Well, you can easily turn hinting off.

But if you like hinting have a look at this patch which enables subpixel hinting within freetype.

Note, these are different patents - the article is talking about the expired Apple patents (http://freetype.sourceforge.net/patents.html) not the CT patents (http://david.freetype.org/cleartype-patents.html) which have not expired.

Linux distributions (like Ubuntu) have been turning on native hinting in FreeType for ages, no news here. I myself is disabling it since the autohinter does actually better job, I prefer or my fonts look the same (the autohinter does a decent job) over having hinted and unhinted fonts look different, I go even further and disable the embedded bitmaps too.

I don't think FreeType implements any of ClearType patents, they however had implementation of native hinting patented by apple but is deactivated by default, so that if you can license the patent from apple (or live in country where it does not apply), you can compile FreeType yourself and enable it. So what changes now is that it is enable by default, which hardly makes any practical difference for many distributions.

I guess the David Berlow version of this is that FreeType, having fantasised Apple patents that would prevent them from implementing full hinting support in their renderer -- and having apparently fantasised being contacted by Apple's legal department about said fantasy patents -- have now fantasised the expiration of those patents. Fantastic!

I don't think Freetype were ever contacted by Apple. As I recall the issue was first raised by the makers of a competing rasterizer.

I recall the earlier version of the FreeType patents information page mentioning something about contact with Apple's legal department. Unfortunately, that version doesn't seem to be cached.

An earlier version did recommend Freetype customers contact Apple legal if they wanted to license the patents.

Also... http://web.archive.org/web/*/http://freetype.sourceforge.net/patents.html

Update...

I found a cached version:
http://web.archive.org/web/20080430122831/http://www.freetype.org/patent...

Near the bottom it says:

It is true that we have been contacted by Apple's legal department, but that has never been in the clear intent of suing us, which isn't too surprising given that FreeType doesn't harm Apple in any way.

On the contrary, because of FreeType, Apple has already seen new patent licensees in the embedded market.

Although it's certainly possible Apple raised this issue first this was the source I was referring to... http://lists.gnu.org/archive/html/freetype/2000-02/msg00017.html

Thanks, Si.

On the subject of Linux font rendering, one thing that surprised me is the very great difference in text rendering quality between different browsers. Firefox rendering in Linux is very poor compared to Opera. [These comments based on admittedly limited experience looking at my wife's Ubuntu install.] I wonder if that will change now, or if it is something independent of FreeType.

I'm surprised to hear that "most Linux distributions" already had hinting turned on, as claimed by the article.

Perhaps its author means commercial distributions which could pay to license the patents, such as SUSE or Red Hat, as opposed to the free community distributions which are what are used by most individual users of Linux.

>I guess the David Berlow version of this is...

...that there was never anything for FT, or any other developer to be afraid of from Apple over TT.

The David Berlow version of this was, that groups or individuals who were trying to, or actually were worrying developers away from TT technology, were doing a disservice to the general user community.

Clear now, I am sure ;)

Cheers!

David: Clear now, I am sure

Clear as ClearType can be.

Good!

Now, somewhere there must be a FT implementation that uses all the TT hinting language at all sizes with antialiasing... right?

Cheers!

And,

sii> As I recall the issue was first raised by the makers of a competing rasterizer.

Who might that have been? What does it mean to compete on rasterization? and are they finished with being scary patent-mongering user-haters yet?

Cheers!

What does it mean to compete on rasterization?

Presumably that one has a commercial rasteriser implementation that one is seeking to license to customers for money, and then these pesky open source folk come along with a free rasteriser. Now, presuming that one has paid a license fee to Apple to implement their patents in your commercial rasteriser -- whether one needed to or not --, one might see it as unfair competition for another rasteriser not to have paid Apple to license those patents.

According to the allegation to which Si linked, Sampo and Bitstream are the ‘scary patent-mongering user-haters’. I suppose Sampo knows as much as anyone about TT patents, since he wrote them. Apple, for their part, have apparently been happily collecting patent license fees from anyone offering to pay them, even as they appear to have taken no legal action against those who didn't. All very messy.

JH> According to the allegation to which Si linked..

That's all nonsense John. A 10 year old link to an email from no one I've ever heard of, full of uncertainty and a bunch of assumptions.

What I meant by competition doesn't exist in rasterizers. This has all been a user-hating attempt to keep open source from being better than the OS at something important.

Cheers!

David: This has all been a user-hating attempt to keep open source from being better than the OS at something important.

An attempt by who? The makers of the open source rasteriser? They seem to be the only people who ever publicly stated that the Apple patents prevented them from doing something.

What I meant by competition doesn't exist in rasterizers.

Huh? You didn't say you meant anything by competition. You asked what it meant when Si mentioned ‘makers of a competing rasteriser’. What Si meant is obvious: a different product competing for the same market of customers.

I have no idea if the ancient allegation is true or false, someone who really cared could check the OT-L and see if the claims hold up. But it seems to me more likely that someone who paid for access to these patents would raise this concern, rather than Apple who really doesn’t care.

>What Si meant is obvious: a different product competing for the same market of customers.

I didn't mean the meaningless meaning of competition that's generally used around MS products. Are one of you going to boldly suggest that any substantial number of users choose among platforms by rasterizer, even when each platform has several?

The fact is that Linux users had their font supply and font quality cut by unfounded developer fears. That's all, and it's hardly fantastic.

Cheers!

It is all about embedded systems, no one talked about users choosing different rasterisers.

From http://web.archive.org/web/20080430122831/www.freetype.org/patents.html:

Myth 2: Apple Is Suing (or Sued) FreeType

This complete myth apparently started with this article on the SlashDot news site. Too bad the editors did neither care to check the submitted link nor even tried to contact us, we could have helped them!

It is true that we have been contacted by Apple's legal department, but that has never been in the clear intent of suing us, which isn't too surprising given that FreeType doesn't harm Apple in any way.

On the contrary, because of FreeType, Apple has already seen new patent licensees in the embedded market.

Khaled: It is all about embedded systems, no one talked about users choosing different rasterisers.

Exactly. End users are not the customers for rasterisers: they're the downstream consumers. When Si spoke about competing rasteriser makers, he obviously meant companies competing for actual customers, i.e. makers of rasteriser client devices and software.

David: The fact is that Linux users had their font supply and font quality cut by unfounded developer fears.

Or was it down to developer aversion to licensing patented technology in an open source product?

Or was it down to developer aversion to licensing patented technology in an open source product?

With many free software licenses, you can't technically license patents, because you can't pass the protection you perceive with your code, see §11 patents of the GPL for example. Also, there are the stand against software patents that many free software developers would take, so it is on principle to avoid software patents, neither licensing them nor intentionally violating them.

>...he obviously meant companies competing for actual customers, i.e. makers of rasteriser client devices and software.

Thanks for the decrypt. But "obviously"... Quartz and ClearType, (the rasterizers) are not for licensing as far as I know. We are obviously talking here about the scaler, for which these ex-patents were made and obviously not rasterization. Is that not obvious?

Cheers!

David: But "obviously"... Quartz and ClearType, (the rasterizers) are not for licensing as far as I know.

As far as I know too, but FontFusion is for licensing, and Monotype's UFST is for licensing, and they're in direct competition with FreeType for the embedded market.

Does anyone other than you use the term rasteriser in such a precise way that it is distinct from scaler? Apple seem to use the terms interchangeably, as do other people who show up in Google searches for ‘font scaler’.

Since the business of licensing technologies for font rendering tends to involve whole solutions for embedded systems, distinguishing bits and pieces of those renderers seems irrelevant to the discussion about who would or would not have had a vested interest in persuading FreeType that it would need to license patents from Apple.
___

I think Khaled has nailed it. The issue for FreeType was most likely inclusion of patented IP in open source software, irrespective of whether Apple intended to enforce patent licensing.

What I meant by competition doesn't exist in rasterizers.

exhibit a (note the domain):

http://freetype.com

JH> Does anyone other than you use the term rasteriser in such a precise way that it is distinct from scaler?

I know you don't like hints, but I'm only tying to help. Them patented instructions, in post-aliased screen font environments, were not part of the solution of hints that worked... I said so long ago and since. I came back to cover your fantasies of my fantasies in this otherwise gentle conversation.

Anytime anyone wants to talk about all of raterization, (sizing, scaling and rasterizing) and use the term rasterizer, that's fine. Anytime anyone wants to talk 'bout one of them and use the term rasterizer, that's for less expert forums, I think. The patents, I think were for code that does scaling useful in aliased rasterization.

>Since the business of licensing technologies for font rendering tends to involve whole solutions for embedded systems,

I had not noticed this tendency from the Apple-MS TT license onward.

>You didn't say you meant anything by competition.

Hmmm? I mean(t) that in the post-aliased world of screen fonts, FT solutions could bring competition to the Quartz and ClearType scaler and rasterizer solutions and bring true competition to many parts of the market vertically.

>What Si meant is obvious: a different product competing for the same market of customers.

That, to me, is not obvious... Or are we so different, my brother, (the same market of customers), that we should be so devoted to totally different scalers and rasterizers? ;)

Or would you suggest that the market of font software implementers gives their customers a choice? Or would you suggest that the choice these implementers have is wide enough to justify use of the terms like choice and competition?

It all, choice and competition, depend(ed) on developers getting amp-ed about hints that work if they need them, for the market transitioning from A to AA fonts. I see one 'r two companies that've been in the hinting business constantly since the mid-1990's. This miscreation of a fantasy problem with TT patents is one reason.

And should I ask sii... who jumped off the hinting bandwagon recently? Head 'em up, move em out all day long here to ensure windows users don't barf at their type. Must be for the sake of competition, eh!?

Cheers!

David: Them patented instructions, in post-aliased screen font environments, were not part of the solution of hints that worked...

So what impact did not supporting them have on FreeType's competitiveness in the same post-aliased screen font environments?

64K apologies. I used “rasterizer” as short hand for what people like Bitstream, Monotype and others refer to as “font engines” in their sales blurbs. Was specifically referring to those "font engines" that claim to handle TrueType fonts.