@uli

AFAIK - there is a distinct difference between what the copyright laws and copyright case law might or might not define as infringement and what the contract laws that apply to the EULA agreement might or might not allow you to do.

All I can advise is that if you don't like this particular provision of the contract, either ask them to change it before you buy or buy elsewhere. There's no law I know of that says you have to accept the terms of a non-negotiated contract.

Rich

[tracking]

That would clearly be an infringement. Even just opening the font in
an editor and not saving anything out is usually technically verboten.

hhp

Fortunately there aren't really any good reasons these days to convert a font from TrueType to PostScript. Correct me if I'm wrong.

Mr. Fink:

> All I can advise is that if you don’t like this particular provision of the contract, either ask them to change it before you buy or buy elsewhere

This would be no solution, as "derivative works" are forbidden by most EULAs.

Mr. Coles:

> Fortunately there aren’t really any good reasons these days to convert a font from TrueType to PostScript. Correct me if I’m wrong

You are wrong. There are millions of people, who convert TrueType to PostScript.

Mr. Papazian:

> That would clearly be an infringement

If this were true, then millions of people would commit infringements.

Wrong, billions.

Have you been on US freeways? People who
drive at the speed limit get hit from behind.

hhp

> You are wrong. There are millions of people, who convert TrueType to PostScript.

They might do it, but why?

@ Stephen

Fortunately there aren’t really any good reasons these days to convert a font from TrueType to PostScript. Correct me if I’m wrong.

1. A well hinted PostScript font is better than a badly hinted TrueType font. And it is easier to hint a PostScript font.

2. If you have two different formats in a document, which you print, I think, that the printer has to convert one of them. I trust more in FontLab than in my printer driver.

Probably there are more reasons.

@ Uli

If this were true, then millions of people would commit infringements.

Yes. Me for example. I give a f*** in those ridiculous restrictions. If there are bugs in fonts, that I can easily fix by myself, I fix them. In case of TT-PS-conversions, I ask the foundry for a PS version, because the conversion from PS to TT is lossy and most typeface designers draw the outlines in PS modus. And in case of PS-TT-conversions I would ask the foundry, because I am able to hint a PS font, but not a TT font, at least not in a high quality.

Internally, I'm with Arno; I harbor no qualms when I open up, mess with, and output any font for my own use. But I won't take on a commission that involves violating a EULA. Being in the font biz is why I draw the line there. Not being an officer of the law however, I have no issue with carving up traffic like a hot knife through butter.

hhp

@all

Uh, with web fonts there is a great need to convert OTF to TTF. Internet Explorer can't handle OTF. This is a situation that will go on for years.

I really can't predict about TTF to OTF conversions but OTF to TTF will happen on a big scale.
It's a sure bet.

Regards,

Rich

> Internally, I’m with Arno; I harbor no qualms when I open up, mess with, and output any font for my own use. But I won’t take on a commission that involves violating a EULA.

You completely misunderstood me. I did not think of the few nerds who use font editors. I thought of the millions of ordinary people, who use Microsoft's Word and Adobe's Acrobat.

Experiment 1

Open Word, type the letter "A" and mark it up with the TTF font "Palatino" supplied with Windows, but also purchasable at the Monotype website. Then select Acrobat and convert this Word file to a PDF file. The glyph of the letter "A" contained in this PDF file will be encrypted by this hex string:

adff0000063a0d948415829305bc07949305bc8baa8e998f089990989396
98089698a1b6acd308f4f77805f7a5f8fd05bef70dbef70dbef70d08cc06
f840fe8205c4fb1a059c659e62a15e08a05e9a7194830893839585988608
9787a688b68b089383055a0783830545904b8d528b08fb0288fb0389fb03
8908839305bc07939305d68cb98e9e90089e9094978b9d088b99869f82a3
08fb13f7c805fcd406fb01fb9705785d81698b76088b7a957f9d83089d83
b886d58b089483055a078383053490408d498b08428b40894086088b06f8
5ef8c115f88e06fb8ff8f105fb93fcf105090e

But this is not the TrueType glyph of Palatino "A". It is a PostScript Type 1 glyph of Palatino "A". Adobe's Acrobat ignored the Monotype EULA and the US Copyright Law and converted this TTF font to a PostScript Type 1 font.

This means that you would have committed a copyright infringement and/or a contractual EULA infringement, because the Monotype EULA stipulates: “You agree not to adapt, modify, alter, translate, convert, or otherwise change the Font Software, or to create Derivative Works from Font Software or any portion thereof.”

Experiment 2

Open Word, type the letter "A" three times and mark them up with the TTF font "Palatino", and then select the character options "Outline", "Shaded" and "Relief". In this way, Microsoft Word generates derivative versions of the Palatino "A". The result will look like this

www.sanskritweb.net/temporary/pala-outline.pdf

This means that you would have committed a copyright infringement and/or a contractual EULA infringement, because the Monotype EULA stipulates: “You agree not to adapt, modify, alter, translate, convert, or otherwise change the Font Software, or to create Derivative Works from Font Software or any portion thereof.”

If fonts were "Font Software", I say IF, which means, IF fonts were computer programs in the sense of the US Copyright Law, then millions of Americans would commit copyright infringements by using word processing programs and by creating PDF files.

If fonts were "Font Software", then thousands of Monotype customers who bought TrueType fonts and who signed this bizarre Monotype EULA would commit EULA infringements by using these TrueType fonts.

Why not take this up with Monotype rather than the court of public opinion?

> Why not take this up with Monotype rather than the court of public opinion?

Monotype's EULA is only an example. The same applies to Linotype, Ascender, Berthold, T.26, Letterhead, Typotheque, Underware, etc. etc. etc.

Uli, if Monotype doesn't mind, take a hint.

hhp

IF and only IF fonts were copyrightable "software" or copyrightable "computer programs", then it would be forbidden by the US Copyright Law "to create Derivative Works from Font Software or any portion thereof", as claimed in the Monotype EULA and many other font "software" EULAs.

IF fonts were copyrightable "software" or "computer programs", then it would be forbidden to convert fonts from one format (e.g. TT) to another format (e.g T1).

IF fonts were copyrightable "software" or "computer programs", then Microsoft and Adobe would be the biggest criminal offenders of copyright worldwide due to the millions of criminal font conversions done automatically for millions of computer users worldwide. Examples:

Windows NT automatically converted T1 fonts to TT fonts. Microsoft did not ask the millions of Windows NT users, whether they wanted to commit the criminal offense of copyright infringement, but Microsoft automatically committed these copyright crimes for computer users worldwide, IF and only IF fonts were copyrightable "software" or "computer programs".

And Adobe committing infringements in the opposite direction from TT to T1 does not ask the millions of Acrobat users, whether they want to commit the criminal offense of copyright infringement, but Adobe automatically commits these copyright crimes for computer users worldwide, IF and only IF fonts were copyrightable "software" or "computer programs".

I think you would have a very difficult time establishing to a judge's satisfaction that computer fonts are not software as defined by US law. The law is very broad in this respect:

A “computer program” is a set of statements or instructions to be used directly or indirectly in a computer in order to bring about a certain result. (US Code, Title 17, Chapter 1, Section 101)

I don't think Microsoft and Adobe would have a difficult time demonstrating that the inner workings of Windows NT and Acrobat violate neither the spirit nor the letter of the EULAs in question and that in any case Microsoft and Adobe should be held harmless.

@ Uli

I think, that you mess up EULA with copyright laws. With regard to the question, whether fonts are software or not, I partly agree: If all fonts are software, then any information, that can be stored on a computer, is likewise software – bitmaps, vector graphics, movies, music, texts and so on. That would be absurd. But how about the layout features of OpenType fonts? Aren’t they programs? And if the features are programs, are those OTF, that contain layout features, then programs likewise? I would say, there are not, but they are software. But how shall the user decide, which OTF are software? OTF don’t contain layout features necessarily.

aric:

> in any case Microsoft and Adobe should be held harmless

Are there any other companies which, in your opinion, are above the law and should be granted immunity from prosecution?

Windows NT automatically converted T1 fonts to TT fonts.

I did not know that. But I remember, that I was not able to print Type1 in the same quality anymore, when I switched from Windows 98 to Windows XP, because my inkjet had no PS abilities. Does anybody know, whether Windows 98 also has internally converted Type1 to TrueType? And if, does anybody know, where it is documented?

> Does anybody know, whether Windows 98 also has internally converted Type1 to TrueType?

No. (I still have Win 98 on one of my HD partitions.)

@uli, @aric, @arno, @all,

To an extremely large degree, the "law" regarding software as copyright and EULAs, which are really a part of contract law, is being made up by judges in the courts because the legislated law is inadequate, insanely complex, and in parts, vague. Perhaps this is for the best because in the US Congress, at least, you can bet the industries dependent upon copyright will buy off the US Congress and draft the legislation themselves (as they always have, historically).
So there are very few "bright lines" as they say in IP law lingo.
There are two general trends: EULA's, as contracts, no matter how broad, hold up in court very well. Infringement of software copyright on the other hand, has been interpreted more narrowly. This is because software is not a purely creative act like a poem - software is aimed at solving a problem and, therefore, different software which all aim to solve the same problem will tend to converge. If everyone on this thread were to go off and write a poem, there would be obvious divergence between each and every one of our efforts. (No limericks, please.) However, if we all created a Windows executable for copying a file from one directory to another, there would certainly be similarities. Therefore, the trend in determining copyright infringement of software has evolved to be much more narrow - as long as your software is not a byte-for-byte copy of something, there is considerable leeway.
That's my understanding of the situation.

rich

Are there any other companies which, in your opinion, are above the law and should be granted immunity from prosecution?

This is my assertion: in the examples you cite, Microsoft and Adobe should be able to demonstrate to the satisfaction of a judge that they have broken no laws and that they have violated no contract to which they are party.

Read this

www.sanskritweb.net/forgers/ssi2.pdf

and have fun (or puke).

The type is too small.

I read it, but it's not entirely clear to me what the copyright notice is intended to cover. I'll grant you that Adobe doesn't have the right to assert copyright ownership of the implementation of the letterforms, the associated metrics, etc. whether in TT or T1 format (unless in fact they hold the copyright for those letterforms, metrics, etc.), but surely they have a right to assert copyright ownership of the rest of the code responsible for embedding/subsetting the font.

aric:

> but surely they have a right to assert copyright ownership of the rest of the code responsible for embedding/subsetting the font.

Adobe did not embed the ripoff TT font ExpressSSK, but Adobe converted this TT font to a PS font. Adobe did not have a right to do that.

Look at the encrypted hex string of the capital S:

<1C60D8A8C9B6465286517DFB6D7C6C14A043374D961E136CA717E576CB9E
0ABEB37363E10081011882767160AB68D42427361B5BA3C4541EBE814357C975
F2D902D25AD05822832B08F28F4BC30F02C999C03E4B79B6A108B3D318DB316B
1FF7D11CBAB6E0F316E088BCD91128AEEE8037121C95EC6316063465BFF7E90D
ED5468A14E191112A80B7C128091249FB305C2E3CD8C3B9E1C0C9896F63695F0
977B0553958238285F6E7FB94FB7A4818712>

to which Adobe claims copyright (in the year 2009, nota bene).

This hex string decrypts to

0 659 hsbw 460 623 rmoveto 37 -27 18 -38 0 -50 rrcurveto 97 5 rlineto -9 75 -32 57 -55 38 rrcurveto -49 35 -64 18 -78 0 rrcurveto -69 0 -58 -17 -47 -33 rrcurveto -54 -38 -27 -51 0 -66 rrcurveto 0 -75 39 -53 79 -31 rrcurveto 72 -19 72 -19 73 -18 rrcurveto 78 -23 39 -38 0 -53 rrcurveto 0 -48 -18 -37 -36 -25 rrcurveto -31 -22 -41 -11 -50 0 rrcurveto -65 0 -50 14 -36 29 rrcurveto -39 31 -19 46 0 63 rrcurveto -102 -9 rlineto 0 -87 31 -65 63 -43 rrcurveto 54 -37 74 -19 93 0 rrcurveto 74 0 63 18 51 37 rrcurveto 57 41 29 55 0 72 rrcurveto 0 84 -39 60 -79 36 rrcurveto -1 1 -72 20 -144 38 rrcurveto -78 22 -39 35 0 50 rrcurveto 0 39 18 29 37 19 rrcurveto 28 15 35 7 41 0 rrcurveto 53 0 43 -12 33 -25 rrcurveto closepath endchar

This is the PS outline of the Akzidenz Grotesk "S" ripoff labeled ExpressSSK.

What makes you think that Adobe has the right to convert a TT ripoff font to a PS ripoff font? Do you think that the font forger Paul King granted to the Adobe CEO Shantanu Narayen in the year 2009 the right to convert his SSI (Southern Software Inc.) ripoff TrueType fonts to Adobe Inc. PostScript ripoff fonts?

As far as I can tell, Adobe would not be considered an end user in this case and as such would not be party to an end user license agreement. If you disagree, and if you can find a way to construe yourself as an interested party in this matter, get a lawyer and sue Adobe. Until the courts rule in your favor, most of us are going to assume that in cases where the EULA says we can embed/subset a font, the organization behind the EULA will allow us to use those fonts in PDF files, even if that means that the internal representation of those fonts changes from TT to T1.

aric:

You should work as the legal assistant of Adobe's CEO Shantanu Narayen, since you possess the great persuasiveness to make dimwitted people believe that Adobe is permitted by law to convert font forgeries from TT to PS and to make dimwitted people believe that Adobe becomes the copyright owner of converted font forgeries.

Uh, with web fonts there is a great need to convert OTF to TTF. Internet Explorer can’t handle OTF. This is a situation that will go on for years.

Oh, great! I'll be deploying all my fonts in OTF, then. The hell with Internet Explorer!

"Check, please."

hhp

Open Word, type the letter “A” three times and mark them up with the TTF font “Palatino”, and then select the character options “Outline”, “Shaded” and “Relief”. In this way, Microsoft Word generates derivative versions of the Palatino “A”. The result will look like this

No, no. You see, if you take a Type 1 font, and look at the vector information encoded in it, and then put those vectors in a TrueType font, then you're violating copyright law.

But if you just take a Type 1 font and print with it... well, that's what you're supposed to do with a font.

If you then take the printed image, and perform an "outline", "shaded" or "relief" transformation on the printed image, you aren't creating a derivative work of the Adobe Type 1 Font... you're merely creating a derivative work of your own printed document.

Now, what if you print the whole alphabet, and then shove it into an autotrace program, and make a TrueType font that way?

Since you have different vectors, and no hinting, you have filed off the serial numbers. And since type designs are, unfortunately, not protected by copyright, you haven't violated copyright law.

However, since the font was licensed, not sold, even to use it for printing, you needed to accept the terms of the EULA, and it may well be that doing this will violate the terms of the EULA.

Note, however, that if you print a book in this font, and someone else picks up a copy of that book, likes the font, and makes an autotrace version of it after scanning a page in... well, he violated your copyright. But he didn't "sign" the EULA. So rip-off copies of fonts are still impossible to stop with EULAs alone in the absence of legal protection for font designs.

quadibloc:

Your legal argumentations are entirely wrong:

- Palatino is a German typeface.
- Linotype is a German company.
- Hermann Zapf is a German designer.
- German law protects font designs.

While the Latin character set of Palatino is too old, the Greek character set contained in the 1998 Palatino version pala.ttf is probably a new design and hence probably protectable by German design law. If this should be the case and if the Greek character set should have been successfully registered as a new design, then "outline" etc. modifications of this Greek design made by Microsoft Word, as exemplified here

www.sanskritweb.net/forgers/kreter2.pdf

would be illegal according to German design law.

And the Postscript conversion of these outline modifications made by Adobe Acrobat in the PDF file kreter2.pdf would be illegal too.

Hence both Microsoft and Adobe would have committed an infringement of the German design law, unless they had been granted by Hermann Zapf the right to "to adapt, modify, alter, translate, convert, or otherwise change" the Greek font design of Palatino.

It is good that German law protects font designs, preventing the nonsensical situation I outlined in my post.

However, it is difficult for users and creators of Type 1 fonts to sue Adobe, and for users and creators of TrueType fonts to sue Apple (and, likely, its licensee Microsoft) because, of course, they had to license the font technology from those companies in the first place. This may place some limitations on what would otherwise be their rights.

Also, let us say that I have purchased a copy of a TrueType font, and am not licensed to convert it into an Adobe Type1 font. I use that font in a document which I am printing on a PostScript laser printer.

In doing so, I do not create a .pfb file that sits on my hard disk that I can use for printing documents with programs that can only use Type1 fonts.

However, it is true that the inner workings of the PostScript print driver take the vectors of my TrueType font, and send them off to the printer in a form that looks like the vectors of a Type1 font. Infringement?

No. That's because the "conversion" is used solely as a temporary means of printing, locked inside the programs, and not made available. If the font supplier were to use the EULA as a means to say "just because you bought this font, you aren't really licensed to print with it", they would be committing fraud, and the criminal law takes precedence over contract law (and, for that matter, even copyright law).

As for creating an outline or shaded version of Palatino's Greek character set, it may well be you do have a point, and it is not legal to use such a feature with some of the fonts one has on one's computer, depending on licensing, given that the font designs are protected by law - and the font design, of course, is still there after something has been printed.

Under U.S. law, though, except for trademark rights to the name of the font, most of the rights pretty well end at the time of printing - although the situation may have improved a bit recently.

quadibloc:

Your above posting is very interesting, but you mention additional legal problems which should be discussed in a separate thread.

However, one of the numerous Monotype vs. Adobe lawsuits, which fits this thread, should be mentioned here, namely this lawsuit

www.steptoe.com/publications/338b.pdf

It is funny that Judge Leinenweber, who tried to grasp the embedding bit, did not mentally realize that Adobe converts entire TT fonts to PS fonts thus dumping the embedding bits into the trash can.

Although I normally don't bother engaging with Uli, I'll make a brief exception.

Uli: "Adobe converts entire TT fonts to PS fonts thus dumping the embedding bits into the trash can."

Adobe: "Note, however, that it is possible to add fsType information to Type 1 fonts. See “Fonts that Use FSType Instead of fsType” on page 19 for more information"
http://www.adobe.com/devnet/acrobat/pdfs/FontPolicies.pdf

Cheers,

T

Mr. Phinney:

> See “Fonts that Use FSType Instead of fsType” on page 19

This paragraph on page 19 begins with the sentence:

"When TrueType and OpenType fonts are sent to Distiller, they are usually converted into one of the following PostScript representations: Type 1, Type 42, or CIDFont."

That is what I said at the beginning of this thread. But who allowed Adobe "to adapt, modify, alter, translate, convert" the fonts made by others?

Let's assume, I publish a new book, and then discover that Adobe "adapted, modified and translated" my book into another language. How come that Adobe is allowed to do that? Is Adobe above the law?

What law or laws do you suppose Adobe breaks in programming Distiller in such a way that TrueType and OpenType fonts are converted into a PostScript representation for embedding in a PDF file?

My understanding is that I cannot use any font I want in a PDF document, I can only use fonts that are licensed to be embedded that way. Adobe, however, is not responsible for my use of Adobe Distiller; when I "print" a word processor document out as a .PDF document, I am the one who is taking the action of embedding those fonts in the document.

Since Adobe does not provide tools for extracting embedded fonts from .PDF documents, but rather the fonts are embedded there as only an essential part of allowing such documents to serve as a form of electronic paper, they aren't engaging in contributory infringement either.

The law does draw a distinction between translations that are performed internally and externally. Thus, a microprocessor might convert the instructions of a computer program into a microcode-like form, and then even cache the instructions in this form within the chip to speed up tight loops by avoiding repeating the instruction decode step. Running a computer program on such a microprocessor does not constitute "translating" it into another language for copyright purposes, because it does not have the consequence of allowing people to use the program on other computers with a different instruction set.

Currently, there is a lawsuit going on between IBM and the maker of an Itanium-based machine that emulates IBM mainframes, because it uses just-in-time compilation that performs the translation externally - so this is a complicated area of law. On the other hand, Apple felt that they were unlikely to be sued by software vendors for their system for providing PowerPC emulation on Intel Macintosh computers using the same technology.

Embedded fonts in a PDF document aren't inside a chip, but since they are locked inside the document, font vendors may allow their fonts to be used for creating such documents. But you are correct that they're not obliged to allow this. But the onus of compiance is with individual users of Adobe products, not on Adobe - Acrobat Distiller has a legitimate and valuable purpose which has nothing to do with stealing fonts.

aric:

> What law or laws do you suppose Adobe breaks in programming Distiller in such a way that TrueType and OpenType fonts are converted into a PostScript representation for embedding in a PDF file?

This law:

US Copyright Law, section § 506 c (Fraudulent Copyright Notice)

Since you want to defend Adobe, you ignore the other Typophile thread:

see www.sanskritweb.net/forgers/interstate.pdf

You're right, I was forgetting that thread. In that thread, it was established that the supposed fraudulent copyright notices aren't even part of the PDF file; they are not inserted by Distiller at all, but by the printer driver into the "printed" PostScript code. They assert copyright over the code the driver has generated, not over any other intellectual property in the file, whether font-related or not.

It seems that no one can persuade you to believe that your interpretation of the situation is flawed. It also seems that you cannot persuade anyone that your interpretation is correct. Sounds like a stalemate to me; I'm throwing in the towel.

On a completely unrelated note, I wish you a very happy new year.

> On a completely unrelated note, I wish you a very happy new year.

I wish the same to you.

The documentation

www.sanskritweb.net/forgers/interstate.pdf

has been greatly enlarged and now also includes legal remarks for Germans on page 89 of this document.

quadibloc:

> Under U.S. law, though, except for trademark rights to the name of the font, most of the rights pretty well end at the time of printing

UK is different. In UK, there is a provision for the printing of books etc.

The "Copyright, Designs and Patents Act 1988", which is the current UK copyright law, specifies in section 15 for "typographical arrangement":

"Duration of copyright in typographical arrangement of published editions
Copyright in the typographical arrangement of a published edition expires at the end of the period of 25 years from the end of the calendar year in which the edition was first published."

This funny provision even applies to newly typeset public domain books, e.g. the works of Shakespeare, and even to PD books newly typeset in PD fonts, e.g. books typeset in an old Garamond font, with the consequence that you are not allowed to reprint an old public domain book newly typeset in an old public domain font.