That sounds very damaging. This is *exactly* the sort of information that leads people to think some things are ok.

Wow, this sounds like a declaration of war to me. Apple declares war on type designers and font foundries. Who would've thought?

Perhaps if enough people complain, they will change their website copy. Just like that whole wrong-quotation-marks thing in the New York Times anti-Vista ad…

Si, any chance that you (or other attendees) could publish an online re-cap of the Business of Type event? I'm not asking for full presentation decks or discussion minutes or anything; I know that some of that must be private. But just an event summary… like you commonly see on blogs after TypeCon or the ATypI conference?

It isn’t Apple’s job to educate designers about font licensing. It would certainly be good of them to point it out, but nobody should expect them to do it. If the foundries aren’t licensing their fonts for web embedding, it’s their responsibility to be very upfront about it. This has been coming for years, foundries really should have started doing more than just tucking a “no embedding” provision into EULAs, which most people never bother to read. Put the terms up there in big red letters that people will actually see before they click “buy.”

It's not Apple's job to educate, but when they make misleading and arguably false claims like "limitless choices" (from the press release) and "use any font they want" (from the web site), then there's a problem.

I pinged Apple on Saturday and followed up this morning as well. Maybe I'm an optimist, but I'd expect the text will be changed some time this week. As I said on Friday in our panel discussion on Web fonts, this is the sort of thing that can easily happen at any big company when the PR/marketing folks go off and write something without quite enough careful vetting by the technical folks. Not that I'd know that myself, nosirreeBob.

Cheers,

T

>It isn’t Apple’s job to educate designers about font licensing.

Sorry, but I disagree. There are various options re. developer education around technologies that redistribute fonts. I believe that it absolutely is the responsibility of people who make font-exploiting technology to help developers and users stay legal. Here's a recent example from Microsoft - http://windowsclient.net/wpf/starter-kits/sce/sce-howto-using-fonts.aspx

I can understand why a Apple might want to stay silent, and that might have been the safest thing for them to do legally. But in this case they are giving plain bad advice to web page designers, and that needs to be fixed.

>Si, any chance that you (or other attendees) could publish an online re-cap of the Business of Type event?

We're working on that.

>Perhaps if enough people complain, they will change their website copy. Just like that whole wrong-quotation-marks thing in the New York Times anti-Vista ad…

I think this is a good test to see how seriously they take this problem. The ad in question was fixed within 24hrs, correct? And that involved editing a multipart Flash animation. This should be a quick fix. Clock is ticking.

"I find it hard to understand why they’re telling web designers that they can post any font to the Web."

It's called 'marketing'. Marketing usually has little to nothing to do with the product's inherit abilities or lack thereof.

"I believe that it absolutely is the responsibility of people who make font-exploiting technology to help developers and users stay legal."

IANAL, but I think this would be really bad precedence. Just because you CAN copy a book on a photocopier, should the photocopier manufacturer be liable for you infringing copyright?

It'd be NICE if apple's marketing folks ran this kind of stuff through their developers and added important details like we're discussing here, but I can't say Apple HAS to do that on any legal grounds.

"But in this case they are giving plain bad advice to web page designers"

Eh...any competent web designer isn't going to believe anything on a marketing web site anyways. Mom the scrapbooker might believe it, though. So, yea, some concern there.

> Just because you CAN copy a book on a photocopier, should the photocopier manufacturer be liable for you infringing copyright?

Don't newer scanners and printers prohibit you from copying and printing money?

I've seen plenty of photocopiers that have warning notices on them, and I've seen plenty of photocopiers with no warning (or one buried in the manual). What I've not seen is a photocopier with a notice from Xerox saying - "go ahead use me to photocopy anything you like, it's all good."

>should the photocopier manufacturer be liable for you infringing copyright?

I'm not saying the're any liability if you stay silent, what I'm saying is that it's good customer service to give your customers a heads-up.

If web designers can post "any font" for use with Safari, what exactly makes this a 'marketing' issue?

Cheers!

"Don’t newer scanners and printers prohibit you from copying and printing money?"

They try. But that's a counterfeiting issue rather than IP infringement.

"what I’m saying is that it’s good customer service to give your customers a heads-up"

No argument there. Let me know if you ever find a software company that does that. ;o)

"what exactly makes this a ’marketing’ issue"

The text on Apple's web site (aside from the support section) is, in all likelihood, owned by marketing.

Saying you can upload any font for use with safari is true, just greatly overstating the feature both technically and legally. But that's what marketing does. The fine print/asterisks/caveats/gotchas are reserved for the back pages of the documentation or the knowledge-base articles, or some blog post somewhere in the future...

>Let me know if you ever find a software company that does that.

;-) You are too funny!

I would guess the amount of people using Safari is limited and the damage is probably limited if they take action.

Don’t newer scanners and printers prohibit you from copying and printing money?

Not the hardware, but some scanning software and Photoshop will not scan US money properly. Most color laser printers sold in the states tag every document with unnoticeable serial numbers. But doing so with fonts means bringing back font DRM, which some important entities are openly against.

The purpose of my comment was not to elevate Apple’s actions above criticism. I just wanted to point out that this will not be the first time this happens, that the type industry should expect these incidents to happen, and foundries need to start being proactive and creative about licensing and educating users before it becomes a problem—i.e. before Firefox or IE jump on board. Some people are already doing so—creative license models, Adobe working with the AIGA on student fonts, Microsoft’s event hosting and of course, all the industry types talking about it here and elsewhere. But if you guys slip, you could end up in a nasty hole like the record industry, and that’s not something anyone in the industry really wants, is it?

Hm, I'd like to play the devil's advocate for a moment ... if the text read "With NCSA Mosaic, web designers can go beyond text and use any image they like to create stunning web sites ...", would you want the Web back without images? Or would anybody come to believe from that sentence that no legal restrictions would apply? That would be just naive.

Okay, granted, there are too many naive people doing so-called web design. It all boils down to that people must be educated to respect intellectual property more. (I know, it's easy to say.)

"...Not that I’d know that myself, nosirreeBob."

The corporate world has its own axis :-)

ChrisL

Hm, I’d like to play the devil’s advocate for a moment ... if the text read “With NCSA Mosaic, web designers can go beyond text and use any image they like to create stunning web sites ...”, would you want the Web back without images?

Mr. Kutilek hits the nail on the head:

While fonts are not copyrightable in most countries (although Typophiles wish they were), images or pictures ("pictorial works" as per US Copyright Law) are copyrightable in most countries, but many Typophiles do not hesitate to upload copyrighted pictures to Typophile threads without having obtained the permission to do so from the picture copyright owners. The Typophile website threads abound with pictures, for which the publication rights have not been acquired.

Look, for instance, at the start of this thread: Even this tiny picture (screen shot) inserted by Mr. Daniels probably is copyrightable, although this might be disputed at court, but by browsing through other Typophile threads you certainly will find many copyrighted pictures, for which the uploaders never acquired the picture publication rights.

If you download the pictures made by others, "as they're are needed", and upload them without permission to your website, why do you complain that Apple Inc. intends to do the same with your fonts by downloading them without permission, "as they're needed" by Safari?

Interesting point, Uli.

The thing is, a tiny 72 dpi image does not have scalability and usability as a piece of softwarre. I would not have a problem seeing images of type at 72dpi on web sites but I would objectto the useable font being extracted by the browser and made available as free software to pirates as well as naive users who would assume anything on the web is free. Think of Si's image as a small quote of text as opposed to stealing an entire novel and selling it as your own.

ChrisL

Uli is back! Wooo!

Wooo!

I agree with Chris (Apple lawyers please don’t come after me) ;-) there’s a difference between the hi-res © photograph and a 60px x 60px picture of a Seahawk helmet (at least I hope so). In the case of the Safari scheme we’re talking complete, unobfuscated, unsubsetted source fonts. However, it’s been my experience that drawing analogies between fonts and other forms of IP, be that pictures, music, books, etc., almost always break down.

Note Apple said you CAN, as in, you are able to technically. Apple didn't say you MAY use any font. Mincing words yes, but that's what marketing is about. Web designers can post any font. That doesn't mean they may. I can use any picture I want on my website, of course as well *

Yes, it's bad advertising that I agree should be changed, but it's not encouraging people to violate their licenses either.

* Provided I obtain proper permission.

«El futuro es una línea tan fina que apenas nos damos cuenta de pintarla nosotros mismos». (La Luz Oscura, por Javier Guerrero)

In some cases, because we’re having discussions about the images or their subjects, and the images are being used in part or at small scale, and for no profit, use of said images constitutes fair use. At least in does in the USA, where Typophile is hosted and maintained. Redistributing an entire font, especially when done to make a profit, is very different.

>* Provided I obtain proper permission.

I think adding "(license permitting)" after "any font" would be a good quick fix.

Uli has a point, but it's not really about 'lo-res' vs. 'hi-res'. It's more of an issue of fair use and/or common sense.

In the Netherlands (and I guess in most of the world) people are supposed to know the law. In other words: everyone has to find out if there are rules regarding the things they do. May it be driving a car, selling fruit, designing a website — you need to know what’s forbidden, restricted, allowed, etcetera.

Same goes for using some typeface on your site with the technology Apple provides. If the law, the typedesigner or the typeseller forbids this, you just can’t do it. Even when Apple suggests you could. But then, Apple also suggests you can make a video masterpiece using Final Cut. : )

. . .
Bert Vanderveen BNO

The headline for this thread is dynamically created from a commercial font using Flash and Javascript.

Why is Safari making this exact same thing possible with CSS wrong and illegal if doing it with Flash and Javascript is not?

Yeah I agree with Bert. Sure there's maybe like some soccer mom in Minnesota who thinks all fonts on the internet are free... but the majority of professional designers know the general laws surrounding copyright. They just choose to ignore them.

I think many people are missing a bigger issue here which is that Safari is only implementing something that is already a published web standard from the Worldwide Web Consortium. Whatever the marketing hyperbole from Apple, web fonts will eventually be a feature of every browser and designers and font foundries need to recognise this.

>The thing is, a tiny 72 dpi image does not have scalability and usability as a piece of software...

Ah.. But that all depends on the person, his abilities and the useage. I know ways to take 72dpi images from the net, tweek and rez up them in photoshop and use them on the side of Buses.

But again as was stated before by Aluminum about Uli's point is that it's more about fair use and/or common sense.

I think part of the problem lies in how easy it is to pass a font from place to place with out retribution. Where a copyrighted image when the same is done.. the creator pushes much harder to bring retribution for it's mis-use.

P.S. Pardon my grammer I'm more of a visual person.

Who sez you can't scan money?

http://gethelveticaoffourmoney.com/

"While fonts are not copyrightable in most countries"

They certainly are in the USA, which is the country Apple is most primarily operating in. The fonts in question are the actual digital font files. As far as I know, most countries recognize digital fonts as software and as therefore subject to copyright as software. Certainly the USA does.

Cheers,

T

>The headline for this thread is dynamically created from a commercial font using Flash and Javascript.
>Why is Safari making this exact same thing possible with CSS wrong and illegal if doing it with Flash and Javascript is not?

You’re missing the point, it’s not "wrong" or "illegal" to post "a font" to a web site – the point of this thread is that Apple’s promotional material suggests that Web designers can post “any font”, when the reality is that no known commercial fonts can be used under the scheme. If Adobe’s material said the same thing about Flash embedding it would be wrong, just not as wrong as Apple’s statement as many commercial font licenses allow for Flash embedding.

>I think many people are missing a bigger issue here which is that Safari is only implementing something that is already a published web standard from the Worldwide Web Consortium.

Actually been discussing the “bigger issue” on Typophile since the issue resurfaced in 2006. I started this thread to specifically address the Apple positioning of the technology.

“While fonts are not copyrightable in most countries...” (Ulrich Stiehl)

"They certainly are in the USA, which is the country Apple is most primarily operating in. The fonts in question are the actual digital font files. As far as I know, most countries recognize digital fonts as software and as therefore subject to copyright as software. Certainly the USA does." (Thomas Phinney)

Mr. Phinney:

This is the famous fairy tale you have been propagating for many years.

Let us assume, Germany belongs to your "most countries"
("As far as I [Thomas Phinney] know").

If so, then quote the German law and/or the German Supreme Court decision ruling that fonts are copyrightable. You will not be able to do so with regard to Germany, because your story of "copyrightable fonts" is a fairy tale.

And you will not be able to do this with regard to France, Belgium, Italy, Spain, Portugal, etc. etc. etc. etc. etc. etc., because your story of "copyrightable fonts" is a fairy tale. Only little children think that fairy tales become true by telling them over and over again.

"Certainly the USA does"

I do not know an American Copyright Law nor do I know an American Supreme Court decision ruling that fonts are copyrightable.

You, who you as an American are claiming that this is so, should at least be able to specify the section of the American Copyright Law ruling that fonts are copyrightable and/or to specify the American Supreme Court decision ruling that fonts are copyrightable. But even with regard to your own country, you cannot specify this to make your fairy tale become true.

PS: The design of "new and original" fonts may be registered in certain countries, e.g in Germany, on the basis of a design law ("GeschmMG" in Germany), but this has nothing to do with copyright laws.

the point of this thread is that Apple’s promotional material suggests that Web designers can post “any font”, when the reality is that no known commercial fonts can be used under the scheme.
Again, the key word is can. You CAN post any known commercial font, provided it's not in any odd known format. Whether or not you MAY is another question entirely.

Teacher, can I go to the bathroom?
— Well, I don't know, can you?

Teacher, may I go to the bathroom?
—Yes, you may.

Again, this is not to say that I don't think it's a bit misleading, but still, we can't fault them for being inaccurate.

«El futuro es una línea tan fina que apenas nos damos cuenta de pintarla nosotros mismos». (La Luz Oscura, por Javier Guerrero)

Don't you see that new fonts will get exposed to new customers? It shouldn't even be illegal for copying and font foundries can still web-embed their fonts.

I have to second Jens and guifa here.

They say you can use any font. And yes, you technically can. Whether you legally are allowed to or not is a completely other question and its up to you to check that.

They merely claim to provide the technology, so that you can, what they actually do.

On the other hand, Apple has its HQ in the US. And over there they have a strange interpretation of law (why would a microwave producer have to put a warning sign "not for drying pets" onto it, to avoid ligitation? -- actually, is that an urban legend or not? Even if so: very telling).

So it may be a concern for _their_ legal department not to claim such a possibility of using any font.

Something which doesn't seem to be an issue in the discussion so far is safety.

It is possible to build fonts which crash rasterisers. I can think of a couple opportunities in a .ttf or .otf which can cause serious problems in the app and the OS. Fonts are part of a text engine, usually integrated deeply into the OS. Fonts are subject to ambiguous standards and can contain many values which catch the rasteriser unaware. Integrity tests for fonts usually catch some of the problems, but never all of them. (Aside, it seems the PostScript Type 1 format is great for non-downloadable fonts). So far, bad fonts were always dealt with by the person who installed them. Even when something goes wrong, the user will usually know what happened and how to solve it.

In the context of a website there is no sanity check on the server end, it is up to the browser to make sure the font is fit. The user will be unaware of any potential problems in a webpage which is about to open.

Speculating, if the fonts-for-download is part of WebKit, it won't have its own rasteriser and the Doomsday.ttf might be possible. Speculating even more, Mail.app uses webkit, it would be possible to make an html mail message which links to malevolent fonts. It might even work on the iPhone, depending on how much code is shared between the iPhone and MacOSX.

Perhaps Apple has separated the rasteriser in Safari from the rest of the OS. Perhaps Safari also checks the integrity of the font after download. In that case a bad font would just not load. But if a bad font isn't caught, and with a rasteriser in Safari, it could crash the app, but not the OS. In the worst possible case, if Safari uses the OS rasteriser, it could trigger a series of crashes which eventually take the whole OS down. And that is quite unacceptable.

At the very least, the downloading of fonts should be covered in the security prefs of Safari, and off by default. A bad font can do more damage than a run away javascript. Even Java and Flash don't have access to the core OS functions as fonts do. But these do have their own controls in the security prefs.

And that's just Safari. Will Firefox be able to follow to same safety precautions Safari might have?

Of course, all of these problems can be fixed in rasterisers and font handling code. But given that rasterisers, after almost 20 years of improvements, still contain problems, makes me wonder how fast browser and rasteriser builders can respond.

Just my 2¢ of gloom.

Dez:

"The thing is, a tiny 72 dpi image does not have scalability and usability as a piece of softwarre. I would not have a problem seeing images of type at 72dpi on web sites but..."

Sii:

"I agree with Chris (Apple lawyers please don’t come after me) ;-) there’s a difference between the hi-res © photograph and a 60px x 60px picture of a Seahawk helmet..."

Although it is awfully boring for me to explain again and again the most simple notions of copyright law, I will try to explain them again. For this purpose I refer to a copyrighed picture in this new Typophile thread

http://www.typophile.com/node/43945

The "Ali boxing poster" shown in this thread is indubitably a "work of art". The date 1975 proves that this poster was made roughly 30 years ago. If the poster was made in the USA, then it will be protected by copyright for a period of 70 years post mortem auctoris (after the authors's death). So nobody except the author (here the designer or painter) or his/her assignee (e.g. the company which acquired the picture rights) is allowed to publish this picture.

It does not matter, whether this poster is photographed or scanned or reproduced by whatever technical means, and it does not matter, whether this reproduction is done in high or low resolution: In all these cases, you have to acquire the publication rights from the author or his/her assignees.

Even if you want to reproduce this poster in a very tiny size and in very low resolution, e.g. as a postal stamp, you will have to obtain the right to do so.

Therefore, the publication of this Ali poster as a scanned reproduction in low resolution at the Typophile thread

http://www.typophile.com/node/43945

constitutes a copyright infringement, unless he who uploaded this picture has acquired the right to reproduce this poster on the Typophile website.

Let us assume this legal information just given by me were not correct, and everybody were allowed to upload scans and photos of any pictures to any websites, then for example, everyone were allowed to scan all pictures painted by Picasso and upload them to whatever internet sites. This of course is definitely illegal and would constitute an infringement of copyright, in the very same way, as the publication of the Ali poster constitutes an infringement of copyright.

End of tutorial. I need some fresh air now.

click around on these two sites ....

good reading on Fair Use

Chilling Effects

and read this

Fair Use on the Internet (pdf)

i haven't gotten through all of it, but it seems that Typophile's use of images would constitute Fair Use under most reasonable interpretations. If it wasn't 4 a.m., I'd make a more detailed post.

also interesting is the notion of an Implied License when you post something on the internet.

Whenever an author posts anything to the Internet, he or she should reasonably expect that it will be read, downloaded, printed out, and perhaps forwarded on a limited basis. The author impliedly grants a limited license to users to utilize the work in this manner.

It doesn’t matter whether fonts are copyrightable or not. That only affects your ability to sue people who have been given copies illegally by someone who owns a legal copy. It doesn’t change the fact that fonts are only distributed subject to a licence agreement; if you don’t comply with the agreement, you could be sued for breach of contract, even if in your country fonts are not subject to copyright.

In a sense, therefore, the issue of copyright here is a red herring. The real issue is that fonts are licensed, and that the licence agreement (which is a contract between two parties, and applies regardless of copyright law) would in many cases be breached by distributing a font file in the manner presently supported by Safari. And since many users don’t bother reading or following licence agreements, the likelihood of a breach is high. Hence the complaints here about Apple’s marketing of Safari’s new feature.

Personally I would like to see support for subsetting and maybe a new file format for web fonts that can't be easily installed by end users. This has been tried in the past (WEFT, for instance), and I think it's the only way that those of us who would like to be able to embed fonts on the web are going to be able to convince the commercial type foundries to permit this type of usage. It also has the advantage of not requiring every visitor to your site to download large font files when all they needed was the letters ‘a’ through ‘z’. But it needs to be standardised across platforms and browsers, or nobody will want to use it.

--
http://alastairs-place.net

Erik

I would hope that Safari is running its own text renderer (even if just a separate instance in a sandbox of sorts) because it is lacking many text rendering features standard in CoreText. Neither required ligatures or kerning render, both of which feature in any standard NSTextField that you can do in Interface Builder in a dummy app (one that just has a single NIB file and no code).

I'm not sure why Apple turned off those features, but I do know they (oddly) have inconsistent font technology between their programs. Why would it be that TextEdit (which uses the available-to-anyone NSTextView) have a more advanced engine that Safari or even Pages? My only guess is that they don't expect very large documents in a program that would use NSTextView, or that it's not as highly optimized as the engines used in Safari or Pages where speed is more important, especially for long HTML pages or doctoral dissertations, and they've cut out features or done other optimizations.

One of the main features I've noticed (and irritated me to know end with some of my term papers) is that Pages does not support combining diacritics very well (needed for putting the tilde on a lc Q).

«El futuro es una línea tan fina que apenas nos damos cuenta de pintarla nosotros mismos». (La Luz Oscura, por Javier Guerrero)

i'm not sure anyone else noticed this, but Safari now runs on Win XP and Vista, so this issue is not limited to just Mac users.

I think Erik's point related to font rasterizers (converting a vector glyph outline to a rasterized image of the requested glyph), not text layout (where features like contextual substitution are done), which are normally separate issues handled by different applications.

It wouldn't surprise me if the Safari browser (for the desktop) uses it's own rasterizer; on Windows it uses it's own rasterizer which produces results similar to the Mac OS. But I am not sure that would be true on the iPhone, where there could be advantages to sharing the system's rasterizer, and the architects may have seen this as a safer environment than the desktop (which would probably be a bad idea!)

Although it may seem like it's a stretch that a font could crash a system or application, it is entirely possible, and I'd guess that this aspect of "font downloading/embedding" has not been sufficiently tested.

Jason C

Re. security (off-topic I know) - Safari's lengthy legal page includes a reference to Freetype (ironic given the patent tussles between Apple and Freetype but that's another story), the Open Source rasterizer, so I imagine they're using it. Freetype is mentioned several times in this Linux Watch article - http://www.linux-watch.com/news/NS5622797384.html - about the importance of keeping components up to date. So the next time the Apple pop-up ad appears asking you if you want to update Safari, iTunes and ice water etc., you better click 'yes'.

The issue raised at the head of this thread seems to be less about copyright protection and law than the responsibility of a company to keep its public statements in line with protected practice. Copyrights exist to protect an author's economic interest, and the basic test of infringement is whether or not the original owner of the copyright has suffered economic loss (actual or real potential) from the unauthorized use of the item for which the copyright was recorded.
web designers can go beyond web safe fonts and use any font they want…
A true statement, as long as the requisite licensing fees have been paid or grants of use received. Like most spin meisters the Apple flacks don't detail the costs. How many politicians this season have detailed what costs their programs would entail?

Uli,

As usual, you resort to insults.

It is a simple fact that the US Copyright Office accepts the copyright registrations of digital fonts, and does so following an explicit policy decision made in 1992 (quoted below in the court judgment excerpt).

To the best of my knowledge, you are correct that the US Supreme Court has not ruled on the applicability of copyright to digital fonts. That's because they haven't needed to, because no case has been appealed all the way to the US Supreme Court. Things don't need to be ruled on by the Supreme Court to be law in the USA.

However, Judge Whyte's 1998 summary judgment in the Adobe vs SSi case in Federal District Court did rule on this issue, and upheld the principle that digital fonts are subject to copyright. http://directory.serifmagazine.com/Ethics_and_Law/Copyright/judgement.php4

The relevant section of the judgment follows:

* * * * *

In a 1988 Copyright Office Policy Decision, the Copyright Office determined that digitized typefaces were not copyrightable because they were not computer programs and required little selection or arrangement beyond that dictated by the uncopyrightable typeface design. "Policy Decision on the Copyrightability of Digitized Typefaces, " 53 Fed. Reg. 38 1 lo-381 13 (September 29, 1988).

However, in 1992 the Copyright Office issued a final regulation regarding the registrability of "computer programs that generate typefaces" which appears to back off the 1988 policy decision, although the parties dispute this. The 1992 Regulation states:

After a careful review of the testimony and the written comments, the Copyright Office is persuaded that creating scalable typefonts using already digitized typeface represents a significant change in the industry since our previous Policy Decision. We are also persuaded that computer programs designed for generating typeface in conjunction with low resolution and other printing devices may involve original computer instructions entitled to protection under the Copyright Act. For example, 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 unprotectible shape of the letters nor functionally mandated. This expression, assuming it meets the usual standard of authorship, is thus registrable as a computer program.

"Registrability of Computer Programs that Generate Typefaces," 57 Fed. Reg. 35 at pp.6201-2 (February 12, 1992)). Plaintiff contends that the 1992 Regulation supports protection for the scalable typefonts at issue. Defendants contend that the 1992 Regulation reaffirms the 1988 policy decision stating that digitized typefaces are not protectable.

Defendants base their assertion on the fact that the 1992 Regulation specifically states that it "does not represent a substantive change in the rights of copyright claimants." Id. at 6202. They argue that the effect of the 1992 Regulation is merely to revise the procedure in referencing font-depicting data in the registration process. Thus, defendants contend that the numerical reference points that define an outline of a glyph are unprotectable as a matter of law.

Defendants argue that after one has filtered out the unprotectabie elements of plaintiffs software in order to compare what remains, one finds that minimal, if any, protectable expression remains. Thus, defendants state that "merely manipulating an unprotectable font image to create another, slightly different (but still unprotectable) font image cannot possibly give rise to protectable expression . . . ." Defs. Memo., pp. 15-16. Defendants assert that no matter what points are selected by the Adobe editor performing the process, they correspond directly to, and hence are determined by, the unprotectable font shape. Therefore, because the output is not protected and there cannot be any creativity in what the editor does to obtain the output, nothing is protectable.

Adobe contends that King copied literal expression.7 Adobe contends that while the shape of the glyph necessarily dictates some of the points to be chosen to create the glyph, it does not determine all the points to be chosen. Thus, each rendering of a specific glyph requires choices by the editor as to what points to select and where to place those points. Accordingly, Adobe asserts that the selection of points and the placement of those points are expression which is copyrightable in an original font output program. The actual code is dictated by the selected points.

Adobe relies on Feist Publications v. Rural Telephone Serv. Co., 499 U.S. 340 (1991) and argues that because the glyph coordinates are not prescribed entirely by the shape of the glyph they involve creativity and are protectable. "To qualify for copyright protection, a work must be original to the author. [citation omitted]. Original, as the term is used in copyright, means only that the work was independently created by the author . . . and that it possesses at least some minimal degree of creativity." . Id at 345. Further, "the requisite level of creativity is extremely low; even a slight amount will suffice. The vast majority of works make the grade quite easily, as they possess some creative spark, no matter how crude, humble or obvious it might be." Id. Thus, Adobe argues that because there is some creative choice by the editors who select the on-curve and off-curve reference points, Adobe's font software programs are protectable.

Defendants contend that FontMonger and Fontographer go beyond the code and extract the coordinates and express the coordinates in absolute values. Thus, defendants assert that the source code is not being copied but only the coordinates or points which are not protectable. Defendants then assert that FontMonger and Fontographer create code based on the selection of the points which is necessarily dictated by the shape of the glyph. Defendants contend that because they only extracted the points, they have not infringed because the points themselves are not protectable.

The evidence presented shows that there is some creativity in designing the font software programs. While the glyph dictates to a certain extent what points the editor must choose, it does not dictate every point that must be chosen. Adobe has shown that font editors make creative choices as to what points to select based on the image in front of them on the computer screen. The code is determined directly from the selection of the points. Thus, any copying of the points is copying of literal expression, that is, in essence, copying of the computer code itself.

Further, the selection of points is not dictated by functional concerns only. See 57 Fed. Reg. 6201-2. Defendants argue that efficiency is the key which is driven by the goal of minimizing the number of reference points. However, simply because there are several ways to create the same glyph, some being more efficient than others, i.e. using fewer points, does not mean there is no creativity in the process of creating the software to produce the glyphs. That some creativity is involved is illustrated by the fact that two independently working programmers using the same data and same tools can produce an indistinguishable output but will have few points in common. Accordingly, the court finds that the Adobe font software programs are protectable original works of authorship.

* * * * *

After this, I think I need to revert back to my principle of not responding to Uli, which I let lapse after nearly a year.

Bye,

T

I love an Apple :)

-------- http://www.theapplemuseum.com

Just get rid of the "any" - that will do.
Could you add "any of your fonts" instead?

André

"Could you add “any of your fonts” instead?"

Perhaps "any of your licensed fonts which permits it"

ChrisL

i'm fairly sure that posting reduced jpegs to a discussion group is protected as fair use, but those days may be numbered. even fair use seems to be in jeopardy. check this out; it sends chills :

http://williampatry.blogspot.com/2008/04/fair-use-three-step-test-and-eu...

i'm not arguing against your livlihood, folks, and anyway that's a different thread.
but as usage restrictions become more draconian and unreasonable it seems more important than ever to set precedents for open sharing where possible.

Here are two short answers to Mr. Phinney's reply:

1)

Mr. Phinney's statement "most countries recognize digital fonts as software" was a fairy tale and will remain a fairy tale, because Mr. Phinney could not prove that "most countries" recognize fonts as software.

2)

I predicted elsewhere that Mr. Phinney would again quote this old hat, the SSI "preliminary judgement" alias "summary judgement" (in Germany "einstweilige Verfügung").

Everyone knows that "preliminary judgements" are "preliminary" and hence cannot be generalized, because no trial took place. Furthermore, Mr. Phinney did not disclose the most "relevant section", namely this one:

"Based on the foregoing, the court grants Adobe's motion for summary judgment on its claim of copyright infringement of its Utopia software program, denies defendants' cross-motion on the copyright claims, denies Adohe's motion for summary judgment on the patent claims, denies defendants' countermotion on the patent claims and denies Ares' motion for summary judgment on the third-party complaint."

This means: Firstly, the preliminary judgment refers to the font Utopia only, to no other fonts sold by Adobe. Secondly, the preliminary judgment denied Adobe's motion. In my document about this preliminary judgement

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

I stated:

"Adobe did not dare to seek a trial due to the fact that Adobe’s motions concerning Slimbach’s patent claims were denied by the district court."

I do not know an American Copyright Law nor do I know an American Supreme Court decision ruling that fonts are copyrightable.

Nor, apparently, do you know US law. There is no need for there to be a US Supreme Court ruling for something to qualify as legal precedence (as a minor addition/correction to Thomas's statement about the US Supreme Court not needing to rule on something to make it law, well, no court can make anything law in the US, they only interpret extant law). As well, there is no such thing as a "preliminary judgement" in American law, so I'm not sure what you mean by that but based on a quick question to a German friend (note I explicitly don't claim to know the German system and am only speculating), I think you've confused summary judgement with preliminary injunction, which are very different (and this confirmed by the BEOLINGUS dictionary). A summary judgement is only preliminary in that it takes place before the primary case, to rule on whether or not claims are triable, and my guess in German would be "beschleunigtes Urteilsverfahren". Courts can still create precedence in these judgements (in the US), as evidence is presented just as they would be in a normal case.

SSI would want the court to rule that copyrightable did not extend to the font software thus avoiding the entire court case. The court stated that they are protectable, and thus Adobe can continue in their case against SSI, which would then involve proving that SSI did indeed violate the copyright. If SSI had succeeded, there would have been no need for Adobe to attempt to show how their software was copied, because the court would have ruled that it wasn't copyrightable. This is the point of a summary judgement in US law. Regarding the copyrightability and generalization of the ruling:

"The parties do not dispute that defendant King had access to plaintiffs programs.5 Thus, the primary issue in this case is whether the material at issue is protectable expression under the Copyright Act"

"Thus, defendants contend that the numerical reference points that define an outline of a glyph are unprotectable as a matter of law."

"Accordingly, the court finds that the Adobe font software programs are protectable original works of authorship."

This is indeed a ruling that can be generalized. Otherwise the court would have stated "Accordingly, the court finds that Utopia is a protectable original work of authorship." But it didn't, leaving the judgement open for reference in other cases.

«El futuro es una línea tan fina que apenas nos damos cuenta de pintarla nosotros mismos». (La Luz Oscura, por Javier Guerrero)

Thanks ChrisL: “any of your licensed fonts which permits it” will do just great!

Uli is right about the images. We take images from other websites all the time. Do I profit from it? No. But it is analogous to using fonts without proper licensing.

Mikey :-)

guifa:

A summary judgement is only preliminary in that it takes place before the primary case, to rule on whether or not claims are triable, and my guess in German would be “beschleunigtes Urteilsverfahren”.

Such terminological discussions do not make much sense, because English is not my mother tongue and German is not your mother tongue. The term “beschleunigtes Urteilsverfahren” is a criminal proceeding in Germany, and the term "preliminary judgement" is used by me and others as another expression or as a synonym for "summary judgement" in the USA.

Instead of terminological discussions, you should look at the reality which will open your eyes so that you will recognize that the SSI "summary judgement" does not prove that the Adobe fonts are copyrightable in the USA or elsewhere.

For instance, Paul King has been selling the SSI fonts alias Adobe fonts many years after the lawsuit. And these SSI fonts alias Adobe fonts are even today still downloadable from numerous websites. (Miss Tiffany does not like links to such sites, so that you will have to browse for these sites for yourself.)

All this would not have been accepted by Adobe, if the Adobe fonts alias SSI fonts were copyrightable.

What is more, Adobe's entire Font Folio collection (the unmodified fonts, not the SSI clones) is downloadable for free from numerous websites. This would not be accepted by Adobe, if the Adobe Font Folio fonts were copyrightable in the USA or elsewhere.

And as regards font CDs, Adobe's entire Font Folio collection can be bought for a few cents as commercial ripoff CDs. I documented this at my website

http://www.sanskritweb.net/forgers/index.html#OSTEUROPA

All this would not be accepted by Adobe, if the Adobe Font Folio fonts were copyrightable in the USA or elsewhere. The SSI "summary judgement" alias "preliminary judgement" is entirely useless and proves nothing.

Uli: This means: Firstly, the preliminary judgment refers to the font Utopia only, to no other fonts sold by Adobe. Secondly, the preliminary judgment denied Adobe’s motion.

Utopia was used as a test case in the Adobe vs. SSi lawsuit, and was accepted and understood as such by both parties. Of course Judge Whyte's acceptance of Adobe's motion for a summary judgement on the copyright of the Adobe font applies to the other fonts in the case: it was acceptance of the copyrightability of fonts (at which point SSi quickly agreed to settle out of court: they understood that the judgement applied to all the fonts, not just Utopia, even if you dont').

What Judge Whyte denied was the motion for a summary judgement on the specific patents claimed by Adobe. So he accepted the motion regarding the general issue of font copyright, which he judged in favour of, but he denied the motion regarding the patents. Note that this does not mean that he denied the patents, but that he denied to motion to make a summary judgement on them: the patent issue would have to be decided in court. The copyright issue could be decided by summary judgement, since the US Copyright Office had already established in principle that fonts could be copyrighted as computer programs. What the judgement needed to establish was whether the fonts, and particularly the outlines, met the requisite legal criteria of creativity to be protectable by copyright. And Judge Whyte decided that yes, they did, and that summary judgement is now legal precedence. I don't know what the role of precedence is in German law, but in in US law, derived from English common law, is is extremely important.

Your later argument that infringing fonts are still available through various sources is not evidence that fonts are copyrightable. Unlike trademarks, which one is obliged to defend or lose, one is under no obligation to defend one's copyright, and the copyright is retained regardless of whether the holder actively prosecutes infringement. I happen to know something of the material cost in time and energy that the SSi lawsuit put on Adobe's type group, that it delayed the development of new products and tied up key people for more than a year. Frankly, I'm not surprised that Adobe are not launching lawsuits at every infringement of their copyright. Like reasonable people, they pick their fights.

...and the term “preliminary judgement” is used by me and others as another expression or as a synonym for “summary judgement” in the USA.

It is not a synonym. There is nothing 'preliminary' about a summary judgement, any more than there is about a summary execution.

vanblokland:

I think your comments are spot on. I was playing with this feature yesterday and was able to crash Safari, using a font described here. Precisely the problem you described and a relatively trivial example (i.e. overflowing the argument stack). On the Mac, Webkit is passing the downloaded font data off to the ATS system service which does the rasterization:

http://trac.webkit.org/projects/webkit/browser/trunk/WebCore/platform/gr...

So bugs in ATS can be exploited using Safari. It's quite possible that this code could be tightened up and made more secure but I think some thought needs to be put in to making sure it gets there and not just assuming that it is secure. With Firefox, we have the same concern, relying on underlying platform API's needs to be done carefully.

Keep in mind that I think the same thing is probably possible with downloaded PDF files containing embedded fonts. Using Freetype or another rasterizer is possible but definitely not an easy task, especially when you want the code to be efficient.

John Hudson:

One thing I've never quite understood is to what exactly the Adobe vs. SSi case really applies to. The hints? The outlines themselves? Without hints is a font really a "program"?!?

Put another way, if someone takes a TrueType font and copies the outlines to make a SVG font (no hints) are they in copyright violation? Would the same user be in violation if they used the outline within a vector graphic of some sort and distributed that (i.e. something that still contains the outline)?

I don't think these are just theoretical issues but real issues that web designers need to be able to understand. "No embedding" flags, originally intended for use with Word documents, don't really explain much in this context.

Question: Do fonts embedded in PDFs use a system renderer?... like ATS?

In my opinion, outlines alone could not be regarded as computer programs. But my opinion does not count. If I remember correctly from having read earlier, either by (imho false) court judging or by copyright office regulation, font files per se (so even without hints) are considered font "sofware".

Actually, Uli is wrong in a lot of respects (because even if we disapprove of certain laws, they are laws), and while I disapprove of fonts being protected as "software", I do think they need to be protected in some way (as "work of art" e.g.).
His argument in one of his writings is telling however: If you draw an outline and save the font, the stipulated "software code" is generated automatically. Your work was not authoring software code, but of designing (as an artist) outlines.

aszszelp:

His argument in one of his writings is telling however: If you draw an outline and save the font, the stipulated “software code” is generated automatically.

I avoid the word "software" as a legal term, because it does not occur in German copyright law. Even the US copyright law does not use the word "software" (but the word "software" occurs in the American "software rental amendment act").

Could you please specify where I used the legal term "software"?

LOL!
Uli, is your essence of existence to argue?
You are questioning someone (who is – partially and with truly strong limitations – backing you) in the point that supports your case?
That's hilarious!
:D

" We take images from other websites all the time. Do I profit from it? No. But it is analogous to using fonts without proper licensing."

It is? It is not universally the case that it is unlawful to use images, and a whole body of law concerning 'fair use' tells that tale. Fonts, unless one is using the software in some journalistic pursuit of typographic commentary, have no fair use outside of license. SSI, was a troll under the bridge between proprietary and common font formats. They have no more foot stomping or hand biting to offer the world.

"Could you please specify where I used the legal term “software”?"

No. The continued circular opinions/arguments represent you as a waste of time with a mouth, but little or no ear function. When one does not understand the law even if one reads it, one does not understand typography even when one sees it, and ones grasp of type technology is second only to many billions of others, I think it is fair to say one should sit down, silence themselves and listen. Perhaps that applies to you.

Cheers!

I know Uli was being rhetorical, but he seems to have a typically incomplete knowledge of copyright law. He needs to look up concepts like fair use (U.S.) and fair dealing (Canada, U.K., some other countries). In short, you can post screenshots with attribution in many contexts without infringement. The same goes for photographs and other copyrighted works.

I don’t see how fair use or fair dealing would cover embedding fonts in Web pages even if they were copyrightable (or even considering that font files actually are copyrightable). Hence, Uli’s point is rather off topic. But the objection “The Web is all about replication of copyrighted works, so why should fonts be any different?” is one that merits discussion.

—
Joe Clark
http://joeclark.org/

I think (c) is off-topic as is the security issue (although I can't understand how Apple would not have fixed an exploit that Adobe patched 5 years ago). I'd suggest a separate thread, but haven't we been around that bush many times in the past.

The reason I started the thread was that I found Apple's positioning of the technology problematic and somewhat misleading, especially given the sensitivity and controversy swirling around the subject.

Anyway James Arboghast sent me the following notes off-list (too busy to contribute himself) suggested I post them here...

> CSS font embedding is mainly a distribution issue and a commercial
> rights issue. Here's why:
>
> * In principle, copyright is concerned with both intellectual property
> rights (art and design) and distribution rights (commercial vending).
> In practice, when copyright "law" and litigation are applied to
> unauthorized distribution of fonts, the principal of law is applied, not the letter.
>
> * Irrespective of state-defined copyright legislation in a given
> country, and whether or not digital type outlines and fonts are
> "copyrightable", font makers and vendors are automatically granted
> commercial rights---specifically the right to profit from their
> invention, design, and sale of their commercial font product(s)---like
> all other producers of artwork, design or software products.
>
> * The commercial rights of font makers, including the right to profit
> from their invention, design and sale of retail font products, are
> upheld on a case-by-case basis, usually in the "court of professional opinion".
>
> * Unauthorized distribution of commercial font products is typically
> prohibited by individual font product EULAs, which are "legally binding"
> contracts between the font maker or vendor/distributor, and all users.
> EULA contracts are enforcable by private litigation, independently of
> state-defined copyright legislation. EULA and commercial rights of
> font makers are enforcable, and have been enforced,
> by private litigation.
>
> * Retail OpenType fonts are commercial software by definition, as they
> contain OpenType feature programs. They are also commercial products,
> irrespective of whether or not they are software programs or
> "copyrightable". All commercial products are protected from
> unauthorized distribution by commercial law in the U.S. and Germany,
> Australia, Great Britain, France, Poland, New Zealand, Japan, Norway,
> most western countries.
>
> * Retail TrueType and Type1 fonts, whether they are definable as
> "software"/"programs" and "copyrightable" or not, are nonetheless
> commercial products issued with EULAs and protected from unauthorized
> distribution by commercial law in the U.S. and Germany, Australia,
> Great Britain, France, Poland, New Zealand, most western countries.
>
> * Copyright legislation in individual countries has nothing to do with
> the price of fish in China Also try: "the funny fish forging industry"
> ;-)
>
> * CSS font embedding is mainly a distribution issue affecting the
> commercial rights of font makers and vendors/distributors.

jdaggett: One thing I’ve never quite understood is to what exactly the Adobe vs. SSi case really applies to. The hints? The outlines themselves? Without hints is a font really a “program”?!?

Read the text of Judge Whyte's summary judgement on the copyright question, which Thomas posted above. It is very clear. It is important to bear in mind that what US copyright law considers a computer program and what it considers a sufficient degree of creativity to qualify for copyright protection are not what you or I might consider them. The USCO defines a computer program as 'a set of statements or instructions to be used directly or indirectly in a computer in order to bring about a certain result'. That is a very generous definition, especially if one pays attention to the allowance for both direct and indirect use. I don't doubt that some people might want to argue with that definition, or propose alternative definitions, but for the sake of US copyright law that is the definition that counts.

The other thing that is important to bear in mind is that, as Judge Whyte noted in his judgement, citing precedence, 'the requisite level of creativity is extremely low; even a slight amount will suffice'. That is to say, it is not necessary for a work to exhibit a huge amount of creativity -- what I'm tempted to call a 'romantic amount', since many popular notions of creativity are strongly influenced by romanticism --; the mere presence of creativity, in even a slight amount, is deemed worthy of protection.

Bearing those two things in mind, this is the key passage of Judge Whyte's judgement (by emphasis:

'The evidence presented shows that there is some creativity in designing the font software programs. While the glyph dictates to a certain extent what points the editor must choose, it does not dictate every point that must be chosen. Adobe has shown that font editors make creative choices as to what points to select based on the image in front of them on the computer screen. The code is determined directly from the selection of the points. Thus, any copying of the points is copying of literal expression, that is, in essence, copying of the computer code itself.'

"The code is determined directly from the selection of the points. Thus, any copying of the points is copying of literal expression, that is, in essence, copying of the computer code itself.’"

As the saying goes, A hah! Now I understand what they meant. Granted, I think this vastly understates what is involved, both technically and creatively, in designing a typeface but I at least see what lawyers have hung their hats on as a point of legality--or the basis for their litmus test. Whatever we typography lovers and type designers really think or express as what is involved--or what could be stolen is most likely too complicated to prove or disprove and would involve varied opinions of type geeks instead of absolute sightings by judges and lawyers.

ChrisL

Is Chief Justice John G. Roberts a computer programmer?

Let's assume, John G. Roberts, Jr., Chief Justice of the United States,

shown here: http://en.wikipedia.org/wiki/John_G._Roberts

is sitting in front of his computer, and after starting Coreldraw, as a first step, he draws with his mouse two egg-like concentrical circles as an "O" letter, such as shown here:

see http://www.sanskritweb.net/forgers/computerprogramm.pdf, page 3
(screen shot of step 1)

Then, as a second step, John G. Roberts, Chief Justice of the United States, saves this "O" letter as a CDR Coreldraw file, such as shown here:

see http://www.sanskritweb.net/forgers/computerprogramm.pdf, page 4
(screen shot of step 2, top)

Question 1: Is John G. Roberts, Chief Justice of the United States, a "computer programmer", who has just written a CDR "computer programme" called "O"-letter?

Your answer to question 1: Yes or No and Why?

Then, as a third step, John G. Roberts, Chief Justice of the United States, saves this "O" letter as a TTF TrueType file, such as shown here:

see http://www.sanskritweb.net/forgers/computerprogramm.pdf, page 4
(screen shot of step 3, bottom)

Question 2: Is John G. Roberts, Chief Justice of the United States, a "computer programmer", who has just written a TTF "computer programme" called "O"-letter?

Your answer to question 2: Yes or No and Why?

Is Chief Justice John G. Roberts a computer programmer?

I apologize, but I think it needs to be said…

"Is John G. Roberts, Chief Justice of the United States, a “computer programmer”"

What does being a programmer have to do with anything?

I thought we're talking about the definition of software.

If you're asking is "did John G. Roberts, Chief Justice of the United States, by saving that file, create a unique piece of computer code (ie, software) that describes a particular visual design that he created" then the answer is "yes".

But, again, IANAL and all that really matters is what the lawyers and judges think in a particular case. ;o)

I'm going to point out that Apple says you CAN use any font, not that you MAY use any font. It's trivial semantics, but potentially essential to any legal argument.

John Hudson: The code is determined directly from the selection of the points. Thus, any copying of the points is copying of literal expression, that is, in essence, copying of the computer code itself.

So from this description that would imply that someone who uses a glyph outline in a vector drawing program and distributes that outline contained within a vector file format like SVG is actually violating the copyright of the font in the U.S, since that file contains a copy of the protected outline.

And why wouldn't the same apply to the copy of FB Moderno contained in Headline.swf used to display the content title on this page, a font I have not licensed?

aluminum:

What does being a programmer have to do with anything?
I thought we’re talking about the definition of software.

That's a good joke.

What you call "software" is called "computer programme" in copyright laws.

"Computer programmes" (in your non-copyright-law parlance "software") are "literary works" written by authors. There are copyright laws which explicitely define that "computer programmes" are "literary works", e.g. the United Kingdom Copyright Act:

see http://www.sanskritweb.net/temporary/UK-Copyright-Act.jpg

The author of the "literary work" of a "computer programme" is called the "computer programmer".

Copyright protection only subsists in original works of authorship, e.g. in "computer programmes" as "literary works", written by "authors", e.g. here the "computer programmers".

For definition of "authorship" etc. see e.g. the American Copyright Law:

see http://www.sanskritweb.net/temporary/USA-Copyright-LAW.jpg

I hope this answers your question:

What does being a programmer have to do with anything?

And I hope that you now understand that "talking about the definition of software" means "talking about the definition of computer programme"

Mr. Hudson stated above in this thread:

The USCO defines a computer program as ’a set of statements or instructions to be used directly or indirectly in a computer in order to bring about a certain result’.

This is not only the definition used by the USCO (US Copyright Office), but much more important, it is the definition in the US Copyright Law itself:

see http://www.sanskritweb.net/temporary/USA-Definition.jpg

If you’re asking is “did John G. Roberts, Chief Justice of the United States, by saving that file, create a unique piece of computer code (ie, software) that describes a particular visual design that he created” then the answer is “yes”.

The answer is "no":

Firstly, a person who is not a computer programmer and who has not learnt computer programming languages and who is therefore not able to write computer programmes cannot be the writer of a computer programme.

Secondly, a person who draws circles with the mouse in a vector drawing programme (e.g. Coreldraw) makes circle drawings, but he does not write computer programmes.

Not to pick on FB Moderno too much but it looks like the Single User License Agreement of the Font Bureau disallows embedding altogether:

4. EMBEDDING RESTRICTED
(a) Except as may be specifically permitted herein, all embedding and transmission of the Font Software is prohibited.

(c) Licensee is not permitted to distribute documents using or containing embedded copies of the Font Software on the Internet or for other forms of distribution or downloading to unauthorized users.

So unless a user gets a special license from the Font Bureau, use of FB Moderno in webpage titles like the one at the top of this page is not allowed.

«Firstly, a person who is not a computer programmer and who has not learnt computer programming languages and who is therefore not able to write computer programmes cannot be the writer of a computer programme.»

That completely contradicts what you yourself said earlier in the same post.

If I use a WYSIWYG programming tool to create a simple program [please note that there is no such thing as a “computer programme”], then I am the creator of that program, despite the fact that I don’t know C++ or any other programming language well enough to actually write the code itself. If I create an entire font in FontLab and save it as a TrueType Font, then that is considered software and I am considered its creator, despite the fact that I did not actually code the file myself.

Computer programmer is (at least here) a protected title that requires at least a bachelor’s degree in computer programming, so obviously, the answer to your question is neither yes nor no, but rather, “Yes, if he has a degree in computer programming as well as being a judge”, which is of course completely irrelevant to the discussion.

Oisín:

> there is no such thing as a “computer programme”

If you don't accept the reality

http://www.sanskritweb.net/temporary/UK-Copyright-Act.jpg

then this is not my problem.

As regards the copyrightability of computer programmes which are not written by computer programmers, but which are generated automatically by computers themselves, see here

http://www.law.duke.edu/journals/dltr/articles/2001dltr0024.html

Generally said, this discussion got way too out of merit; Apple are just trying to market their cr*ppy (no offense, I am a Windows user, and under Win Safari simply sucks!) browser, boasting about its features. Possibly they should have used 'you ARE ABLE' in the place of 'you CAN', but the sense is not that different. There are no technical limitiations on such embedding in their implementation of the current CSS standard for web-fonts embedding, you have to admit it. Only the legal ones.

And, honestly, most laymen do not care about EULAs and never read it. The corporations do, though :)

@acnapyx
What do you mean with Safari (Win) sucks? I've recently downloaded Vers. 3.1 and it seems to work properly (loading times, css-rendering compared to firefox) under Win XP Prof. Is there a security problem? Okay, i don't like the blurish macintosh type rendering, i prefer cleartype.

«if you don’t accept the reality

http://www.sanskritweb.net/temporary/UK-Copyright-Act.jpg

then this is not my problem.»

You misunderstand. There is no such thing as a computer programme. When referring to computer programs, the word ‘program’ never has an -me at the end.

I also d/l-ed Safari for Win but dumped it very soon.

> what do you mean with Safari (Win) sucks?

Well, here are my impressions, even if this is slightly offtopic here:
- loading times (browser): ages.
- loading times (pages):
without flash - slower than IE6, IE7 and Firefox.
with flash - faster than Firefox and IE7, but slower than IE6.
- options and settings: too limited for my liking.
- CSS rendering: well, almost perfect, I admit it.
- type rendering: blurry (on CRT I do not use any smoothing, if possible).
- cache: IE, Firefox and Opera allow browsing the cache. Why Safari should not?
- plugins (Firefox-style)? Bwahahahaha.
and many, many more.

Regarding the support of web-fonts, right now there are no alternative browsers to compare with, so... still waiting for next version of Opera, which supposedly will support it.

Would I use Safari for everyday browsing? God, no! Would I remove/replace my default browser for Safari? At least, not now. I'm keeping it only for web design compatibility checks.

Maybe under Mac the things are different, but I seriously doubt it. This is a design issue, not a platform one. Apple simply do not provide enough advanced options for customizing the browser design/behaviour. Remember the discussion if there is an option to disable the web-fonts? Of course, Apple are new on this market, but Safari on my WinXP desktop still looks like an oversized iPhone :D

Oisín:

> You misunderstand. There is no such thing as a computer programme. When referring to computer programs, the word ‘program’ never has an -me at the end.

Google has 13,000 hits for websites in Great Britain, where this spelling still occurs:

http://www.sanskritweb.net/temporary/programme.jpg

I think this thread has struck the Law of Diminishing Returns and has been sent off on the path to entropy.

ChrisL

"Secondly, a person who draws circles with the mouse in a vector drawing programme (e.g. Coreldraw) makes circle drawings, but he does not write computer programmes."

That's like saying a painter doesn't make paintings...his paintbrush does.

"http://www.law.duke.edu/journals/dltr/articles/2001dltr0024.html"

That read seems specific to Fractals. A minute subsection of the complete works of computer programs. It's also an opinion piece that simply described the various potential approaches.

But the argument seems to be 'does the creator of the work own the copyright, or does the creator of the software that the creator of the work used to create their own software own the copyright?'

To go back to the painter, that wouldn't that be like arguing: does the artist owns the copyright to the painting or does the manufacturer of the paintbrush and paints own the copyright?

"Apple simply do not provide enough advanced options for customizing the browser design/behaviour. "

Nope. Apple doesn't emphasize user customization in most of their UI design philosophies. Subjectively, of course, that makes it a bad design for folks that prefer that.

Honestly, I've found nearly all browser suck in Windows. Safari is OK. Firefox is OK, but has constant memory leaks and weird behavior. IE7 has an incredibly obtuse UI but doesn't suck nearly as much as IE6. I actually rediscovered Opera as of late and kind of like that.

I like FF exactly because of its customizability. However its memory leakage is such a PITA. Hope v3 will be better in that respect.

>I think this thread has struck the Law of Diminishing Returns and has been sent off on the path to entropy.

True, and Apple's Safari page remains unchanged. Not heard from many type designers, so maybe this positioning isn't considered a big deal? However, if foundries do feel the text is misleading and want to be put in contact with the right people at Apple then please feel free to contact me off-list and I'll put you in touch so you can let them know your feelings on both the positioning and the technology itself.

"...I’ll put you in touch so you can let them know your feelings on both the positioning and the technology itself."

Thanks, Si. Consider me among the group.

ChrisL

My last word on this (well, in fact a quote from the very good feature about the value of typefaces):

A free society is dependent on voluntary compliance with the law. If we ever get to the point that we must rely on enforcement for compliance, the game is lost.

"A free society is dependent on voluntary compliance with the law."

That assumes the society both knows the law and has not been confused by posted statements from what they perceive to be reliable sources.

ChrisL

@jdaggett: FB Moderne used in this way here on Typophile is being used with permission from Font Bureau. No license broken if permission is given.

@Miss Tiffany: Good. But I'm guessing that a lot of web designers don't realize that the use here is a special exception and not generally allowed. In general, it seems like fonts embedded in Flash files have the same problems as downloadable fonts in Safari, namely that the actual details of the EULA effectively disallows their use in this way without special permission.

A more subtle point is that it seems like the FB Moderne EULA also effectively disallows the use of glyph outlines contained in a vector formatted image that is publically available, since "font software" seems to cover the set of points that determine glyph outlines.

There's a broader problem in that font makers have historically turned a blind eye to possible EULA infractions with respect to Flash and other similar technologies because they don't understand them, or feel they are innocuous (don't think they result in much extraction and unlicensed reuse) or don’t want bad PR amongst key customers, etc., etc., Now that there’s a far more serious threat looming they can’t afford to turn a blind eye and simply hope that their loose fonts don’t end up in the wrong hands. I don't see a double standard here - pick your battles, as they say.

You are correct that most foundries' EULAs do not allow embedding on the web in any form. Font Bureau's EULA, specifically, does not allow embedding of any kind without an extended license.

In this instance I would hope most people would understand that Font Bureau is featured in this way because they have chosen to be.

I think font vendors' EULAs should be absolutely explicit about what they allow in terms of Flash SWF embedding, raw fonts on Web servers, EOT usage, etc.

Cheers,

T

Not heard from many type designers

Probably because most of us realize that what we have to say won't make a blind bit of difference.

I think font vendors’ EULAs should be absolutely explicit about what they allow in terms of Flash SWF embedding, raw fonts on Web servers, EOT usage, etc.

I would like to do that, but have a hard time figuring out what's in my best interests. Generally, I wait to see what organizations with the resources to address such issues comes up with, and take my lead from them.

>I would like to do that, but have a hard time figuring out what’s in my best interests.

This was the focus of my talk at the Business of Type event - a framework for how type designers can learn about these different technologies, and make informed decisions around how they license their IP for use with them.

A good initiative Si, but the event was too far away for me to attend.

I think font vendors’ EULAs should be absolutely explicit about what they allow in terms of Flash SWF embedding, raw fonts on Web servers, EOT usage, etc.

Frankly, some of the contents of font EULA agreements seem byzantine beyond belief.

As an experiment, I poked around in Font Book and picked "Century Schoolbook". Not a normal Mac OS X system font, probably installed with MS Office. The font information shows Monotype as the foundry along with a license URL. Hmmm, page not found, okay, dig through the Monotype fonts.com site to find their single-user EULA agreement, which I'm assuming is what applies to me. Section 12 basically disallows use of a font in anything other than a non-editable static graphic and only for "personal or internal business use" and not "commercial" use. Definitions for these terms are given but it's not quite clear where "for use in a website graphic" falls. It also appears to disallow anyone from sending anything containing a glyph outline, such as a path within an Illustrator file or a vector graphic formatted image like SVG, to a user without a license.

But wait, dumping out the OS/2 table for the font shows that fsType embedding bits are set to "editable embedding", so maybe it really is okay to use in a website graphic? And maybe this could be used with Safari as long as the font is not cached?

Seems like font users who in good faith try to determine "can I use this font for this purpose?" need something that's a little simpler and more straightforward.

Fonts should come in two forms:

* lead = no embedding electronically
* electronic = embed as you see fit/need.

;o)

>Not a normal Mac OS X system font, probably installed with MS Office.

This font would be covered by the Office EULA not a Monotype stand alone license. Document embedding (as defined in the OpenType spec) is permitted, and explicitly so in the most recent version of the Office EULA...

:D

:D ?

Was a reply to aluminum.

This font would be covered by the Office EULA not a Monotype stand alone license.

How would a font user know that? I don't think Office ever announced that it was installing those fonts as part of the install process and the data within the font itself indicates something else (i.e. that the license is covered by the Monotype EULA)!

I'm sure you're right about this but my guess is that other applications also install fonts without explicitly altering the license notice within the name table strings. Makes it difficult to track down the EULA that actually applies. Same probably goes for all the lovely fonts that ship with Mac OS X, are they covered by the Apple EULA when the font data indicates something else? Hard to figure out from a font user's perspective.

Document embedding (as defined in the OpenType spec) is permitted…

Hmmm, are you distinguishing document embedding from a webpage's use of downloadable fonts? I know others have argued that this is really font linking rather than embedding, but I thought Microsoft claimed that the meaning of those embedding flags extended to the use of downloadable fonts in EOT form.

from :

http://www.fonts.info/info/press/free-fonts-for-font-face-embedding.htm

Safari 3.1 for Windows and Mac supports the embedding of “sfnt fonts” (TrueType, OpenType PS, OpenType TT) using the font-face declaration. Technically the fonts are not embedded in the website, but they are simply linked like an image file. Thus the fonts need to be stored on a public server. Since you cannot upload commercial fonts to a public webserver, you are limited to freeware fonts. FDI fonts.info believes in the future of web fonts, so we decided to provide webdesigners with a set of high-quality web fonts supporting a wide range of character encodings.

>Hmmm, are you distinguishing document embedding from a webpage’s use of downloadable fonts? I know others have argued that this is really font linking rather than embedding, but I thought Microsoft claimed that the meaning of those embedding flags extended to the use of downloadable fonts in EOT form.

Yes, my reading of the OT spec is that the embedding permissions relate to fonts being embedded within document files. With EOT the font is not within the file but is tied to it via the URL-binding system. MS consulted with the industry at the time. However, over the years many vendors have put additional restrictions in their EULAs, around commercial embedding (eBooks), and technologies they don't like, such as PDF, or more commonly EOT.

@Ch: Wow! Nice demo page too! Shows the Safari-specific CSS styles (fill, stroke, transitions, etc.), good stuff.

Happy ending - my opposite number at Apple reports that the Safari promotional pages have been updated...

http://www.apple.com/safari/

http://images.apple.com/safari/docs/Safari_Product_Overview20080602.pdf

Well done Si!

It might have been related to our ongoing harassment of the good folks at Apple over the past two months.

T

>Well done Si!

>It might have been related to our ongoing harassment of the good folks at Apple over the past two months.

Yep, I'm sure that various people have been nagging them - and it just took a while for the bureaucratic cogs to turn.

Excellent. Thanks for the news Sii.

j a m e s

Actually, I found something concerning the implementation of @font-face in Safari which might be considered a bug. However, I have not found any page on the apple homepage to report Safari bugs. Any idea where and how to report?

Szabolcs

You should be able to find contact info for Dave Hyatt online - he's the Apple engineer responsible for Safari’s support for raw font downloads.

Thanks, Si!

(I'll just have to find the time to put my observations together)

Szabolcs

http://bugreport.apple.com is the preferred route especially if you have more technical details, but you can also go View->Customize Toolbar and activate the Report Bug button or (at least on a Mac), you can go to Safari->Report Bugs to Apple.

«El futuro es una línea tan fina que apenas nos damos cuenta de pintarla nosotros mismos». (La Luz Oscura, por Javier Guerrero)

Thanks, Matthew, I just could not find that site! (I noticed the bug report form built into Safari, but it was way to generic. Also, it was so impersonal, you could never know whether the mails incoming to the mailbox from that form are not redirected directly to /dev/null ;-) )

You know, that's something I really disapprove of with Apple. Absolute intransparency. Also: even bugreport.apple.com is a) well hidden b) you need a registration (oh, the first option they offer you is 3000+ USD/year... of course, if you scroll down, you'll find the 99USDp.a. for students, and at the very bottom, as if they did not want you to notice, the free "OnLine Membership").

The stipulated aim of making everything as easy for the user as possible, "you don't have to care about technical details" is good. It should be the primary task of good USER (oriented) interface design. However, it should not completely put the user under tutelage, if he _wants_ to care for technical details.

At least in the current situation it gave me the feeling, that bug reports are actually _not_welcome_. Or why else would you make it so hard to post improvement suggestions?

Let me have been mistaken. Let me find out, I was wrong!

Szabolcs

The bugreport system is open to members of Apple Developer Connection, and they do make it clear that it's free (I don't think I've had any indication to the contrary), it's designed for developers more than for end-users, because end users tend to not be able to describe in the clarity and detail that is necessary generally to diagnose problems.

They want registration for more serious natures so they can follow up. If they have a problem or a question they want a way to contact you and it integrates within their main bugtracking system for all software. I'm sure also WebKit has one as well. While I still have some bugs that have been left inattentive for 3 years, others have been rapidly fixed, and not necessarily in the order I would have thought (eg, easy things are unfixed, hard things get fixed). Partly it depends on how many people submit the same complaint.

«El futuro es una línea tan fina que apenas nos damos cuenta de pintarla nosotros mismos». (La Luz Oscura, por Javier Guerrero)

back to the original topic:

Apple says you can use any font (obviously according to it's licence) but I think they refer to any type of font.

I mean, if it only supported TTF fonts what good would that be. Still according to licences, but you still wouldn't be able to use *any* font in that case, only a fraction of them.

Also like a photocopier, some photocopiers can only photocopy a small area, so a big one that says, "This photocopier can photocopy ANYTHING" doesn't mean "You are encouraged and legally entitled to duplicate material you may not otherwise have licence to distrubute as long as you use this machine!"

Give apple a break, they could easily have made their own custom font format for safari fonts only, but they didn't…

I think you're reading too much into this.

Apple marketing used wording that some people in the font community objected to. They made the Apple font people aware of the issue. Apple changed the wording. End of story.

Good to know.

As I have described above, this was the impression I got "from the outside".
So, according to you, it's worth getting into it. And I will, when I find the time (I like to do stuff either completely or not at all: no half jobs. This goes for bug reporting (being as specific, as possible, describing the problem with technical details, etc.).

Szabolcs

Safari bugs are not really "Apple bugs". A better place to report them is the Webkit page:
http://webkit.org/quality/reporting.html

What are the bugs?
(Can you send me an email if you don't want to describe them here?)

Ralf

Ralf,

I wanted to write you directly anyway (in a different matter), so I will.
However let me ask, are you somehow involved in Webkit.

(--- the bug is not a rendering bug, so maybe it's not Webkit, after all... (Webkit is the name of the _rendering_ engine, isn't it?))

Szabolcs

Webkit is the application framework which several browsers (including) Safari are build on. Safari versions can be a little behind the current Webkit versions. So it is a good idea to check if your bugs might already fixed in the lastest nightly build of Webkit. http://nightly.webkit.org/

I have no connection to Webkit, I am just VERY interested in webfonts. ;-)

What they intended to say was simply that standard Web formats for embedded fonts now work on Safari. The phrasing, unfortunately, could confuse some people about licensing issues, but it was clear enough to me that this wasn't what they were talking about.

I don't think so, if you recall in 2008 they only supported raw TTF files, and that wasn't considered a standard web font format back then.

PS. it's always interesting what dead threads the spammers bring back to life, and even more interesting to see how people respond to them five+ years after they died. ;-)