Slashdot Mirror
Microsoft Typography Withdraws Free Web Fonts
jonadab writes: "Microsoft Typography has for years provided a set of very nice
True-Type fonts for free-as-in-without-monetary-cost, including
the excellent Andale Mono (the only scalable fixed-width font
I really like). They are gone. Here
is the Microsoft page where they formerly were, which now tersely
explains that they're not available any longer. There is an
article
about this on extremetech. According to the article, Microsoft
says the withdrawal of the fonts at about the same time as the
LinuxWorld is coincidence. The article also references a Debian
package that has been removed from the distro because of this.
If I understand my rumours correctly, it was a package that
downloaded the fonts from MS, displayed their EULA, and allowed
the user to extract and install the fonts. It was possible to
do the same thing using other distros.
Guess it's time for the OSS people to make some decent-looking
scalable both-screen-and-printer fonts (preferably TrueType).
At minimum, we need nice-looking serif proportional (to replace
Verdana), a sans proportional (to replace Georgia), and a
mostly-sans fixed (to replace Andale Mono), all with good
language support.
This should have been done a long time ago, since the MS fonts
were, albeit $0, not licensed in an open fashion. We always
knew we were relying on MS Typography's generosity, and that
these could disappear at any time. But now the need is more
urgent."
"In a statement, a spokesman for Microsoft said that the company withdrew the free fonts for several reasons. "Most users who wanted the fonts have downloaded them already," a company spokesman wrote in an email to ExtremeTech. "They ship with recent OS's - Windows XP and Mac OS (via Internet Explorer). Microsoft has also found that the downloads were being abused - repackaged, modified and shipped with commercial products in violation of the EULA [licensing agreement]."
;) (joking, joking, out down the chair)
So, everyone who already wanted them had downloaded them, they come with XP and OS X, and people were abusing them.--Damned OSS hippies
Didn't know you could determine that everyone who needed them already had them. Interesting. I'd like to see the metric used to determine that.
Sent from your iPad.
1. GRANT OF LICENSE. This EULA grants you the following rights:
* Installation and Use. You may install and use an unlimited number of
copies of the SOFTWARE PRODUCT.
* Reproduction and Distribution. You may reproduce and distribute an
unlimited number of copies of the SOFTWARE PRODUCT; provided that each copy
shall be a true and complete copy, including all copyright and trademark
notices, and shall be accompanied by a copy of this EULA. Copies of the
SOFTWARE PRODUCT may not be distributed for profit either on a standalone
basis or included as part of your own product.
So uhm, looks like I can distribute it without charge. Someone give me a place to stash 1.5M:
-rw-r--r-- 1 jmd jmd 1524606 Dec 7 2000 truetype.tar.gz
PFAEdit is a sophisticated graphical editor for designing and editing Postscript fonts.
MLT - simple and robust open source multimedia framework for Linux
I found a nice program a couple of days ago.
Try pfaedit. It supports TTF fonts as well as bitmap fonts and has a lot of good features. It supports simple latin-1 fonts as well as unicode fonts and author seems to really know what he's doing since website tells a lot about differences and inner workings of different font types. Pfaedit seems to try its best to convert everything necessary so user doesn't have to worry about them too much.
It is a work in progress but I think good artists can make miracles with it. Website also has good documentation altough I think in-program documentation could be a bit better (just to know where to start). I tried it myself a bit but since I'm no artist..
Website also links to other free font editors but pfaedit seems to be most mature. Most of others only support bitmap fonts.
Well, Linux has always had a problem with nice looking fonts. It doesn't have any.
And who wants to program fonts when they're trying to program something cool? Font making is generally not covered in Computer Science classes.
My suggestion? Pay to have them done by a professional. Bang together a donation page and try to set up a deal with someone who can do the work. If you name the font set after the company and put contact info in there, it's free advertising.
I'm sure they'd offer a discount if you did something like that.
Democrats or Republicans. They are both taking us to the same place and they are not afraid of us anymore.
Erik Dalén
designing fonts is not rocket science, but it comes pretty close. typography might even be the equivalent of rocket science in design.
what we certainly don't need is hundrets of people making up amateur open source fonts, but a few people who know what they're doing.
what might be possible is to find and old font (most common fonts are quite old, and the other good fonts usually are based on them), or a former-eastern-block font and reconstruct it. but you still need quite some experience to do this. i personally wouldn't even try.
--
making up good sigs is a hard thing to do.
opentype overview
-Kevin
Actually, I would imagine that Slashdot is one of the very few sites that doesn't suck, and uses what ever fonts your browser specifies. Of course, since I have fonts disabled, I only see Times New Roman and Courier New in my browser anyway.
<RANT>
Do any of you adults out there remember when Tim Berners Lee came up with this stuff, and how HTML was just supposed to be a recommendation on how to present the data, and not a formal definition of what it's supposed to look like? Hmmm? If you develop a site and you want it to look a specific way (then you're... nm), then use flash or pdf or postscript. But if you give me HTML, I'm going to render it the way I want to see it rendered.
</RANT>
-brian
Also, remember that you aren't just designing an ASCII character set. You need a math font, such as the STIX project, and what about Chinese, Arabic,...?.
Anyway, to answer your question, Knuth's Metafont is a standard part of TeX. It's a special-purpose programming language for designing scalable fonts. Way ahead of its time! The problem is that its output isn't in any modern format. There are various conversion tools, but I don't know how good they are (pktrace, textrace, ps2mf, Mathkit,mktekpk).
There are also some free font-design tools that I know even less about: PfaEdit, TTX (converts between TT and XML, so you can edit by hand).
Find free books.
You're right of course. However, making font design tools widely available is still a good idea. If we want more good fonts, then we need more good designers. And if we want more good designers, then we need to give people who aren't designers yet the tools to get there.
What do you think Hermann Zapf's first font looked like? Probably horrible.
--Lawrence Lessig for Congress!
These fonts are still available from the Corefonts project. This is perfectly legal and in accordance with the EULA; see the copy of Microsoft's FAQ. The project also includes "a source rpm that can be used to easily create a binary rpm package that, when installed, gives access to Microsoft's TrueType core fonts for the Web."
It is hard to read what M$ intends to do by removing free TT fonts from public download, but I cannot see it as a good thing. Basically, M$ is creating a condition in which browsers running on *nix may not (at some point) be able to render Verdana, which is probably one of the most common fonts on the Web. If Verdana is not installed on (say) a Linux PC, all its browsers (Mozilla, Konquerer) will need to degrade to another alternative non-serif font, unless Verdana can be installed in some way or licensed for distribution with Linux distros.
Keep in mind that M$ commissioned one of the great designers (Matthew Carter,of Bitstream, now of the firm Carter and Cone) to design these TT fonts for onscreen legibility. It will not be easy to replace them (Verdana in particular) with another freely-available font.
However, the OSS community is is dire need of a set of fonts that compete with those available on the M$ platforms, both for on screen use and for printing, especially if it hopes to expand onto the office desktop.
Suggestion to the OSS community: have the emerging alliances between the various distros (e.g.,LSB) create a shared fund, used to commission someone to design a serif and non-serif font for general use on all platforms (including Linux). The goal should be to create a font as good or better than the ones that Matthew Carter designed. And give Matthew Carter first dibs on trying to best himself, thereby ensuring that whatever succeeds Verdana will be of the same style and eloquence as Verdana itself.
In the meantime, (and this may be flamebait) distros may wish pay the evil empire to license Verdana and Georgia for distribution with Linux.
Creating clear, scalable, attractive fonts is neither easy nor cheap -- and the people who care about and need quality fonts are users, not programmers. Given that free software is driven by the needs of technocrats and not by the desires of users, there is little likelihood that high-quality "free" fonts will emerge.
The technocrats argue that "making fonts can't be that hard" and "just whip some out in the Gimp", betraying their ignorance. Technocrats won't stand for a non-programmer making such "it's easy" comments about writing a complex application, but they hypocritically think they are so wise as to belittle the complexities of designing quality fonts (or user interfaces, or whatever else isn't considered "elite" enough for their full understanding).
Microsoft is not stupid; it has identified weakenesses in free software, and is exploiting one (the lack of fonts) to its advantage. People in graphic arts or publishing have no interest in free software because it, quite frankly, does not care about them.
The Mac, which has excellent font support, proves that this is not an issue of free-versus-Microsoft or Unix-versus-Windows; clearly, the Unix-based OS/X provides the kind of font support that users need. The reason for good fonts on the Mac is motivation: Apple cares about meeting the needs of graphic artists and publishers.
The downfall of free software is its elitist and myopic attitude. Microsoft knows this, and can use its power to provide the "niceties" (like quality fonts) that free developers ignore.
All about me
we certainly don't need is hundrets of people making up amateur open source fonts, but a few people who know what they're doing
I disagree. I think hundreds of people making up amateur fonts is exactly what we need. After a while, a few of these people will get really good at it, and then we'll have the latter half. Meanwhile, and more importantly, a font design sub-culture will have been established.
The only way to learn rocket science is to DO rocket science. I have never, ever seen a difficult field that could be learned any better than by just flat out trying to do it and puzzling through every obstacle.
In light of the observations above on the Georgia et al. EULA, does anyone have the EULA for arialuni? Perhaps it was offered on the web with similar terms.
Designing a font is nigh-on an artform. For it to work properly, first of all, you need to create between 70 and 130 characters (as a minimum) which are all consistent, work together properly (i.e. fit properly next to and above/below each other) and, most importantly, look good.
That's which someone can "'own' a fucking font" (in your words)... It takes a lot of work (sometimes years to do a whole Unicode font) and costs a lot of money to do. Take a look at the majority of free fonts on the market - if they were developed for free, chances are they have a lot of characters missing (especially accented characters needed across the world outside the US) and a lot of bugs.
...should be taken out and shot. Personal prefernce I know, but years of seeing shitty PowerPoint presentations and Word documents laid out in it have convinced me it's the sloppiest, ugliest, most unprofessional-looking typeface there is. It's not even good for lettering comic books.
The only good use i've seen for it was when I got a credit card in the mail. It was in an envelope, badly printed with my address in blue Comic Sans. Inside that envelope was the real one, a regular windowed envelope marked "disguised mail". The Comic Sans had done a good job looking unprofessional, to hide the fact it was a letter from the bank.
Hundreds of people learning to make fonts won't result in good fonts, just hundreds of people frustrated at the amount of time they wasted in making fonts nobody uses.
Designing fonts from scratch takes years to learn; even copying fonts takes quite a while. I've worked with type designers and have, in fact, created my own fonts, one of which is a rendition of an older font (from the 30s) called Albertus.
It's a tedious process even with good tools. It's mostly about drawing and then matching those drawings to PostScript-possible splines.
Unlike kernel development or software collaboration, in which hundreds of people can each contribute something that winds up in the final results (or even tens of thousands), font design is a lonely profession with lots of abandoned work.
Freelance tech journalist for the Economist, MIT Technology Review, Macworld, and others
Sounds like a genetic algorithm might work well, with a human viewer providing the fitness test.
Given that having a human decide the fitness of each generation will increase generation times, we could speed it up again by enlisting all those hundreds of amateurs who can presumably recognize a good font even if they can't produce one.
The difficulty would not be producing the individual letters, but keeping the "look and feel" consistent across all characters in a font set. The genetic algorithm's "genes" (units of iinheiritance) would have to consist of higher level abstraction, such as "serif" or "bold" or "elongated". These higher abstractions would then be applied to create a character set with a consistent look, perhaps in a way analogous to embryogenesis.
Opinions on the Twiddler2 hand-held keyboard?
Probably my favorite thing about Andale Mono is that the zero has a dot in the center, making it trivial to distinguish from the letter O, which does not have the dot. Few other monospaced fonts today have that feature.
To programmers, that's a big win. In fact, making C-syntax characters look different ("1" v. "l", "{}" v. "()", "O" v. "0", "." v. ",", ":" v. ";", "'" v. "`") should be a priority for anybody working on an Andale Mono replacement. (Andale Mono could be improved on a few of these).
I've often wondered if I might even use a font where a "{" had an extra do-hickey (not quite sure what that would be) to distinguish it from "(". Even if it didn't look like a traditional "{" it might be a win. (But of course, I'd have to see it first).
(P.S. Since I'm dreaming, I might as well wish for a pony, too...)
Heh, sorry for the misleading subject. Actually, in the US you can CAN own a font, you just can't own a typeface. A font is a computer program, and as such, is protectable under copyright law. The name of a font or typeface (like Helvetica) is a trademark, and as such is protectable under trademark law. However, the design for the typeface itself, although protectable in many parts of the world (Europe, Australia), is NOT protectable in the United States.
This pisses off font designers in the US. Ironically, the preceedent for this dates back to the 18th century, when US font manufacturers (who made their fonts by pouring lead into moulds), wanted free license to rip off their counterparts in the Old-world. They got fonts declared non-protectable, much to their chagrin several centuries later...
Back in modern times (about 10 years ago), this loophole was exploited by fly-by-night punks (precursors to spammers) who created "shovel-ware" CDS, packed with fonts created by scanning in the output of established fonts. The lazier ones omitted the step of printing out and rescanning typefaces, and instead resorted to "jiggling" the coordinates in an existing font and selling the output as their own, or by ripping off commercial/shareware/freeware authors by taking just the font and renaming it. These guys (the ones who skipped the scanning step) got slapped with a lawsuit by Adobe and a bunch of other font producers, and have since disappeared.
The point? You can own a font, you can own the name of a typeface, but you can't own the design for a typeface in the US (with one exception - if you can get the US Patent office to grant you a design patent, you can own the design.)
And, creating typefaces (and going one step further, turning them into fonts) is a difficult and underappreciated occupation in the US, so don't be surprised if few people (if anyone) rise to the challenge of creating one for free.
You'd probably be interested in ProFont - a font designed for programmers, which has existed for years, but few outside of the Mac programming community know about it. It was specifically designed to be readable at 9 point, with similar characters distinctly different, as this page demonstrates. The full distribution includes TrueType, Type 1, and bitmap versions of the font for Mac and Windows. You can also download a look-alike bitmap version for Windows here.
I've been using ProFont for years as the font in my editor when coding, and found it very helpful.
Do we need a Free Font Foundation?
I've tried for some time to get some high quality fonts "donated" to Gnome or XFree86; although this work is still continuing, we're not getting very far. Here's why. Maybe you can help.
It's *difficult* (as others have said) to design a successful typeface. For a poorly hinted font, an hour or two on each character design will get you basic latin one in about five weeks, and then you spend another two weeks with hinting. If that sounds a lot of time, remember that you need to adjust sidebearings (nn sit further apart than oo, or you'll get spots of light and dark on a page/screen, for example) and kerning (Wa closer together than Wh, "r," closer than "n,", "fk" further apart to aviod a glob at the top.
It turns out that an R isn't simply a P with a tail, an E sn't an F with an extra leg, in most designs, particularly the more calligraphic such as Palatino.
So, it's a lot of work to make a font, and for Linux and the Free Software movement, we want fonts that support as many languages as possible, and as many scripts as possible, so that as many people as possible can use the software.
That means even mnore work, and a lot of time from people who are primarily creative artists and designers, with a strong techincal background.
There are three main font formats in widespread professional use today: TrueType, Type 1 and OpenType.
It turns out that TrueType fonts are more expensive to produce in high quality than Type 1 outlines, because with Type 1 outlines, most of the hinting is in the renderer, so the code is only written once; with TrueType, individual fonts have bytecode instructions to do hinting, and it's different for each font.
OpenType lets you embed both Type 1 and TT outlines in the same font file, along with metadata for supporting lots of languages. So if yuo use Type 1 outlines, you avoid the Apple patent on TrueType.
One way forward would be to gather enough money to pay some font designers to make some new fonts. Another way would be to make a one-time payment to buy rights to existing fonts. Probably best would be a mixture: start with existing fonts and extend their Unicode coverage.
What would a Free Font be? Probably we need something slightly different from the GPL. In particular, it might not be OK to redistribute a modified Free Font without making clear that you have changed it, because otherwise you could reduce its quality or destroy the artistic integrity of the design, and give the artist who designed it a bad reputation.
Font *outlines* (i.e. the design of a typeface) are protected by copyright outside the USA, because they are recognised as artistic works. In the US, they are not protected, for historical reasons. In both cases, the font *names* are often registered trademarks, so you see Palladium because Palatino is a trademark, I think of Linotype; Dutch instead of Times (Monotype), Swiss instead of Helvetica, and so on.
This means it's not OK to start with existing designs, unless they are old enough - e.g. using the original designs of William Caslon from the 1720s is OK, using Adobe Caslon is not OK, at least not without permission.
So, we need type designers to give permission, or to make new designs.
We need more work on the FreeType Type 1 support, so that we don't have to worry about the software patent on TrueType rendering.
We need an independent legal entity so that designers have someone to negotiate with, and so that money can be paid to them. Maybe the Gnome Foudnation or XFree86.org would do, as long as the fonts can be used with any software, not just Gnome or the X Window System.
I do not have enough time to do a lot of work here, but I *am* willing to help introduce people to font designers and other resources, and to help explain the technological issues.
Hacking on a font renderer takes serious skill, as does designing fonts. But maybe programmers can contribute to FreeType, and to pfaedit (how about a Gnome port, too?) and to ghostscript. Programs like Mandrake's FontDrake can be worked on (it's GPL'd I think).
Who wants to help build a font portal, somewhere people can download Free Fonts from, and with links to font designers who can help customise fonts, and to non-free fonts you can buy?
Who wants to donate a server and some bandwidth?
Set up a mailing list?
Remember, we need fonts that are Free, not just ones that don't cost anything, and we need high quality, and support for lots of languages.
If you read this far, my thanks, and let's make something happen. Post here, or feel free to send email [liam at holoweb dot net, will work]
Liam
Live barefoot!
free engravings/woodcuts