RSSMozilla aims to revolutionize Web layout with new Firefox font support
By Scott M. Fulton, III | Published October 21, 2009, 5:54 PM
One area of Web standards where both Mozilla Firefox (version 3.5.3 CRPI: 7.34) and Opera (version 10 CRPI: 6.38) have an edge over Google Chrome (build 3.0.195.25 CRPI: 15.85) is in the field of page-designated font rendering. It's where the code for the Web page specifies which fonts to use, and even triggers the downloading of those fonts where necessary. Actually, Opera 10 has led the way in scalable Web fonts support although Firefox 3.5 has followed close behind.
The problem here has been with the extremely proprietary nature of the fonts used for the Web. They actually are TrueType and OpenType fonts, the majority of whose licensing prohibits their use for anything other than installation in commercial operating systems on a per-desktop basis. Even though some typographers have created free renderings of their commercial font products (here's a favorite of mine: Museo Sans by Exljbris), there's some question as to whether type designers are technically allowed to use the proprietary underpinnings of font technology (mostly contributed by Adobe, Microsoft, and Apple) for use on the Web.
Now, in an effort to resolve this little dilemma, Mozilla is announcing that forthcoming daily builds of version 3.6 (presumably the Beta 2 Preview editions) will support CSS3
@font-face embedding using a font format that is not TrueType or OpenType. It's being called Web Open Font Format (WOFF), and its purpose is simply to repackage the same spline data that appears in TrueType and OpenType font files, in a format and with licensing that's tailored exclusively for use on the Web.
Leading the move toward WOFF is Mozilla contributor Jonathan Kew. In a document Kew co-authored for the W3 Consortium, he writes, "A WOFF font file is simply a repackaged version of a sfnt-based font in compressed form. The format also allows font metadata and private-use data to be included separately from the font data. WOFF encoding tools convert an existing sfnt-based font into a WOFF formatted file, and user agents restore the original sfnt-based font data for use with a Web page." (Here, "sfnt" refers to a specific type of spline geometric data, and is a term based on how early Macintosh systems tagged spline data appearing in OpenType files.)
The metadata will enable foundries to include licensing information which could, for example, restrict a downloaded font's ability to be used anywhere on the user's system except on the page that triggered its download. Or, it could enable an open font intended to be used multiple places, to be shared freely.
What WOFF could also enable -- as part of a future wave of developments that could be ready for Firefox 3.7 next year -- is for Web designers to use the variations that are present in font files, but which HTML and even CSS have never directly covered: Many fonts include hints for how renderers should typeset small caps, and how it handles ligatures -- like connecting capital "T" with small "h," or small "f" with small "t." Without a mechanism in place for addressing the special capabilities of many fonts, CSS can't get a handle on them.
So Mozilla engineers proposed an amendment to CSS just last June 29: a new property called font-variant that enables exclusive properties of embedded fonts to be declared outright. In a build of Firefox 3.6 that was altered by Kew for testing this feature, he was able to produce the lavish typographical poster you see here, using a font called Megalopolis by Jack Usine that's rich with ligatures, using only HTML.
With full font features enabled, diacritical marks, monospaced numerals, and capital letters and descending lower-case characters with sweeping swashes only where applicable (like the beginnings of sentences), would all become addressable by browsers, making them effectively the typographical layout engines that engineers always knew they could be. The result could be a Web where usable text may appear as clearly as on the printed page.
The biggest hurdle the Mozilla engineers may face with this feature is a familiar one: contending with Microsoft's Internet Explorer trying to implement the same feature, but in a different way. As Mozilla contributor Christopher Blizzard acknowledged in a blog post yesterday, "IE currently tries to download all fonts on the page, whether they are used or not. That makes site-wide stylesheets containing all fonts used on site pages difficult, since IE will always try to download all fonts defined in @font-face rules, wasting lots of server bandwidth."
This will be a good thing as long as every browser does it using the same code. I don't want to write one way for IE and another for Firefox.
Score: 2
|Great new feature.
Score: 2
|*laughing*
Microsoft designed this for IE4. Embedded OpenType [Wikipedia].
Yeah. Internet Explorer 4.
It was submitted to the WC3 for CSS3 in 2007. Again, as a standalone submission in 2008. The "innovation" mentioned above is the result. All a part of Internet Explorer 4, released in 1997.
Funny how the article totally fails to mention this. *grin*
(Should have waited to post this until all the MS Trolls had come out of the woodwork decrying Microsoft's failure to keep up...)
Score: 3
|Yep, I personally wrote a web app on Nov. 16, 2000 (I'm looking at the files right now) which needed to render a wide report in Arial Narrow on an intranet where many desktops lacked MS Office and therefore didn't have Arial Narrow installed nor licensed to be permanently installed. The "Installable Embedding" mode was enabled in arialn.ttf so I used Microsoft's WEFT to repackage it as an *.eot file and I added a stylesheet with the @font-face rule. Worked OK, but I never used the feature again after this project because Microsoft never garnered enough support from the W3C nor from web developers to make it worthwhile for Microsoft to improve the tool support and simplify the whole process.
In fact, Microsoft originally submitted it to be part of CSS1 on April 19, 1996. The CSS3 submission in 2007 and standalone submission in 2008 were just two of many resubmissions. In 1996, typeface foundries strongly opposed it on idealogical (profit-driven) grounds and web developers opposed it due to various IE4 shortcomings. My project just needed a simple tabular layout but designers who wanted to do serious layout work where things like kerning and dropcaps actually matter were not impressed. Some of these criticisms were really the fault of CSS' primitive state in the late 90's, but the bottom line was that OpenType Embedding was an innovation that was simply way ahead of its time.
Mozilla's gripes about IE8's lousy @font-face implementation are completely accurate but they don't point out that IE8's implementation has hardly changed since IE4, so Mozilla is really comparing a browser they will release in late 2009 with the browser which Microsoft released in September 1997. I would be shocked if Mozilla's implementation wasn't a LOT better, after having 12 years to think about it.
Score: 4
|Yeah, I used primitive embedded font support in my corporate website back in 1998, a dark age when I was contractually forced to work with IE-compliance in mind and forget about Navigator. The feature was promising, but its awfully buggy implementation gave me headaches for days.
So +1 for you getting the historical facts almost straight, and -1 for you trying to sound like a pro-MS smartass.
Score: 1
|Embedded OpenType. Yes, Microsoft developed a way to embed OpenType way, way back.
And that's the problem. You can embed OpenType fonts, but their licenses issued by their foundries don't always permit them to be shared. WOFF is the possible solution, a way to take the same splines in the OpenType font and repackage them in a way that _they can only be used on the Web_. This way, they don't violate the license.
Which...gee, that's what the article said. And I thought you used to read them all the way through.
-SF3
Score: 3
|So the total lack of mention that this was originally done by MSFT way back in '97, or that this is in fact, the direct result of their multiple submissions to the WC3 was unnecessary and in no way a failure on your part to actually provide detailed, relevant information regarding the tech?
But the need to bash IE for "not doing it right"...that was necessary and relevant?
Throughout the entire article, Mozilla is the good guy, even Opera...and the only mention of Microsoft? Negative snark regarding how they are "doing it wrong".
Credit where credit is due, dontcha think? They were "doing it" before anyone else. Long before...and the only reason Mozilla and Opera even can is because of what Microsoft did in '97.
But heck, leave that bit out. Wouldn't want to be called Microsoft shills, now would we?
Score: 0
|LOL, you're so wasting time on these drones, Scott. Their contorted logic and selective historical presentation says it all, and that's why their such entertaining little trolls. You have to remember that most of them have a vested interest in Microsoft's success. Sadly, many of them have such empty souls that they end up worshiping the company..which is ultimately to be pitied.
Score: -1
|Scott,
PC_Tool has a very low reading comprehension (he watches too much Faux News). Microsoft did not do this in 1997 or any year. Very different things.
Score: -2
|"Their contorted logic and selective historical presentation"
Baseless accusation with no example or objective data to back it up? Check.
"You have to remember that most of them have a vested interest in Microsoft's success."
Insinuation of being a paid shill, again with no source or proof? Check.
"Sadly, many of them have such empty souls that they end up worshiping the company..which is ultimately to be pitied."
Emotional rhetoric designed to engender a aggressive emotional response? Check.
You sir, have trolling down to an art-form. I bet mommy is so proud.
Score: 2
|Scott, it remains to be seen whether the WOFF implementation will be challenged by foundries. I can't speak for PC_Tool, but as I've said, I'm not just blowing hot air based on skimming some articles; I personally used EOT and still have all the code and files, including hex dumps of TTF headers. I know first hand that Microsoft's solution included multiple provisions designed to address license issues, and when I look at what WOFF is doing, it's really not a great deal different in terms of being able to respect all the wishes of every foundry, given that some of those wishes may be unreasonably restrictive of "fair use." Some foundries like Bitstream are more progressive and generous, so they will tend to be lenient and easily satisfied. Others like Monotype are going to be more protective of certain typefaces which cost them a great deal to acquire or develop. In some ways, WOFF's approach is a little like The Pirate Bay claiming innocence because "no actual files are hosted on our servers." Being technically true didn't prove to be an effective defense for The Pirate Bay, and I doubt that WOFF's approach to repackaging will be seen as substantially more defensible than EOT's was.
Score: 2
|It's obviously a next step since so many have been pushing the use of applications through the browser.
The problem is getting popular typefaces into patent-free formats that don't ruin them. The majority of zero cost or low cost font files is that they aren't (exceptionally) precise and for the majority of consumers, it's not been a problem. Designing for the web is still a moving target, despite standards.
Score: 2
|Actually, IE has been supporting font-embeding since version 4.0
http://msdn.microsoft.co...VS.85,lightweight).aspx
Score: 1
|You are right.
They submitted this to WC3 in 2007 and what you read above is the result. :)
Score: 0
|Mozilla; Lets just invent our own standard and call them web standards, BAHA, who's going to argue with us? and later blast everyone else for not following the standard, its genius
while its a good idear, Mozilla, Microsoft, Google, Apple along with the Community, COULD agree to get together and talk about these proposed standards...
Score: 3
|You're confusing Mozilla with Microsoft.
BTW the WOFF is already supported by a couple dozen organizations, actually Mozilla is surprisingly late to the party: http://blog.mozilla.com/...ts-web-open-font-format/
As for Mozilla's proposed CSS extension, that's all it is, a proposal, to W3C who actually writes the standards for CSS.
It truly is tough to develop webpages with such a limited selection of fonts you can be sure all users have, thus web fonts are very useful. This may help web designers further refine the look of their pages' fonts.
Score: 3
|Didn't we already go through this kind of scenario with Microsoft and it's Internet Explorer?
Score: 0
|Yup. We did.
Apparently, Microsoft was ahead of it's time. *wicked grin*
Score: -2
|"artfuldodga" wrote :
Mozilla; Lets just invent our own standard and call them web standards, BAHA, who's going to argue with us? and later blast everyone else for not following the standard, its genius
------------
"The MAZZTer" wrote :
You're confusing Mozilla with Microsoft.
------------
Actyally he isn't when Mozilla (hence Firefox) became a player in the browser market, IE had a market share like 95% of the market, making anything it had a DE FACTO standard.
So instead of implementing things the way the marked expected... Well, I always rant about this.
I remember writing about this on my blog way back (http://pjondevelopment.5...orer-firefox-opera.html and http://pjondevelopment.5...e8-and-acid-2-test.html). People tend to forget history quite easily, specially when it's Internet history.
Score: 2
|