US copyright office treats a digital outline font, in toto, as a computer program (while being unfortunately vague about which elements are protected). I also claim that a federal court in Adobe v SSI ruled (in a summary judgment, as a matter of law) that the placement of points in those programs in particular is an element of creative expression protected by copyright. I made no comments about my own opinions on what ought to be, and my opinions are irrelevant to the question of "are you allowed to...?"

My own opinions on "what should be" are that *typefaces* should be protected under copyright as creative works, but that US copyright terms are insanely long and should be shortened dramatically.

I do actually partly agree with one thing Uli is saying, although he's been rather oblique about it. Given that digital fonts are protected by copyright, if loading them into computer memory can constitute infringement, then there would seem to be a potential issue with many normal and accepted uses of fonts that font vendors are actually okay with, even involving third parties who have not been involved in the font licensing process. Opening a PDF loads its fonts into memory, and the person opening the PDF might not have any involvement in licensing that font.

Of course, there would seem to also be some very simple solutions to this. One would be to consider such actions to be fair use by the viewer when they were occurring simply for the purposes of viewing documents with embedded or linked fonts, and the viewer was not duplicating the font (which would not mean that setting up that situation by embedding or linking the font is automatically fair use). Another would be for the font vendors to clarify in their copyright statements what uses they permit for unlicensed people. This would be similar to copyright statements that grant certain exemptions beyond fair use. Uli seems to think this is a big deal. I think it's a minor issue, and if it ever came up in court in a more worrisome context, font licensors and copyright holders would deal with it more actively. They'd all go talk to their lawyers about the “right” way to do it, and one or more legally viable approaches would emerge. They might or might not be like the ones I outline here... again, I'm not a lawyer, nor are any of the other folks in this discussion.

Actually, the whole discussion has veered off of "are you allowed to...?" The OP was asking about the legality of selling a new digitization, and had no intention of using existing digital fonts as a basis for said digitization. The comment was made that a new digitization was fine as long as one didn't directly input existing digital fonts as the basis for the new version and we wandered off....

Anyway, back to my day job. Time to go look at some font metadata.

Cheers,

T

Nick Shinn's picture

Luc doesn't promote piracy.
His web site is value-neutral, in an academic manner.
It includes links to a great many font sites, including pirate sites, because its reach is broad.

Search engines do a better job of promoting piracy.

Uli's picture

William Berkson:

Since you address me directly, I will answer directly.

Since I have written several books about computer programming, I know what I am talking about. If someone claims that the data points of glyphs are computer programs, I cannot but laugh.

Look at this trivial BASIC two-liner:

1 DATA C,O,M,P,U,T,E,R
2 FOR I = 1 TO 8: READ LETTER: PRINT LETTER;: NEXT I

The data segment is line 1, and the code segment is line 2.

Data and code may be mixed, but they are and remain two completely different things.

For TTF fonts, I have proved irrefutably, that the sequence of numbers describing the x-y coordinates of the outlines of TTF glyphs are no code, but data only.

see www.sanskritweb.net/forgers/computerprogramm.pdf, pages 5-7

(This does not apply to TTF hints, which are code, but this is not the topic of this thread. For TTF hints as code, see above file computerprogramm.pdf, pages 8-10).

Theoretically, it would be possible to program a huge mixed code-and-data rasterizer including as code segment the rasterizer proper and as data segment the entire Adobe Font Folio library. Nevertheless, you could distinguish the code (= rasterizer proper) and the data (= fonts proper).

Since you are no computer programmer, let me explain it in this metaphorical way:

If you have a cup of coffee, you can distinguish between the cup (code) and the coffee (data). The cup remains the same, but the beverages may change (coffee, milk, water). Likewise, the rasterizer (code) of a given operating system remains the same, but the fonts (data) may change, i.e. the same rasterizer code may process different font data (e.g. Helvetica, Times, etc.)

If a judge shows this sequence of numbers

-227 0 0 227 160 160 228 0
0 -228 -153 234 333 0 0 -333 etc.

(see www.sanskritweb.net/forgers/computerprogramm.pdf, page 6)

to a mathematician and claims that this sequences of numbers is a computer program, the mathematician will ask the judge, whether he is sane in the brain.

Judge Whyte, who obviously thinks that anyone, e.g US Chief Justice John G. Roberts, is a computer programmer, if he draws e.g. an egg with a drawing program, will never understand that drawings are no computer programs, nor will he ever understand that sequences of numbers are no computer programs.

see www.sanskritweb.net/forgers/publicdomain.pdf, page 2

Since Adobe knows that the Whyte judgment is nonsense, Adobe never dared to sue anyone selling the Veracity alias Utopia font during the past decade.

Uli's picture

Thomas Phinney:

> what the law ought to be is irrelevant to the current discussion.

The Whyte judgment ruled:

"King also infringed when he loaded into his computer's random access memory copies of the Adobe program. See Sega Enters. v. Accolade, Inc., 977 F.2d 1510 (9th Cir. 1992). His fair use defense fails because his intermediate copying utilized unauthorized copies of the Adobe programs in an apparent attempt to avoid being a party to Adobe's license agreement."

This is what IS (according to Whyte), not what OUGHT to be (according to Whyte).

But this has the legal consequence (according to Whyte) that anyone viewing the PDF

www.sanskritweb.net/forgers/veracity.pdf

commits the crime of copyright infringement.

For instance, Thomas Phinney, who downloaded this PDF file committed the crime of copyright infringement, unless he is "a party to Adobe's license agreement" .

Thomas Phinney, who all of a sudden claims that one OUGHT to consider downloading of PDF files containing fonts as fair use, now deliberately disrespects the Whyte judgment ruling that this is NOT fair use and that anyone downloading PDF files with embedded fonts is a criminal, because, as Thomas Phinney correctly states, "Opening a PDF loads its fonts into memory, and the person opening the PDF might not have any involvement in licensing that font."

For instance, if you search in Google for "filetype:pdf"

www.google.com/search?&q=filetype%3Apdf

you get 168.000.000 hits (one hundred sixty eight millions).

This is the number of possible crimes committed by downloading PDF files.

Hence, the bizarre Whyte judgment makes the entire world to criminals.

Nick Shinn's picture

If someone claims that the data points of glyphs are computer programs, I cannot but laugh.

And if someone were to claim that computer programs are data?

All software is data, including code.
That is why it is called software—the electronic information which is stored in, and used by, physical hardware.
The distinction between code and data is specious, in so far as digitized information is concerned.
It's all bytes, which is pure data.

…but some data are more equal than others

Uli's picture

> All software is data, including code.

I suggest to try to read some books about this topic, e.g.

http://www-cs-faculty.stanford.edu/~knuth/taocp.html

Many students who tried to understand books on programming and e.g. who tried to understand the distinction between code and data soon discovered that they are not intelligent enough to talk about this difficult subject. Many students, who thought that they are clever enough to understand the Knuth books never managed to understand them. Ask Professor Luc Devroye, if you do not believe me.

Uli's picture

Thomas Phinney:

> Opening a PDF loads its fonts into memory, and the person opening the PDF might not have any involvement in licensing that font. Of course, there would seem to also be some very simple solutions to this. One would be to consider such actions to be fair use by the viewer when they were occurring simply for the purposes of viewing documents with embedded or linked fonts, and the viewer was not duplicating the font (which would not mean that setting up that situation by embedding or linking the font is automatically fair use). Another would be for the font vendors to clarify in their copyright statements what uses they permit for unlicensed people.

All these proposals are legal rubbish. You cannot force font licensors to adopt fair use rules for all fonts. This could only be done by law, not by contract. But since this strange American Judge Ronald M. Whyte ruled that it is forbidden by fair use to copy unlicensed fonts into computer memory, everyone downloading PDF files with embedded fonts commits the crime of copyright infringement, unless he licensed these fonts before downloading the PDF files. Therefore, those who offer PDF files for download and who think that the bizarre judgment by the strange Judge Ronald M. Whyte is no legal rubbish, must provide the download link with the following proviso (example):

Warning: Do not download the following PDF, unless you have acquired a licence to the Utopia font. If you download this PDF without a licence to the Utopia font, you will commit the crime of copyright infringement according to the judgment by Judge Ronald M. Whyte:

http://www.sanskritweb.net/forgers/veracity.pdf

Uli's picture

Nick Shinn:

> "All software is data, including code."

Professor Luc Devroye says:

- "Programs are data", but
- "Data are not programs".

I agree to the latter.

I cannot agree to the former from the point of view of European copyright laws.

For example, in Europe, programs are protected for a period of 70 years post mortem auctoris, whereas databases are protected for a period of 15 years only:

"In the case of a database ... the term of protection by that right shall expire fifteen years from the first of January of the year following the date when the database was first made available to the public."

http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=CELEX:31996L0009:E...

If fonts were databases, the period of protection would be even shorter than the period of protection for the design of typefaces, which is 25 years:

www.sanskritweb.net/forgers/publicdomain.pdf

Thomas Phinney's picture

"But since this strange American Judge Ronald M. Whyte ruled that it is forbidden by fair use to copy unlicensed fonts into computer memory"

He said no such thing.

The law:

"Notwithstanding the provisions of sections 17 U.S.C. § 106 and 17 U.S.C. § 106A, the fair use of a copyrighted work, including such use by reproduction in copies or phonorecords or by any other means specified by that section, for purposes such as criticism, comment, news reporting, teaching (including multiple copies for classroom use), scholarship, or research, is not an infringement of copyright."

It actually seems improbable that copying a font into memory *in general* could be construed as fair use. Presumably it would depend on the purpose of said copying. That was probably a goofy thought on my part. But in any case, the judge did not say that copying an unlicensed front into computer memory is "forbidden by fair use." That's part of your ongoing campaign to put words in his mouth.

Uli's picture

> But in any case, the judge did not say that copying an unlicensed front into computer memory is "forbidden by fair use." That's part of your ongoing campaign to put words in his mouth.

Here is his mouth again:

"King also infringed when he loaded into his computer's random access memory copies of the Adobe program. See Sega Enters. v. Accolade, Inc., 977 F.2d 1510 (9th Cir. 1992). His fair use defense fails because his intermediate copying utilized unauthorized copies of the Adobe programs in an apparent attempt to avoid being a party to Adobe's license agreement." (www.sanskritweb.net/forgers/veracity.pdf, page 5, section 5)

You should accept your defeat. And you should admit that Adobe never dared to sue anyone during the past 13 years for selling Veracity.

Thomas Phinney's picture

That's exactly my point: his fair use defense failed for specific reasons, not because fair use could *never* apply to loading fonts into RAM.

Aside from which, your “never dared” might be somebody else’s “never bothered.”

Uli's picture

> Adobe "never dared" to sue anyone during the past 13 years

> Your "never dared" might be somebody else’s "never bothered"

For those companies which offer for free download unlicensed copies of Adobe fonts, it does not matter, whether Adobe "never dared" or "never bothered" to sue them.

Besides torrent, peer-to-peer and similar websites, it is interesting to note that there are renowned advertising agencies offering the entire FontFolio collection to their customers for free download. These promotion companies do not care to make their download sites accessible via password only, so that Google, Yahoo etc. index the Adobe fonts freely downloadable from these websites. Who searches for Adobe font file names, will find these advertising agency websites offering thousands of Adobe FontFolio fonts for free download.

What is the opinion of Typophile readers?

- Adobe "never dared" to sue these advertising companies ???
- Adobe "never bothered" to sue these advertising companies ???

Uli's picture

How Paul King forged Utopia is shown pictorially here:

http://www.sanskritweb.net/forgers/veracity2.pdf

Richard Fink's picture

Despite my recurring nightmare about receiving one of Thomas Phinney's toes in a zip lock bag via Federal Express. (Careful with those metaphors, Thomas!) I've come back.

Uli said, regarding fair use,

>This could only be done by law, not by contract.

Uh, I wouldn't be so sure about that. The courts in America, at least, seem to have never seen a contract provision they don't like. Product liability laws are the obvious example. EULAs allow sofware providers to weasel out of any responsibility for how their software might misbehave. Consumers use at their own risk - by contract, not statute.

Uli's picture

> Uh, I wouldn't be so sure about that. The courts in America, at least, seem to have never seen a contract provision they don't like.

I was thinking of third-party-provisions. At least here in Europe, two parties A and B cannot stipulate a contract to the detriment of a third party C. For instance, if a third party C downloads a PDF with a font embedded which is copied into the RAM of his computer, this is illegal according to Judge Whyte, but the parties A and B cannot make a contract which allows or forbids the third person C to download the PDF, unless the third party also signs the contract.

Nick Shinn's picture

@Uli: I cannot agree to the former [all software is data, including code] from the point of view of European copyright laws.

My observation was philosophical and etymological, not legal.
It is also the opinion expressed by Luc Devroye ("Programs are data", which you quoted), so please don't patronize me as "not intelligent enough to speak about the subject"!

I have no desire to read Knuth's The Art of Computer Programming, which I suspect contains precious little art.
My literary allusion ("…but some data are more equal than others") was to Animal Farm, the inference being that this is a political issue as much as a legal one.

It is the role of technological progress to reduce cultural artefacts to "content", minimizing their worth, and this is the foundation of the value of new media devices (both hard and soft). But typefaces are not just an old analog form automatically reduced to data (with programs being a more privileged species of code, to the vested interest of technologists). They are written directly into digital media, and even revivals of metal typefaces are necessarily unique interpretations—as would be the case with the 24 pt Erbar font that is the subject of this thread; the specimen has sufficient traces of process that will require "cleaning up", or not, as the case may be, that any digitization would have some measure of originality.

Uli's picture

Nick Shinn:

> My observation was philosophical and etymological, not legal. ... so please don't patronize me

I understand. Sorry.

**************************************************

Indicentally:

During the past few days, numerous people downloaded the file veracity.pdf.

According to Judge Whyte, they committed the crime of copyright infringement.

Because according to Judge Ronald M. Whyte, the downloader of the file veracity.pdf
"infringed when he loaded into his computer's random access memory copies of the Adobe program (scilicet Veracity alias Utopia). His fair use defense fails because his intermediate copying utilized unauthorized copies of the Adobe program in an apparent attempt to avoid being a party to Adobe's license agreement."
(www.sanskritweb.net/forgers/veracity.pdf, page 5, section 5)

For Judge Ronald M. Whyte, it will be easy to identify the suspects who downloaded veracity.pdf thus loading copies of the font Veracity alias Utopia into his computer's random access memory.

Therefore I say to Judge Ronald M. Whyte:

"Round up the usual suspects." (Casablanca)

Richard Fink's picture

Uli - it no longer matters what Judge Whyte thought in 1998. But he was not alone in considering a RAM copy, a copy under the law.

It is now not infringement to view the PDF under the Title 17 statute in the US Copyright law. Loading a "program" into RAM as a normal part of utilization is not infringement.
According the following quotation from Cornell University, this was added to the law in 2000.
Limitations On Exclusive Rights: Computer Programs

Thomas Phinney's picture

I wish Uli would stop lying. I also wish more people would call him on it, but I expect most have realized it is pointless to do so.

Uli's picture

Richard Fink:

> Uli - it no longer matters what Judge Whyte thought in 1998. But he was not alone in considering a RAM copy, a copy under the law. It is now not infringement to view the PDF under the Title 17 statute in the US Copyright law. Loading a "program" into RAM as a normal part of utilization is not infringement. According the following quotation from Cornell University, this was added to the law in 2000.

I understand that you want to help Thomas Phinney, who behaves like a cornered rabbit, but referring to § 117 does not solve the problems of the Whyte judgment.

§ 117 of US Copyright Law is old hat added before the Whyte judgment and enacted December 12, 1980. See here page 1 and page 14:

www.history.nih.gov/research/downloads/PL96-517.pdf

§ 117 states that "Making of additional copy ... is not an infringement for the owner of a copy of a computer program".

It is obvious that the downloader of a copy of a computer program is not the owner of the copy of the computer program, nor does the downloader become the owner of a copy by copying the copy into the random access memory of his computer.

Té Rowan's picture

@Uli - Law or no law, the common concensus is that once you buy a copy of a commercial program, you are the de-facto owner of that copy, only without the rights to copy except as needed for the normal operation of the program. Judge Whyte may have made a precedent, but he does not make law. Only institutions like Congress, Parliament and the Bundestag get to do that. Nope, Ministries/Departments make regulations and must base them on existing law.

Richard Fink's picture

Actually Uli, you might be right about the date, it seems.
Cornell Univ's citation might be wrong or misleading.
Will check.

Rich

cdavidson's picture

This is the first time I have chimed into this debate. Being in the law field myself, I always get concerned when I see other people quoting the law. I wish I knew about IP law, but regrettably I've never had the need to look into it.

I have two questions:
1. Is anyone in this thread a lawyer?
2. Has anyone consulted a lawyer on this topic? Despite their price, they do know what they're on about.

Té Rowan's picture

Lawyer? Not me. Just insane(ly bored). Where's the porridge?

Thomas Phinney's picture

CaleD:

1) Not me. Nor is Uli, nor as far as I know anyone else on the thread.

2) Yes, I have had extensive legal counsel on the topic, many hours of discussion, and at one point vetting an article I wrote summarizing the IP status of fonts. All this mostly focused on US legal status, but with moderate attention to the areas that differ in other countries.

Of course anybody who is concerned about the potential of getting sued (or charged with criminal copyright infringement) would be well served doing something other than reading a bunch of non-lawyers debating the topic in a public forum. If nothing else, perhaps this thread shows what the key legal issues are that everything turns on.

Cheers,

T

piccic's picture

I have got to read all this later on…

Anyway, I still don’t understand why typeface designs (not the digital versions, which are technical instances, as software) are so problematic to be considered per se.
The issue of "copyright" does not coincide with actual protection, right? It seems to me it just involves the problem of making a copy, a question which has become quite a topic with the "immateriality" of digital data sets.

Syndicate content Syndicate content