Spring 2010

Monitor: End of default

The technology that allows us to control type on the Web is available. Now designers and foundries have to decide how to use it best.

The short history of the World Wide Web has seen many rapid developments, in interactivity, speed, motion, colour and image-handling, but typography has customarily been seen as a problem, at least by designers who care about the way text looks onscreen.

Web protocols mean that you can only specify fonts from a very small range of default typefaces that are universally available – installed on every users’ machine in order for them to render correctly.

Within these restraints, any designer or art director who wished to specify type with the precision of a job for print, TV or film would have to create the text as an image or build the site in Flash, but such routes raised many problematic issues of speed and accessibility.

Clients such as corporations and publishers who were accustomed to branding every visible square inch with their custom fonts had to accept the default nature of the Web, and many designers have long resigned themselves to living through the typographic equivalent of the dark ages, relieved partially by the advent of Cascading style sheets (CSS) which allow much greater control over the styling elements of a website, including the size, weight and style of the (still limited range of) fonts.

For a significant minority of designers, the limited type palette became a signifier of authenticity, a cool hair shirt they could wear with pride. The ‘default look’ of non-Flash websites has spilled into books, magazines and music design as a conscious style choice rather than necessity.

For many years, the idea of being able to specify any typeface for a fully interactive, W3C-compliant website remained a far-off goal. One of the first signs that a radical change might be on the horizon was the launch of the @font-face tag which gives style sheets the ability to link to font files stored online and make them available to users’ machines.

While this was great news for designers, who can at last wrest back control of the typefaces that sites are displayed in, it seems, on the surface, to be bad news for type designers and digital foundries, who are concerned about the security of their intellectual property in an online environment in which everything is seen to be free. The problem with @font-face is that it works with OpenType (OTF) and TrueType (TTF) files, ‘real’ fonts in other words, and that anyone visiting a site employing @font-face is sent those real font files in some form.

The catalyst in this rapidly changing area was been the emergence of Typekit, the self-described ‘easiest way to use real fonts on the Web’, a third party, Web-only, ‘fonts-as-a-service’ provider, produced by San Francisco-based company Small Batch.

With Typekit, you don’t buy fonts, but create an account with various subscription options based on price and functionality. You then create a ‘kit’ of typefaces for your website from Typekit’s font library, which (at the time of writing) includes part of the FontFont library, Type Together, Porchez Typofonderie and Mark Simonson. Notably absent are major type designers such as Hoefler & Frere-Jones, Font Bureau and all but one member of Village.

Typekit works by storing the fonts on a series of global servers and delivering them to users as a website demands. The important aspect of this is that the source of the fonts is ‘obfuscated’ in order to deter users from accessing and potentially copying the fonts for their own use.

Many fear that Typekit’s obfuscation techniques are not strong enough, and that the fonts can be too easily pirated. The Typekit blog says: ‘The fact is, for something to appear in a browser, it has to be on the Web. If it’s on the Web, it can’t be completely protected.’ In some cases, the browser has to download the font to your hard drive before it can display them, bringing it, in essence, to your feet. Even Typekit co-founder Jeffery Veen admits that ‘someone with enough time and knowledge can recreate a font from our service.’

Changing by the minute

But Typekit is not the only solution: Typotheque is rolling out its own system (see interview with Peter Bil’ak pp.38) which allows owners of Typotheque font licenses to create embeddable fonts via their accounts. It also promises special online versions of fonts for optimal screen rendering. Other players include fontdeck.com and Kernest.com.

Now new Web Open Font Format (WOFF) typefaces promise another more secure way of delivering the fonts, and you don’t need to use a third party’s technology. WOFF fonts can be ‘self-hosted’, although many user’s browsers will need to catch up with the new type embedding technology.

Many foundries and type designers are still considering their position, but it seems clear that as the market for Web fonts grows, the business model is changing. We asked three leaders in the type industry for their thoughts about online type.

Chester Jenkins

Village

With our background in graphic design – I worked as a graphic designer for a decade before transitioning to type seven years ago and my wife and partner Tracy is a graphic designer, too – I believe that we are able to consider our clients’ needs alongside our Village members’ needs. Too often type people forget who they are making their type for; there is a way to balance what our clients think they want and what we think they need. But I certainly don’t speak for giant corporations like Adobe and Linotype, and I don’t speak for the hundreds of fledgling type designers who distribute through MyFonts and are now on Typekit.

I’m not imagining a shift away from ‘traditional’ OpenType fonts towards Web fonts; I see the latter as a complement to the former. I imagine that most designers wanting to use a typeface dynamically online will also want to use that typeface in static designs for print and other media. Because a Web font is a very specific tool for a very specific task, most people will still need the rest of the tools in the kit. I believe that the third-party / subscription model for Web font licensing will have a hard time surviving when most foundries eventually allow licensees to host Web fonts – in the WOFF format, most likely – on their own servers. (And provided that the end-user license isn’t too arduous.)

WOFF is a format which was co-developed by one of our members, Tal Leming, along with Erik van Blokland and Mozilla’s Jonathan Kew. It is the least imperfect of the formats being developed for controlling type on the Web.

We are 100 per cent against @font-face, because it makes specified OpenType fonts freely accessible by, and to all. It is one thing to unknowlingly ‘share’ a Web-only font file, but quite another to give away an OpenType font file which can be used for anything.

I am sure that there will be rampant piracy of Web fonts, but no more so than the piracy of OTFs, and no more damaging to foundries. (Most pirates would not license their purloined fonts under any circumstance.) Clients presently licensing type legally will be the ones licensing Web fonts for their projects. I think that there is a way to provide both ‘traditional’ OTF fonts and Web fonts in ways that benefit both the foundry and the client.

Jonathan Hoefler

Hoefler & Frere-Jones, Inc.

We’re not quite ready to announce anything, but it probably goes without saying that this has been a subject of great interest for some time now.

What I can say is this: we view Web fonts the same way we view other fonts: as an irresistible challenge, and a genuine opportunity to make typography better. H&FJ’s plans are what they’ve always been: we’re working to make the best fonts we can, and to provide them to both designers and readers in the most practical way.

Stephen Coles

Font Shop International

FontShop is taking some decisive steps. We can’t say we know exactly what’s going to happen with the shift from print media to the screen, but we do feel confident that offering a choice is the best way to get a read on the future. We offer fonts via the Typekit Library (a subscription / service model) and as standalone fonts in the WOFF / EOT format (a per-font licensing model). The stand-alone model is very comparable to the way desktop fonts are licensed now. But instead of per-user, it’s per-page view. There are three broad levels intended to assign an appropriate fee to customers as varied as a personal blogger and a multinational corporation.

For most designers one method works much better than the other and we’re already seeing that in our sales. In the end — once WebKit (Safari) and Chrome catch up — I think the WOFF solution will be the preferred option by a majority of our customers, especially corporate clients.

The main benefit of WOFF from a type maker’s standpoint is that the fonts can’t be used in desktop applications. Because these fonts are downloaded to every computer that views a Web page it was critical to type suppliers that they not be reused as standard ‘print’ fonts. It’s an extra layer of protection from unintentional unlicensed use. Just like desktop fonts can be used illegally beyond their license, so too can WOFF fonts. Fortunately, Web font use is much easier for us to track than print font use.