An alternative suggestion for web fonts licensing

Warning This article was written over six months ago, and may contain outdated information.

Following the recent push by Microsoft and Ascender to revive the EOT format for web fonts, debate has raged over the pros and cons of the two main alternatives: embedding and linking. Richard Rutter came up with the idea to license fonts on a monthly payment basis, with the font being served from the supplier’s server (or a trusted alternative).

I think there are a number of potential problems of practicality with that approach, many of which have been raised in the comments. As a contribution to the debate, I would like to offer the following suggestion:

What if the font vendor were to supply an authorisation code when selling a font licensed for online use, which the user were then obliged to use inside the @font-face rule in any CSS file which called the font in question; perhaps like so:

@font-face {
font-family: MyFont;
src: url('myfont.otf');
/* License: XXX-123 */

This would make it easier for font vendors to crawl websites looking for instances of their fonts, and checking to see if there were a valid license issued for its use on that particular site. This could involve website makers hosting a file somewhere with details of all the sites they manage, to which the vendor could cross-reference.

Of course this wouldn’t stop font piracy, and it wouldn’t be effective against users stealing fonts for use in offline media. But it would allow vendors an easy method of checking that fonts used online were used legally, at least removing that burden from them.

Worth discussing?

10 comments on
“An alternative suggestion for web fonts licensing”

  1. I’m not up to speed on this discussion, so apologies if I’m being slow. Are you suggesting that the file is served from the vendor and the CSS comment must be present in order for the font to work?

    I for one have no problem with paying a one-off licence fee for the font, so your suggestion (if I’ve understood it correctly) seems a reasonable one.

  2. That wasn’t my intention; what I meant was that the file is hosted on your own site and the license included in your stylesheet. That wouldn’t, of course, stop anyone from downloading the font file; what it would do is protect the correctly licensed users from the threat of prosecution, and make it easier for font vendors to find unlicensed copies being used on the web.

  3. Ah, sorry. Is there any mileage in what I thought you were saying?

  4. That’s basically the idea that Richard Rutter proposed; for the fonts to be hosted and served by a protected, licensed, third-party.

    My issues with that are a) the pricing model (see this comment) and b) relying on a third-party server. The usual issue of DRM raises its head: if the font vendor closes down, we lose the font we’ve paid for.

  5. I agree that pricing model wouldn’t work, but (presuming that people are happy to pay, which is the other issue) couldn’t we just pay once, as we do for offline fonts, and store one copy locally for backup?

    My concern about the CSS comment as you propose it is that I suspect only legitimate users would bother to use it (people who omit the comment would be hard to track, as they could easily hide their tracks eg by renaming the font file). Therefore using this method to keep track of illegitimate use would be pretty unreliable.

    And even legitimate users would have to remember to include the comment each time they used the font (something I would struggle with myself!), which would further reduce its reliability.

    And if the vendor sees no benefit in using it they’re unlikely to adopt it.

    I don’t mean to sound negative, I definitely think there’s some milege with your idea.

  6. Don’t worry; I’m always up for a bit of constructive criticism.

    Regarding renaming the font file, I did think of that; but I presume the vendors have some kind of embedded signature in the files, or a checksum, something like that. Maybe not, in which case the whole idea is a no-go.

    It could be that this whole idea is a non-starter; I just think we should explore every alternative before settling on DRM.

  7. Could the licence code actually determine whether the font works, regardless of where it’s stored? ie if the CSS comment isn’t present it won’t display? Or maybe that’s what you initially meant by ‘obliged’? Anyway I’ll shut up now! :)

  8. What do you think about the “MAME” proposal at :)

  9. Hi Dave,

    I haven’t gone through it in detail, but at a first glance it seems quite similar to what I proposed.

  10. Okay cool — I hope to see this promoted if the W3C’s new Fonts Working Group is formed, so I’m happy to see that others have been thinking along similar lines :-)