Nameplate set in Sutro Deluxe and Initials from Parkinson. Your typeface could be next. Learn more.
Ads via The Deck

“Embedded” Web Fonts Return. Uh-oh.

Stephen Coles on October 4, 2007

The web is about to become more beautiful/hideous. Dave Hyatt of the WebKit engine used by Safari:

WebKit now supports CSS @font-face rules. With font face rules you can specify downloadable custom fonts on your Web pages or alias one font to another. This article on A List Apart describes the feature in detail. All of the examples linked to in that article work in WebKit now.

For the uninitiated, this means any TrueType font can be called by a style sheet and then downloaded by the web browser. This reopens the legal can of worms that falls off the shelf every time we talk about font embedding. Good fonts cost money. Like most software, each user or CPU must be licensed to use a commercial font. When you start talking about every visitor of a web page downloading the font — well — you enter very sticky territory indeed. Each foundry has their own end-user license agreement, and many of these EULAs mention embedding in general, but the font industry hasn’t specifically addressed the issue of CSS @font-face rules. It’s a tiny, toothless industry and, as usual, they are a step behind.

So, due to the professional type designer’s understandable desire to be paid for their work, most websites that will take advantage of this new technology will use free fonts. Cue the foreboding horror music.

In general, web designers aren’t typographers. Their specialty is in the realms of interface, hierarchy, and navigation. Their training does not include making decisions about what typeface to use for long passages of text. This is not so much an issue today, where HTML text is limited to screen-optimized fonts crafted by experienced type designers. But now web designers have a sea of crummy fonts to choose from. I’m afraid this does not bode well for readability and aesthetics on the web.

Update — John Gruber, via email, succinctly describes the dilemma:

The fonts you’re allowed to embed legally aren’t worth using; the fonts that are worth using aren’t embeddable.

Update: May 3, 2010 — Lots of stuff has gone down.

See also: Initial Discussion at Typophile : Typophile Webfonts Discussion in 2009–10


  1. Si says:

    I think it’s a mistake to call linking raw fonts to Web pages “font embedding” as the fonts are in no way tied to or embedded within the document file. This is plain and simple font redistribution, which as noted, is not allowed by commercial font licenses, which tend to allow certain forms of embedding.

  2. jennyy says:

    I mean no offence to the typographers who might be reading this, but commercial typography (ie,computer fonts) is the biggest scam since Amway. ($9000 for Adobe Font Folio? Really?)

    Hopefully this development will get people annoyed enough with ugly free fonts that the foundries will start giving away some quality ones.

  3. akatsuki says:

    It would just be nice if we could at least have a “gentleman’s agreement” around a core group of high-quality, free fonts (which do exist, I think). They could be held in a repository somewhere with a link embedded on sites to that repository.

  4. marlwin says:

    Now I understand why examples from A List Apart were so crappy.

  5. Danny Dawson says:

    In general, web designers aren’t typographers.

    I think that this statement is not exactly true. True “web designers” generally have a strong background in the traditional aspects of design, such as typography. This is true of almost every colleague I’ve met who works as a professional web designer.

    The real worry is a resurgence of the geocities crowd of yesteryear, and the professional employment of much of today’s myspace crowd. Thankfully, horrible design tends to congregate and make itself easy to avoid.

    With every new technology comes the power to make great things and awful things. @font-face has the potential to revolutionize the type industry, if only designers drive for the change to occur. Expect to see more refined, Creative Commons-licensed fonts. Expect to see new foundries who specialize in embedding licenses running circles around the monolithic type houses, until the big boys can catch up.

    Of course, it also will make for some ugly blogs. All in all, it’s just words and pictures.

    (In case you were wondering, I self-identify as a “web developer”.)

  6. Matt Pennig says:

    A can of worms, for sure. I think if browser vendors do their stuff right though, font designers will still be happy.

    For example, WebKit could keep fonts strictly in memory and discard them when the page unloads. Downloading a font — which is technically the same thing as creating an unlicensed copy — doesn’t necessarily need to be a scary thing. There are plenty of measures browser vendors can take to prevent that download from getting into the hands of dishonest people.

    That being said, aAnything that makes it easier for irresponsible designers to wreak havoc is a scary thing.

  7. craig bradley says:

    It’s absolutely inane that this is an issue. Saying that everyone who views a web page has to buy the font should correspondingly imply that everyone who views my logo should have to buy the font.

    I realize that browser technology means that to view the font you have to have it, and that therein lies the problem, but still, for this to have legal strangleholds is absolutely insane.

    This requires an engineering fix, I’m afraid.

  8. Steve says:

    I don’t know all the legalities, but from what I gather, the font information is downloaded in a form useful only for rendering of the given page; it’s not like the actual font *files* are downloaded in a usable form.

    Given that, what are the legal issues? So long as the website owner/user owns the font, how is this different from normal printing usage of fonts?

  9. Brian Crick says:

    Never commented here before, but here are my thoughts:

    I agree that most web designers today aren’t typographers, but that doesn’t mean they never will, or never should be. The average graphic designer needs a wide variety of skills to make good works, with a good handle on everything from color to composition to typography, and as the web matures, I suspect that more and more web designers will be expected to have those skills as well.

    Unfortunately, I suspect that the tools to make bad typography need to be out there before the market decides that it needs people who can use said tools well and EULAs that permit that to happen.

  10. Todd Draper says:

    I would think there is a lot of goodwill and good publicity to be gained by someone releasing a few nice embeddable fonts. Maybe with a license that requires a tag and link somewhere on the site?

  11. PapayaSF says:

    When Explorer and Firefox support this, then we can talk. Until that happens, what web developer wants to bother setting this up for only Safari users, along with an alternative display system for everyone else? Not me, and I say this as a Safari user.

    My proposal is for Microsoft and Apple to chip in a little money and license more fonts for free distribution with Windows and OS X and Linux. Wouldn’t it be great if we had two or three times as many “safe web fonts” to use? Especially some condensed and light fonts that look good in large sizes.

    And when are browsers going to support all of the CSS font-weight properties?

  12. Matt Wilcox says:

    While your arguments are true, they also rather miss the point. Because your arguments are true for the moment does not mean they are true for all time. Your point of view projects the current situation into a future time frame, and lo-and-behold the future looks full of bad fonts, bad typographers, and illegal distribution of someone’s property.

    The thing is any new technology has the odds stacked against it. You only need look at MP3 players and how they were received with horror by the industry, and scepticism by CD distributors. Turns out the reality is rather different than they expected.

    What will happen, should web-font support appear in IE and Gecko too, is a shift in the font industry, and a shift in the skill-set of the designer. Why do you think web designers don’t know about typography? Well plainly because the tools are not there for good typography on the web. You don’t learn what you can’t use! Once good typography is a possibility on the web, web designers will learn good typography. It’s only what happened with XHTML/CSS – at first there was a swath of poor sites and abused CSS/HTML – but then it all changes as people learn the skills required to use the technology well. Those that don’t learn don’t survive.

    And will type foundries really miss out on the opportunities here in the same way that closed minded copyright obsessed record companies did? At first, probably, but in the end – no. They will get some good free fonts out (my guess is with limited charsets) – and use them as springboards for other revenue streams.

    I have never understood people who argue against or moan about ‘value add’ features. Web-fonts will not make the web ugly – poor designers will. No technology makes anything happen – it’s the people that use (or mis-use) it that make things happen. And what happens to poor designers when web-fonts are in-vogue? They either learn and get better or they change career – same as any one else in any other industry that can’t use the tools at their disposal well.

    Get ready for some ugly fonts and copyright spats in the short term – but a far more beautiful web and a rejuvenated typography business model in the long term.

  13. larkost says:

    My bigger worry is about security and stability. Because fonts are such a low-level thing that takes a bit of processor time they are done with very little security on them. They would be a great place to hide a trojan.

    As an illustration of why I think this is possible, the first thing I always do on a Mac that is behaving “wonky” is to have FontBook take a pass at validating the font files. Many times I find a bunch of corrupted fonts (or more often ones that were broken right out of the box), and removing them and clearing the caches solves the problem and the computer feels faster.

    And this is just with what users decide to manually install themselves. Just imagine what is going to happen when the font system has to chew on fonts coming from any web-page that decides to send them.

    This seems to me like a big hole that we are going to be tossing good developer time into to keep it from becoming a security and performance nightmare. And it is probably going to have to be re-evaluated for every single platform that WebKet gets ported to.

  14. Skippy says:

    For the legal aspect, I don’t see how this is different from offering any other type of content on your website. If you break copyright you do so at your own risk.

  15. Stanley Seibert says:

    Proper mathematical typesetting is one case where embedded fonts would have been useful years ago. It has been incredibly difficult to render math formulae on the web without reverting to embedded pictures which can’t scale or flow with your text.

    Interesting hacks like jsMath, which uses Javascript to try to render math with a mixture of native fonts and image-based fonts, could benefit from embedded fonts. This package generates really nice looking output in Safari if you have the TeX fonts installed, and acceptable output otherwise. If it could deliver the TeX fonts to people without them already installed, that would be fantastic.

    Of course, since embedded fonts are just now getting added to the development version of WebKit, it might be a while before it shows up in a released browser. But if Safari does start supporting this, you can bet I’ll be hacking jsMath to deliver the TeX fonts to compatible browsers.

  16. Jacob Rus says:

    @Larkost: Dave Hyatt noted on IRC that if fonts are really an attack vector, then having a web browser open PDF files with embedded fonts should make it equally vulnerable; i.e. no new attack vector is being exposed here that wasn’t already.

  17. Jacob Rus says:

    @Stanley: ISTM that SVG-as-images in HTML (once webkit and other browsers support that, and support embedding fonts in SVG) is really the way to go for math formula layout on the web (instead of some jsMath thing), once such output can be automatically-generated from LaTeX.

  18. Steve:

    I don’t know all the legalities, but from what I gather, the font information is downloaded in a form useful only for rendering of the given page; it’s not like the actual font *files* are downloaded in a usable form.

    Sorry the previous comments weren’t published before you got yours in, Steve. As Si noted in the first comment, the fonts are actually downloaded. So redistribution is definitely part of the licensing problem.

  19. Jeremy says:

    Yeah, the legal issues here need to be fixed. Downloadable fonts are a really great thing.

    If I have a font legitimately, I can (if the designers allow it) embed it into a PDF that I distribute. I should also be able to do the same with a web page.

    Too, as stated above, the font industry is badly in need of being shaken up. Their prices are absurd and strongly encourage people to obtain the fonts via less than legitimate means.

  20. Si says:

    >Dave Hyatt noted on IRC that if fonts are really an attack vector…

    If Dave is reading this he should have a chat with Peter Lofting (Apple font guy) on this subject.

  21. Illtron says:

    The concern is silly. If you’re going to steal fonts, you can go out and do that today. Hell, I was given floppy disks full of fonts back when I was using a Mac Plus. If you’re inclined to pay for them, is making them slightly easier to steal going to turn you into a thief?

    Luckily, 99% of the ignorant public think Arial is awesome, and doesn’t give a rat’s ass that they can steal a copy of Helvetica, even though it’s sitting right in front of them.

    And even if they did decide to dig through a site’s CSS files and find the path to the embedded font files, is somebody losing money when they steal something they wouldn’t have paid for anyway? It’s not like somebody’s taking a loss on the cost of raw materials. Honestly, of all the people you know who aren’t designers, how many of them have ever bought a font, or even downloaded a free one that didn’t come on their crappy PC?

    Move along; there’s nothing to see here.

  22. johno says:

    MySpace users will love this.

    Although, it’s a little scary to envisage tens of thousands of free fonts finding their way into a web page near me, I don’t think this should lead one to the conclusion that web embedded fonts is a bad idea.

    I think Richard Rutter’s proposed solution is one that could work.

  23. Hrant says:

    If you shake up the font industry any more, the scant amount of good fruit left will just fall off and rot, leaving only the inedible stuff! :-/ Most fonts out there are in fact priced very low when you consider how long they took to make. Sure, that doesn’t mean one can expect too many people to pay anything for something they can swipe for free with impunity, but the other side of the coin is: don’t expect too many people to make good fonts if it’s so cost-ineffective to spend the time doing so.

    That said, this is a complex situation, with no obvious ideal solution. But ignoring the need to pay people enough to maintain their seemingly small but nonetheless central participation in a whole will most certainly result in a weaker whole. Ergo: be careful what you’re saving money on; you might realize too late that you’ve been pennypinching in the wrong direction.


  24. Ian Ragsdale says:

    Is it too much to hope that font foundries will see this as an opportunity? I’m not a typographer (although I have used Fontographer a few times), so they will probably see it differently, but as a web developer (designer would be a big stretch) I don’t see the point of using any non-safe font at the moment. If I have to create images everywhere I want to use it, its too much of a pain. However, if I could purchase any font I want and easily use it on a page, I’d be much more likely to spring for a non-standard font. Am I the only one that thinks this could actually raise sales of fonts?

  25. Blain says:

    I don’t know the answer, but I do have to ask this. What’s worse for the industry: downloaded fonts, or everyone simply using Helvetica, Arial, core fonts, and nothing else?

  26. Gilbert says:

    The industry is going to have a year or so to sort this one out as in practice there aren’t any browsers out there capable of supporting this yet. The capability in WebKit doesn’t automatically translate into capability in Safari. Opera will be next probably. but again that’s not exactly going to change things. The big question is who out of IE and Firefox will be the first to implement it.

    Microsoft tried this in IE4 but didn’t use an open font format and later removed it, but does that mean it can be quickly reimplemented using the same standard as WebKit. Will this now become a priority for Firefox?

  27. Lanny Heidbreder says:

    As someone who consciously turns off text antialiasing at small sizes, I dread the day when any font can be used on screen, even the good ones.

    Every good, respectable Adobe Pro font that can positively exude craftsmanship in InDesign looks like absolute crap in web copy. They don’t hint those fonts worth crap, because (until sometime in the nearer future) there’s been no prospect for seeing them aliased.

    I know I’m like the last Windows guy on the ClearType bus, and possibly the most type-conscious guy still looking at aliased text anywhere, but I’m sticking to my guns until they force blurry text on me technologically, or else double or triple the ppi of monitors. And until most good typefaces get good hinting as well, you can keep your @font-face.

  28. Patrick Hall says:

    Unless I missed a comment, no one has mentioned what I personally think is the single most important application of font embedding: making languages that don’t use the Roman script more accessible on the web.

    A simple example: when the Amharic Wikipedia added font embedding to the default theme (as well as a Javascript keyboard input method), contributions went way up. Everyone has a right to type and read in their own language, but most big companies could care less about relatively small linguistic communities. Embedded fonts give those communities a fighting chance.

  29. The ball is in the copyright holder’s hands. I don’t understand why this isn’t obvious to everyone: license your fonts on a domain-name basis. Make it really easy to buy these licenses : a website were you can handle all the licenses you have, etc. If you need to, issue serial numbers and embed them into the files, to rat out the pirates. DRM never works, don’t even try. Make it really easy to license the fonts for the web, don’t make them too expensive, and you’ll be making much more money than you are now.

  30. Dave says:

    The problem is not zero price fonts, it is proprietary fonts available at zero price. If the public has freedom to improve fonts, such as by adding hinting instructions, then the quality of typography on the web can go up a lot. The Open Font Library is slowly developing as a central resource of such fonts that come with “software freedom”. And FontForge is a font editor also available as free-as-in-freedom software.

  31. Joe Clark says:

    You need to separate the objective argument of “This will violate license agreements” from the opinion that Web designers suck and will screw this up if we give it to them.

  32. Chris Rugen says:

    I don’t see the threat here, in terms of web readability. The web is already filled with illegible/unreadable drek, and allowing poor font choices will likely only affect those who already do a bad job of it.

    If anything, opening up the world of fonts even a little bit will create a push for more quality. Perhaps it will motivate larger companies to bundle more web-friendly, open-licensed fonts. This can, in turn, pique a user’s interest in quality fonts that require a fee (not for the web, but in other arenas).

    I agree that those who use pirated fonts in professional work are already doing it, and making it ‘easier’ won’t turn those of us who behave ethically and legally into patch-wearing font pirates. FontShop is a good example of a seller that understands the value of a few freebies to draw consumers and sales. It’s just the way of the web.

    I could go on and on…

  33. gillsans says:

    I’m with Ned. I think most of the comments are thinking that the current model is best and should continue in the future. Why can’t there be a new font licensing agreement where you can make the font available for a website to embed? Does anyone really think that the web or electronic content is going to regress or go away? There’s no reason this can’t be a win-win. Technology is only going to improve and there’s a big opportunity to cash in. Type foundries might as well figure it out now before it’s too late.

  34. Ouch. Saying web designers are poor typographers is like saying print designers don’t know how to design for the web. You paint with too broad a brush.

    The type foundries are like the RIAA, they are stuck in an old-school mindset of how type designs should be sold and protected. I am convinced that if they better understood the technologies and worked more closely with the web standards bodies and development communities, a solution could be found that would satisfy everyone. For example, why should a font be any different than a runtime-shared-library that gets downloaded if it is needed? This happens in Flex for example and these libraries can be every bit as proprietary and complex as a font. Yet this model works.

  35. Randy, did you miss the words “in general”?

    On your second point, I agree. Font foundries need to start thinking about creative ways to deal with this, and quick. As others have mentioned, this could be as much an opportunity as it is a dilemma.

  36. Julik says:

    Unfortunately what you are saying is a total FUD.
    Font industry had 10 years (I repeat – 10 years) to sort out the embedding conundrums when the feature first appeared in Navigator, then in IE. The font industry had 10 years to determine how a subset format could have been developed to circumvent the reuse of fonts that could have been downloaded. The font industry had 10 years to develop the same robust technologies that now power PDF font embedding and work, but for the web.

    The font industry preferred to leave the website decoration to the fellas at Microsoft, a result of which is a very _ugly_ If it is now time to face the consequences — the better, because the greed vector is surely depleted some. And if that will be free, non-foundry developed fonts — even better. It might be not the work of a foundry that cares about making a cut (see it as labels vs. indies situation) but it will be an individual’s work.

    I think the font industry that you are talking about so glamorously just wants to put itself into the “white princess” situation. I am 800% happy that this attitude is going to be punished, albeit with a delay of 10 years.

  37. Julik says:

    Yes – to add something:

    The training of good designers does include making decisions about what typeface to use for long passages of text, and it’s a total insult to a whole profession to make statements like this one.

  38. Why can’t there be a new font licensing agreement where you can make the font available for a website to embed?

    That’s not the problem. I would start licensing fonts tomorrow for websites if the font-face command would be implemented in a different way. Right now the font file has to be available for download to work with font-face. So if you put a font on a server everyone can start embedding it, download it and install it. So this is would be like giving away all your fonts for free.

  39. Tom says:

    I find it interesting how skewed these comments are toward the issue of illegal downloading and reuse of typefaces. Surely this is a relatively minor issue, given the ready availability of pirated faces. The main issue is royalties for the legitimate uses of the faces on the sites themselves.

    I wholeheartedly agree that there is absolutely no technological barrier to creating unextractable rendered faces, so the lack of progress must be down to either indifference or even disdain for the medium on the part of the typographic community.

  40. Nicole Dotin says:

    It should be kept in mind that much of the ‘font industry’ is made up of one and two person teams that simply don’t have the resources to tackle an issue like this from the development side. Many type designers would love to see a new revenue stream open up (who wouldn’t?), but simply don’t have the money to develop the technology – which means that while they might have a voice in the matter, they won’t have any control over final implementation. If type designers are cautious or apprehensive about what’s happening now, I imagine that is why.

    It’s one thing to say a certain technology should be developed, it’s another to actually make it happen – across all platforms, all browsers, etc. That’s not a simple thing for any company to do, so if we’ve left it up to the ‘fellas at Microsoft,’ it wasn’t by choice. Well, anybody with a stake in this issue can front the money for continued development, so, who’s got cash? ;)

  41. I am amazed by some of the comments here.

    The font industry had 10 years to determine how a subset format could have been developed to circumvent the reuse of fonts that could have been downloaded.


    I wholeheartedly agree that there is absolutely no technological barrier to creating unextractable rendered faces

    Ha. Ok, some education is needed here. So you want to make a system in which the user can download the font to render a webpage but not be able to use the font for anything else. Such a system has a name, it’s called DRM (digital rights/restrictions management). To put it simply, it is not possible. Not hard, not very hard, not extremely hard, simply impossible. The theory is: you are giving both the cyphertext (your font) and the key (some cryptographic key) to the attacker (the end user). For a nice explanation of this, read Cory Doctorow’s piece in the Guardian.

    So, dear font designers, font distributors, etc, don’t make the same mistake as the recording industry. Don’t treat your customers like criminals. If you move fast, you could considerably enlarge the market for fonts.

    Off the top of my head:
    - create a company (or even a non-profit organisation) that will be a licence broker.
    - that company sells downloadable fonts. A downloadable font is tied to a domain name. A domain name can contain wildcards for sub-domain, and path fragments (that means “*” or “”). To tie both together, you do a cryptographic hash of the domain name, the user’s account identifier (probably an email address) and a secret pass phrase. You then embed that hash inside the font file.
    - so boom, you’ve just solved the distribution problem.
    - now you solve the problem of encouraging people to move to the new model: you setup a web crawler (a “bot”) that parses web pages, downloads the fonts, and checks the fingerprint to see if it has been copied from another domain or not. If it has, the owner of the website is infringing on your copyright, so you can then set in motion the usual legal recourses against copyright infringers. Unlike current font pirates, these pirates are out in the open: it’s very easy to check that the font they have on their website is pirated. So no reputable company would want that.

    To those who say: “this is too much work, it’s too high tech for the poor font companies”, this system could be created in a few days by one developer. It’s not difficult. Please do it so we can buy lots and lots of your fonts.

  42. Tom says:

    Ned, surely my assertion:

    I wholeheartedly agree that there is absolutely no technological barrier to creating unextractable rendered faces

    Is the same as yours:

    This system could be created in a few days by one developer. It’s not difficult.

    After you educated me, you came to the same conclusion as me.

  43. Tom, sorry if I came off arrogant about the “education” part.

    Maybe I misunderstand you, but you talk about creating “unextractable” rendered faces, and that is precisely what I’m saying is impossible to do. There will always be a way to extract font information, and recreate usable font files from it.

    The ‘system’ I suggest is just an efficient way to distribute DRM-free downloadable fonts.

  44. nick shinn says:

    I’d like to second Nicole’s observation.
    The Font Industry in reality comprises three quite different kinds of “foundry”:

    1. Apple-Microsoft-Adobe
    These are a massive transnational oligarchy of companies that make their money off making computers and operating systems. They also invent font formats and have small internal boutique “foundries” that design and manufacture fonts. Because fonts are necessary to their core business, they have made things simple for themselves by developing easily-copied (i.e. poorly protected) font formats, and bundling fonts with their main products. This has lowered the value of fonts and made them items of easy virtue. However, it has nonetheless created a large, diverse and growing font market — both professional and consumer — and enabled an independent sector of hundreds of small foundries.

    2. Legacy foundries: Linotype-Monotype-Berthold
    These are not companies owned by type designers. Although they do develop new typefaces, they are characterized by multiple product lines of intellectual property, dominated by revivals of fonts designed long ago which have already recovered their development costs (Helvetica etc.) They can operate on the Wal-Mart model of low margin on high-volume sales.

    3. “Indie” foundries.
    These are businesses owned by type designers, that develop and market new type designs. As has been noted, we do not have the economic critical mass to have any say in the development of font format software — we work with the technology that Apple-Adobe-Microsoft decides. The retail font market does not provide a very good source of investment capital for the development of new product — many of us rely on commissions from periodical publishers to finance new type designs. It should be noted that while Adobe and Microsoft fonts are distributed by the million, indie foundries make their income off retail products that sell, annually, by the dozen. In recent years, we have paid a lot of attention to developing fonts in the complex and (for us) sexy OpenType format, when perhaps we could have spent more time considering business-technology issues such as web fonts.

    These three kinds of “foundry” represent three quite different business models, and their interests are accordingly quite different.

    FSI (FontShop International) and Veer are retailer-publishers, and fit into the picture as well, forming another layer of complication.

  45. Excellent summary, Nick.

  46. djurban says:

    It is unfortunately the designer’s fault. The times are changing, maybe it is the time to change the way you earn from fonts?

    Maybe allowing the usage of fonts for personal non-commercial use for a small price is not that awful? Imagine a $5 font for non-commercial personal use, just for personal users. No DRM btw, it’ll get cracked and it pisses the client.

    And foundries could still earn big money from commercial clients which are their main clients and which generate their revenue. It’s simple, no commercial company could violate the licence because it would come out, it always does, and the violation fees are big.

  47. djurban says:

    Let me just add one more thing. The market of screen-optimized fonts is wide and open. And the industry is closing their eyes for it. I’ve had situations while designing websites that the company (media company, owner of several newspapers) was ready to really interesting money just to get a screen optimized typeface that would allow them to provide their content on the internet with style.

    They’re using Georgia now, but are completely unhappy because they want a web typeface that relates to their print typefaces.

  48. This is not so much an issue today, where HTML text is limited to screen-optimized fonts crafted by experienced type designers

    Does this include Comic Sans, one of the 10 fonts on the safe list?

  49. Bram Pitoyo says:

    You know, if only sIFR can be made to be easily web-indexable (possible) with relatively light memory footprint (kind of impossible), then we’ll be in business.

    I think that by the time that IE fully adopts embedded web fonts, computers will be fast enough to handle fully-sIFR’d contents and Flash will also catch up in performance.

    Otherwise, you can always go the Sumner’s way :)

  50. Does this include Comic Sans, one of the 10 fonts on the safe list?


  51. nick shinn says:

    It is unfortunately the designer’s fault. The times are changing, maybe it is the time to change the way you earn from fonts…a $5 font for non-commercial personal use…

    DJ, your generalization is a bit out of touch. Many indie foundries are experimenting with differential pricing (and other retail packaging), especially at MyFonts.

    Differential pricing is a de facto aspect of the OpenType format, in the sense that a “Standard” OT font with no typographic features and perhaps no CE support sells at a low price, and a fully-loaded “Pro” font sells at a much higher price. Adobe began this practice, and came up with the “Pro/Standard” concept around 2000; it has now become widespread.

    Further, at least one foundry is offering differential pricing of the same font, with separate prices for private and business use. Canada Type does this, with a price split of 20-35%.

    I think your proposed price of $5 is unrealistic. I doubt that if I reduce the price of a $50 font to $5, that I will get a tenfold increase in sales. The low-margin/high volume model depends on large scale distribution and marketing support, without which it’s commercial suicide. I could be wrong, but I’m not willing to take the chance to find out.

    However, some indie foundries are forging ahead, e.g. Font Garden, which prices around the mark you mention, and their fonts are doing well in MyFonts’ sales chart, so perhaps this is the wave of the future and I’d better get with it.

    Also related is Ascender’s $20 corporate package.

  52. Matt Carey says:

    In general, web designers aren’t typographers. Their specialty is in the realms of interface, hierarchy, and navigation. Their training does not include making decisions about what typeface to use for long passages of text.

    What a huge generalisation! What utter rubbish. Is that an ivory tower I see before me?

    I would suggest that a large number of web designers have had some form of design training. If they have then a large number of those will have had some typographic tuition. That may be as little as one session (though I’m sure it will be much more than that), but even so there will be a certain degree of basic knowledge there. They are not going to be rushing out and setting all their body copy in marker felt for goodness sake.

    What you really need to be worried about are the MySpace generation and the users of MS Frontpage. People who don’t have the design training but are given the ability to design their own pages. When they are given the facility to pick a typeface alongside colours then we are in trouble. I can almost see the MS Frontpage packaging now “Wow. Pick your own typefaces…’

  53. jdoole says:

    I wonder how Canada Type’s separate pricing is working for them. I found it very interesting, and possibly very attractive for some of my customers, but the potential for rampant abuse is there. They took a huge risk with that.

    Font Garden is targeting a very specific customer base and I would imagine you get what you pay for. Scripts don’t have much “staying power” , and I would wager that FG is just a brief flash. Others will come with the same model, but will fade.

    And that’s pretty much the bottom line. 9k may sound like a lot for a font package, but when you factor the design time that went into that work, it’s a steal – as I recall, the average is $3 a font – for the best quality work you can buy.

    It’s disheartening to see that some designers, who essentially sell their creative ideas to eat, don’t have any respect for the creative ideas of other fellow designers.

    The bottom line on embeddable fonts is that the larger players (MSFT, ADBE) just didn’t care to come up with a good solution for embeddable fonts. It didn’t make financial sense.

  54. Matt Titchener says:

    Hmm. Given the number of comments concerning licensing I’m a little confused. While working on the redesign of Boxes and Arrows, font embedding (for IE) came up as a possible solution to a hard problem, namely, indexable custom font headlines.

    During our research it became clear that the WEFT (Web Embedding Fonts Tool) efficiently circumnavigated the licensing issue by creating .eot files that were viewable only – a user couldn’t do anything with the font except display the headline on the screen, and of course it wasn’t even installed onto their machine. How is this is a problem for a foundry? How is this any different from having a “licensed” font on a poster?

    Besides, if a foundry really wants to limit embedding on a page, they can simply place a licence within their font that doesn’t allow .eot files from being created.

    While WEFT might be a proprietary solution, it sure is a good one.

  55. I think I’m a bit slow in coming to this discussion. That’s loads of comments, huh. I just wanted to add my 2 cents worth.

    “In general, web designers aren’t typographers.”
    Many Web Designers may not have training as typographers but this, in no way, reflects a disinterest in the subject. I’ve read many discussions and blogs by web designers and developers who are scoping out what rigorous typography is all about. There have been some great discussions on the matter too, starting here, continuing here

    “Good fonts cost money.”
    There is an obvious solution here. Web pages only really need regular, bold, italic and bold italic versions of a particular typeface. There’s no use for expansive typeface families with loads of weights and styles. Similarly there are characters that will refuse to appear on web pages and are, therefore, not needed.

    So there is an opportunity here for typeface foundries to develop web specific versions of their popular typeface families and because these would be pared down versions of what could be considered “print professional” typeface families, they would naturally be more affordable and easily marketed as specialist tools for web design professionals.

    Dose that makes sense? What do you think?

  56. I think using other fonts can have a major impact on design. When used normally they can improve a website. And of course they can make websites ugly and unreadable, but you can do that now also. Good designers and professionals will use it to improve design. You can also create horrible websites using animated gifs and blinking and scrolling text. But the fact is that if you do you will loose visitors for sure.

    You can already see what sIFR can do for a website design. I’ve created an alternative that works through 100% CSS, so when Web fonts are supported you do not need to adjust your design. It is called True Font Family and you can see it in action here:

    I think it only improves the design because it is mainly used for headers, buttons and menu options. But it is very easy to implement and adjust.

    We’ll see where the future takes us, but I think more options means more options to make the web more ugly, but also more options to make it more beautiful. And more options in the past have only made more possible.

    I understand the copyright issues, but I don’t understand why people are afraid this will lead to impossible sites.

  57. Adam says:

    For those of you who disagree with the use of @font-face, consider the following:

    1) People will always design ugly websites no matter what technologies are available to the masses.

    2) How is this different than the ability to hotlink images? Did the world end when photographs under copyright were able to be linked from anywhere?

  58. Roy says:

    Having been a sign painter for the last twenty-seven years, I always get confused in these discussions regarding font usage and foundry copyrights. And the nitpicking that goes with it.

    When I first started in this trade, I was taught Helvetica, Roman, Script and Casual. When I wanted something different, I bought my letraset catalog and spent hours learning a typestyle that I soon incorporated into my signs.

    Over the years, I’ve collected two bookshelves full of type books. But, upon careful inspection you find there are numerous duplications, some with only the name being changed. Same goes with the digital font packages I have on my computer.

    **You see some web designers actually do know typefaces, even their history. FYI, there isn’t a font out there that wasn’t derived from another, anyone that tells you different has a vested interest in keeping you deluded.

    Personally, I hate designing web pages that only use standard fonts like verdana, tahoma or georgia etc. (there’s just so much you can do with that group) and would love to be able to embed fonts just like my pdf files. I see it as inevitable.

    The prices asked for digital variations is another topic of discussion, as is the creation of typestyles from existing fonts, which I look forward to discussing. :)

    I love this site.

  59. georgie says:

    Sorry, to dig this topic up again, I just chanced across it. This might be a stupid suggestion, but it’s just a thought: why cant the bigger or “experienced” type foundries and designers sell their fonts to browsers, and collect a royalty every time a user has to download a font to view a page. They can also sell their fonts to designers who are creating these pages. There is no point in stopping users from downloading fonts, they may not be in control of what fonts their browser will download.

  60. Bob says:

    As usual you get what you pay for.

  61. swarmofkillermonkeys says:

    Personally, I’m disgusted that the big foundries were able to stomp on this technology for so long. I understand the value of $1000 high-end fonts, I do. But why should the millions on the web, or free font producers like Larabie have to suffer for it? It is philosophically abhorrent, and indicates a lot of what is wrong with western (corporate) culture right now.

    Repeat after me (and Mr. Jobs): theft is a social problem, not a technological one. You know, I hear Steve just might have been onto something there way back in ought-one. What do you think? And in any case, is this more dangerous that p2p downloads? I highly doubt it. And who on earth is going to steal a high end font for things like multiple weights or sophisticated ligatures for a freaking webpage? Those people don’t even do their own layouts, they just use a template that came with their blog software. Grr. Sorry, but this ‘let them eat cake’ idea of restriction for the sake of restriction really chaps my backside, on an otherwise very thoughtful site (thanks!).

    Tasteful use of fonts by amateurs (admittedly I’m not a pro) will always be an issue. We eventually trained people away from Comic Sans in correspondence, we can do the same with web pages. In fact, @font-face is the only way we can do this. The rapid increase in layout savvy that good CSS finally allowed amazed me. The kids learn fast — give them a chance.

    And Adobe? They can learn to compete in this millennium instead of the last, just like everyone else has to. Their terror of some very fine new share- and free-ware fonts by new indie designers is of no consequence to me.

  62. Filipe Abreu says:

    I see the point discussed here as a true typographer AND web designer. I always wanted to use the fonts that I made on my websites, especially my own. I really got frustrated when I realized that wasn’t possible the way I wanted. I either had to use .eof fonts, wich only function on IE or use Flash font embedding, wich isn’t possible (yet) as my site is dynamic and uses PHP.

    But I have learned a lot about this matter so far. I acctually run into a document wich says about a fight between BitStream, a font foundry know by it’s Bitstream Vera type family, and Netscape, the Netscape Communicator (on version 4 at the time) developer (wich “turned” open source wich Mozilla). It seems Netscape had registered the PFR (Portable Font Resource) font file type on dubious ways, as it was developed by Bitstream. At the time Bitstream “opened” it’s code so it could be used for font embedding in Netscape.

    Now this was really font embbeding works! The font really was loaded into the user system, but it could only be used into the especified website environment, and as soon the user leaves the website the font was unloaded. This really worked and as well respected the typographers authorship rights. This is the way .eof (Embedded Open Type) font embedding works in IE and .prf font embedding worked in Netscape Communicator 4 (it’s not supported by prior versions of Netscape neither by any Mozilla/Firefox browsers, due to the Netscape/BitStream disagreements).

    And in this point I support the opinion in the previous post. Some bad intentioned users did seem to find a way of hacking this and acctualy get the font to their use. But still they needed to open the font in a font editing software and redo all the tracking, wich we all know it’s very hard work.

    But for the good sake of the web and as well the updating of the type market (just as the music market updated to the mp3 technologies) the open source community wich is developing the Mozilla Firefox browser is making an agreement wich BitStream to finally work a way to make web font embedding come true. And this is something we all should celebrate, after all the world’s trees are finishing and we’ll have to stop making paper someday. New technologies are at hand (just like Guttenberg did almost four centuries ago) and society is changing too. The ones wh chooses to stay behind it will, well, stay behind.

  63. JNB says:

    Just two cents from a “normal” person:

    The comments from the “designers” around here are really something…

    God forbid someone makes an(other) ugly web page. The world would simply fall apart, wouldn’t it? Thank the Lord for all of you glorious, “Professional Designers”, who can make sure everything stays nice and pretty for us.

    I mean really, who cares if the MySpace crowd gets a hold of stuff that can make there sites even uglier. Means more traffic will leave their site and come to mine.

    As with any profession, the level of competence/experience/skill/taste (and any other words you want to throw in there) varies…and that’s a good thing, IMHO.

    Maybe everyone here has been so tied to their web world they’ve forgotten what the real world is like.

    Ever been to a clothing store? Car Dealership? House Shopping? Furniture Shopping? Seen anything you thought was ugly? Every thought to yourself, “Wow, I’ll bet NO ONE buys that ugly thing,”?

    Well, you may be right, but who are you to tell the guy he can’t make it?

    Take a pill and worry about yourself for a change, everything will be alright. Besides, ugly makes the pretty more beautiful! :)

  64. hrant says:

    Good design isn’t about pretty.

    Please go not care somewhere where
    there are more people who don’t care.


Post a Comment

Comments at Typographica are moderated and copyedited, just like a “Letter to the Editor” in a newspaper. Abusive or off-topic comments will not be published. Compliments are appreciated, but will not be published unless they add to the conversation. Thank you!

Recently Commented

Typographica is a review of typefaces and type books, with occasional commentary on fonts and typographic design. Edited by Stephen Coles and designed by Chris Hamamoto. Founded in 2002 by Joshua Lurie-Terrell. Relaunched in 2009 by Coles and Hamamoto.

Set in Bureau Grot by Font Bureau, Nocturno Display by Nikola Djurek, Fern (unreleased) by David Jonathan Ross, and JAF Bernini Sans by Tim Ahrens.

Brought to you by this month’s nameplate sponsor, FontShop, MyFonts, FontFont, Wordpress, Fused, and the letter B. Read our editorial policy.