We license fonts for print use the way we do because there isn’t a better option. Trying to license fonts on some sort of per-use basis makes business much more complicated, and then means less time drawing letters, which is not something that interests a lot of type designers. But with the web it suddenly becomes much easier to work with a service that charges per-use in some way and cuts users off if they don’t pay. Catching infringers is infinitely easier because someone can just create a spider that looks for your fonts and automatically alerts a lawyer to send out a cease-and-desist-or-pay-me order.

And font designers deserve the money. Users are buying this stuff because it allows them to make more money. Designers don’t usually go out of their way to sell their services at the lowest possible cost just because the clients would like that. Likewise, we should be looking for every angle to make a buck. I’m not in this because I want to be Saint James who helped spread literacy to the downtrodden with cheap type. I do it because I want my passion to pay for vacations in Provence.

My opinion: The winners will be the ones who are willing to put up with a little bit of risk, trust the end users and be flexible with their licensing terms. Not everyone is H+FJ and can demand their own terms without paying a price. Buyers are going to go where the getting is good and easy. (Typekit wins in this area.)

I look forward to hearing what others have to say. Thanks for starting the thread Stuart.

I think there's room for all models.

@diner

Good question. I assume you're type designer. Where do you market your fonts now?

@James Puckett
“And font designers deserve the money. Users are buying this stuff because it allows them to make more money. Designers don’t usually go out of their way to sell their services at the lowest possible cost just because the clients would like that. Likewise, we should be looking for every angle to make a buck. I’m not in this because I want to be Saint James who helped spread literacy to the downtrodden with cheap type. I do it because I want my passion to pay for vacations in Provence.”

Unfettered self-interest with no BS. I'm all for it. This is not to say I agree with you on what steps are truly in your best interest, but let's lay it on the line, by all means.

@aluminum

If I hear the word "model" one more time I'm going to scream. (Not your fault, we all succumb.)
How about you sell me your font under the terms of a license and I pay you for it. Is that enough of a model or do we need something more nuanced than that?

Thanks for chiming in on this discussion Richard, I know you're following the webfonts thing very closely . . .

Yes I'm a type designer selling my works at my own site fontdiner.com and selling collaborative works under the foundries Sideshow, Filmotype and Tart Workshop at fontbros.com, myfonts.com and veer.com

While most discussions have been focused on the technology of webfonts, very few have been focused on the customer . . .

Ultimately a license must be purchased as with desktop use or seat licenses and since there isn't an end user per se to license the works to, the only recipient of a webfont license IMHO could possibly be the owner of the website, even if it's spec'd out by a designer, the person who owns the URL ultimately is the buyer and beneficiary of this type of license. In my mind, given that, the license should be URL specific . . .

Taking that further, getting back to the technology piece, yes, a font could be rented (Typekit) and yes, a font (all webfont assets) could be purchased or given away (Font Squirrel) for a one time fee . . .

I personally agree with James that URL binding is an important piece of this puzzle because we don't want to DRM fonts, but we want to do what we can to assure they're being used in accordance with the license that is being sold for this use . . .

And to Ethan's point, his preference is that the fonts are owned by the designer and they, like all current desktop users should be on the honesty system and can use them on any site under their control . . .

The designer is an agent of the client and even if they license fonts on the clients behalf to do design work, even for web work, ultimately I don't feel you can license a font to a designer who can use the font for other projects and implement the same @ font-face kit for each client's site they work on which is why I think URL binding means the font is purchased PER project and PER client as it's needed . . .

Beyond that, I don't think there is really much more to it . . .

(I'm going to risk making Richard scream. Sorry, Richard. I can't think of a better word for it.)

To me, the "rental" model (e.g., Typekit), is a desirable alternative to the "purchase" model we've had up until now. Whether it's viable remains to be seen, but I'm hopeful.

When I tell non-type people that I make fonts, they always say something like, "So, does that mean you get paid whenever your font is used in an ad or a book or whatever?" When I tell them it doesn't work that way, that I get paid the same royalty whether my font is used in an ad seen by millions of people or a birthday party invitation seen by a dozen, they are surprised. I think a lot of people who "steal" fonts do so because they want to use them for personal projects and don't think it's fair that should they have to pay the same price for fonts as professional users.

The reason the "purchase" model has been used is partly historical, going back to when fonts were only bought by typesetting shops. But it's also because there's been no practical way to measure how much a font is used. Can you imagine keeping track of your font usage and sending reports to foundries and paying licensing fees based on that?

However, when you serve fonts on the web, you can put a meter on usage. This could (and should) lead to fairer pricing for font users. Plus, I think this is a model that is easier for users to understand because it's based on how much they use a font, not simply whether they have it on their computer or not. It puts font licenses on a sliding scale, instead of the current all or nothing situation. Musicians often make more from live performances than from CDs or MP3 sales. The "rental" model would, in effect, be like live performances for fonts.

I don't think the "rental" model will replace the "purchase" model (it doesn't work at all for desktop fonts), but I think that, like markets for other goods and services, there is room for both. Not everyone chooses to buy movies on DVD. Most people are happy to rent. The "rental" model simply hasn't been an option for fonts.

And not to sidetrack this discussion, but I think that this is another part of some other discussions that are going on about rights-managed type sales. I’ve been advocating rights managed type since I realized how hard it can be for design students to afford to buy contemporary type for their projects. In the last six months web font services have eliminated that problem for at one area of design. The same can be said for freelancers, small design firms, and their small business clients who want great type but are better off amortizing type costs across subscription fees. In the next year PLINC will be able to do something similar for display type used on physical mediums.

The benefits of this seem pretty huge to me. Small design businesses are going to have access to oodles of fonts other than what they get with their copy of Type Odyssey. As font designers we get ripped off less and make more money by spreading lots of small transactions across people who would otherwise have just used another classic face on projects with tight budgets. I think that will translate to more cool new ideas because we don’t have to worry so much about people not buying really risky fonts that push the boundaries of form and technology but have to cost an arm and a leg with the existing, mostly royalty-free licenses.

Tracking this thread (and remembering to read it later).

I appreciate the comments and feedback so far but for folks who've been hyper-active on other webfont threads, I'd like to see their thoughts on this one. . . I'm looking your way David Berlow, John Hudson and Si Daniels . . .

@richard:

"How about you sell me your font under the terms of a license and I pay you for it. Is that enough of a model."

Between you and me, sure. But you and I are not the entirety of the market.

@mark simonson
ms>Can you imagine keeping track of your font usage and sending reports to foundries
ms>and paying licensing fees based on that?

Some software works that way. Fees are based on user stats reported by the licensee.

As for the rental "model" [I screamed, the police came, I'm sedated now.]:
ms>Plus, I think this is a model that is easier for users to understand because it’s
ms>based on how much they use a font, not simply whether they have it on their computer
ms>or not.

Things can only have a value if the customer sees the value. Tiny bottles of perfume sell for hundreds of dollars. Someone sees a value in it and are willing to pay.
The question is - does the customer see the font as adding to the bottom line or is it just a nicety and another font - perhaps free - will do just as well. To me, how much I use a font means nothing. But that's me. Plus, extra visitors does not necessarily translate into commercial success. Just more visitors. Figuring out how to make money on the net is hard enough as it is without allowing yet another entity to get at my checkbook. Fonts are not just out there competing with other fonts, they are out there competing with other things that sites must have.
So there's the inevitable comparisons - my Apache web server software is a hell of a lot more complicated than a font, and it's free. Wordpress, free. PHP, free. And I can always stick with system fonts and continue to spruce up my pages with text as images as I've always done. I don't know who would see the value in renting a font on a sliding scale. I just don't know. But it's a big world. I suppose Typekit, if they share the info, will help the type industry find this out.

I'm not even going to try to predict.

@diner
I sent you a message on your contact form.

@James Puckett

What do you mean by rights-managed type sales?

What do you mean by rights-managed type sales?

In the world of stock images there are two licensing models. Royalty-free stock means that once you’ve paid for the stock photos you can use them for just about anything short of pornography, in any quantity, without paying more. Rights-managed images are priced based on nature of use, quantity of reproduction, and how many years you plan to keep reproducing an image. If you just want to use a rights-managed photo on ten coffee cups you pay relatively little. For an annual report with a print run of 25,000 you pay more. And for a book cover with an initial print run of 250,000 that might be reprinted five times you pay through the nose.

To me a rights-managed approach makes much more sense for type designers and users.

Getting back to Mark and James comments, the thought is just because you CAN rent a font in the webosphere doesn't mean you should . . .

I'm trying to reverse the case and say if we as font designers make a majority of our income via seat licenses at $20-$30 a font, wouldn't we make the same or more sales if we sold webfonts for $20-$30 per font per site?

Another thought is that perhaps at checkout of a one time purchase when they enter the domain name, it scrapes an Alexa.com history of the domain and returns a traffic volume statistic that informs the price of the purchase or says we'll be in touch with a price via e-mail . . .

Again, I'm not trying to advocate for fixed price webfont purchase so much as I am to say that it's an option worth exploring and nobody has really considered it when in fact offering webfonts at a fixed price is likely for many foundries to create an new sales channel for older works . . .

…wouldn’t we make the same or more sales if we sold webfonts for $20-$30 per font per site?

That assumes users will be willing to pay roughly the same price that they would for a per-seat license with almost no restrictions on creating static content. I doubt that they will. I think that this is probably because designers, and even moreso their clients, have a really hard time wrapping their head around the difference between a font licensed for static content creating and a font licensed for dynamic content rasterization on multiple computers. They don’t even like buying more than one license for the entire business, even if they have 10,000 employees. So getting them to pay $30 per font for one web site that uses six fonts and will get completely redesigned in two years doesn’t seem likely to me.

To me licensing each font for $5 or $10 a year, per-font, per-site seems like a better option. Or to do what Typotheque does; charging more for no limit on the number of sites/server a font can be used on but requiring clients to own their own licenses and charging for excessive bandwidth consumption.

Think of the way a lot of site hosting is done.

You can install a blog manager on your web server (which you're most likely renting) and deal with all the configuration and other stuff yourself (I do this with my own blog), or you can go with a hosted service like Blogger and let the service worry about that stuff. It's not like one is inherently better than the other, it just depends on what works best for your needs and how much you're willing or able to deal with the technical details.

I think we'll see the same sort of thing with web fonts.

Another thing to think about: Are you prepared to do technical support for users who can't figure out how to get your fonts working on their websites in different browsers?

@diner

You remind me of Nick Nolte's character in the movie Lorenzo's Oil. He keeps saying, "I'm just a simple man asking simple questions." And through asking his series of "simple" questions he saves his son's life by finding an effective treatment for a debilitating and ultimately fatal disease. A disease that, beyond a diagnosis, has the top medical people completely stumped.
It's a good model to follow. (Argh! I said it again!)
I'm actually a little stunned. What you're suggesting is an approach that the retail font folks have treated like the plague.

But why don't you try it and see? If the fonts on diner are all under your control you could do a split run and see what happens. Start a new site that, on the same site, offers a choice between a web friendly presentation and the typical print oriented presentation.

Excuse me for waxing philosophical but...
For awhile, right out of college, I sold a line of big ticket, luxury items. High-pressure sales that demanded you to be a hard closer. During orientation by the regional sales manager he told us something that I never forgot: You can say anything to a prospective customer as long as you say it with a smile. It's true. Whatever happened to asking nicely? And saying "Thanks for your order"?
Are these things considered old fashioned or unnecessary?
Recently I bought a font from a major font design studio and did not even receive a thank you email. Nothing. That's never happened before. Quite the opposite actually.
I bought a font from TypeCulture and not only got a thank you email but actually got a little something in the mail, too. I will very much remember where I was treated as a welcome customer and where I wasn't.
Whenever I get the feeling that a company feels like they're doing me a favor by selling me something, if it's not too late, I walk. And then, as long as I can help it, I won't be back. The reverse is true, too.

If you were to organize your site the right way - font squirrel is certainly a good model but you'll probably want to make some changes to protect your fonts from casual download. And then create a friendly atmosphere and ask nicely and follow up appreciatively. Avoid an attitude of entitlement, apply some commonsense salesmanship, and I think you'll sell plenty.
What have you got to lose?

Regards, rich

Thanks for your thoughts Rich,

I guess I'm waiting for more type designers to chime in as well and state their position on the premise . . .

I'm a little surprised the most eagerly adopted method to retail webfonts is rental only . . . I think it's only a matter of time before web designers and site owners take a pass on that . . . It doesn't seem sustainable to me . . .

Further, I think it's a double-standard to present essentially the same offering that's being sold for a one time fee for a recurrent one when the benefit is no different . . .

BTW, the emperor is naked . . . There I said it . . .

Stuart :D

We don't yet have an interoperable, protected web font format, so I can't blame any foundry for not selling web font licenses outright. There's nothing good to protect their fonts. The best thing font rental has going for it right now is extra protection. Remember, even with mild protection like WOFF, fonts are still sitting on a server for all to see. That's quite different from licensing for print, where fonts stay in a closed environment (sorta maybe).

Hi Stuart!

I am very interested in all models because neither type use, nor type users fall into a single web font box. Unfortunately, if I spend a lot of time on free fonts discussions, I'll be raising my prices. Nevertheless, I think this much is worth every penny.

One primary concern of mine at the beginning of mass market DTP was to ensure that the bottom of the price market was well covered by free and very high quality fonts in the form of OS and application bundles. I think this was accomplished with the help of OS and app developers, and that same effort leaped to web fonts where some of the same bundled defaults have served the web for the last 15 years.

One of my primary concerns now, at 'the beginning' of web fonts, is to establish the position of the bottom of the price market. So, I'd like to introduce you to the EPAR, where the field is being leveled for all font developers by an openly published and developable table that will make it possible for non-commercial font folk to have an effective database of font information without developing the kinds of proprietary databases most commercial font developers have.

I hope the "free font" and "f**k the foundries" activists take this new table very seriously, get involved in the final definition of this table, and then take full advantage of this table by working diligently to turn those heaps of unknown typographic quality into a much more useful heap, perhaps... putting FontSquirrel out of business, but I hope not.

Cheers!

I think it’s only a matter of time before web designers and site owners take a pass on that…

I disagree. It’s not like font designers are the music or movie industries trying to get teenagers to pay for one license to subscribe to some music, another license to get the same music as a ringtone, another for their music games, etc. Fonts are serious tools for serious businesses. The cost of url-bound font license should be paid by the client just as the cost of server hosting, CMS systems, etc. For businesses shifting their ad budgets from print to digital the cost of font rentals is probably going to be irrelevant to executives who are patting themselves on the back for all the paper they aren’t buying anymore. And for a small business it’s probably an easier sell than buying an expensive new type family or two.

Further, I think it’s a double-standard…

So what? Like JFK said, life is not fair. Business is full of double standards. This is nothing compared to some of the corruption and kickbacks that go on in printing and ad buying. As long as there’s no dishonesty involved I don’t think it will be a problem.

"We don’t yet have an interoperable, protected web font format, so I can’t blame any foundry for not selling web font licenses outright."

a) that'll never happen and

b) neither does a lot of media...music, photography, etc...however they've managed to adopt a variety of licensing models as well...including rights managed as James mentiones.

"The best thing font rental has going for it right now is extra protection."

Or, rather, the 'belief' that it's extra protection. ;o)

Alternative viewpoint: What font rental has going for itself right now is ease of use for the customer. They are paying for convenience.

Given all these perspectives, do foundries need a font rental company? I presume if a font designer has a shopping cart set up at their site to retail products, why wouldn't they offer their own rental service?

@ Mark, if it's as easy as Typotheque is making it, would customer service really be an issue if all a customer is getting is two lines of CSS and a step-by-step how to?

@ David, you're likely one of few who has an informed perspective on the whole cycle from the beginning of mass market font offerings and you make some good points . . . I was using Font Squirrel as an example of a one time offering but I agree that we'd be waxing philosophical if we were talking about free webfonts . . . I'm more interested in how this could translate to commercially licensed for pay fonts . . .

Diner: There is a thread I started at Typophile and I would be interested to know what your experience has been so far offering webfonts.

Thanks for inviting me here, I am not spending as much time on Typophile as I wished.

To answer Stuart's question - it is too early to evaluate our webfonts solution, we launched it just a month ago. We've had a very good feedback, but it will be interesting to see how the sales will be on a longer term (and not just the short term affected by the hype). From the webfonts sales that we had, the most interesting thing is that it was people who never bought fonts before. We offer a possibility of webfonts also for a clients who licensed desktop fonts earlier, but that was little used so far.

We've worked on the webfonts solution since March, so had plenty of time to think about the licensing terms. We decided to go with a model, that I as a customer would feel most comfortable - that is as close to possible to one time costs. There is a one time licence fee, but because bandwidth costs money, we had to introduce a monthly limit (500MB per used font, per month, for free). So if you buy a package of 6 fonts, you get 3GB for free, which is a bandwidth of a serious site. Another possibility that we considered, was to give fonts for free, but user paying for bandwidth from the first hit. We decided against it, because it was a lot more unpredictable.

I think there is plenty of room for other alternatives, so I am very curious to see other developments.

Peter
http://www.typotheque.com

Or, rather, the ’belief’ that it’s extra protection. ;o)

Actually, it seems obvious to me that font subscriptions (e.g. Typekit) are better protection than WOFF. Typekit obfuscates font data in a number of ways, and although their techniques are well-understood and reversible (at least in theory), it looks like a greater deterrence than WOFF's wrapper.

During the recent W3C web fonts meeting in Santa Clara, I was interested (but not surprised) to hear of impending WOFF support outside the browser, prompting me to wonder out loud if OS support was inevitable. One response was that Linux system support would happen. If/when OS support emerges, then we are that much closer to WOFF being the equivalent of any other desktop font -- and avoiding that is one of the main reasons for the web font format debate in the first place.

"Given all these perspectives, do foundries need a font rental company?"

Probably not. But it's also a convenience for the foundry. Let someone else deal with hosting fees and server maintenance and the like.

"Typekit obfuscates font data in a number of ways, and although their techniques are well-understood and reversible (at least in theory), it looks like a greater deterrence than WOFF’s wrapper."

It's pretty simple obfuscation and has already been 'cracked' by the Russians. They always figure those things out. ;)

DRM/obfuscation/etc of any sort perpetually fails or becomes obsolete. It's a short term remedy, at best. But if it gets the ball rolling, I suppose that's something...

The format is documented hence no need to 'crack' nor to be Russian.

@ k.l. note the smiley. ;)

> I’m a little surprised the most eagerly adopted method to retail webfonts is rental only

It's still early, Stuart. It just happens that Typekit chose a rental model and they have the most traction right now. There will be other mechanisms and license models. The smart foundries won't settle for just one model. They'll cater to the broad range of customers with different needs.

Mark Simonson said perfectly everything else I was going to say about the topic. The market for fonts is expanding to users who would never have licensed fonts before. Foundries need to be open to new ways of reaching these new customers. It's an exciting time to be in the business.

One more thing concerning the title of this thread. I am surprised no one has mentioned it before: there are very few end users who own typefaces. There are a few major companies such as Audi, BMW, Vodafone, etc, who can pay hundreds of thousands of dollars to 'buy' typefaces, the rest of the world licenses limited rights to use typefaces.

Peter

> Why Rent When You Can Own?

Because it's equitable?

Fonts licensed for desktop use have "the number of computers" as a natural quantum. It's fair, easy, scalable, and has endured for twenty years because it serves everyone equally well. News organizations license fonts for thousands of computers, small businesses license them for a couple, individuals for just one. The market has been very effective in determining prices (despite Ethan's cryptic comment above.)

What hosted solutions offer is this same promise for the web, where "traffic" is the standard unit of measure. Hosting offers a reliable way to meet different levels of use, far more effectively and reliably than any kind of single-price (or tiered-price) solution. Surely we want to make it possible for everyone to use our fonts, not only major corporations but small businesses, non-profits, and individuals as well?

@all

Diner seems to have kicked off some sober analysis and I'm glad to see it.
What you have now are untested hypotheses. Thoughts. (Although Paul's service is more than a thought.)

Heck, anything will work as long as those pesky customers go along with it and are willing to pay. The only way to test a hypothese is to experiment. Try it and see.

For the first time ever, you will be dealing with the general public - everybody and their brother has a web page - not professionals.
Gazillions of them.

Where this will lead you, I can only guess.
I only know what a typical web developer does and needs in order to do their work effectively and efficiently.

Curiously, I've yet to read a business case for using linked fonts over the alternatives. I mean, why do it at all? This is far from obvious. Gotta think about that one.

regards, rich

I'm a little late to the discussion -- it's been a busy couple weeks for us. And I'm not sure there is a whole lot more I could add, especially considering Mark Simonson's excellent note about the rental business. Mark was one of the first designers we worked with at Typekit, and I always appreciate his openness and clarity.

I guess my ultimate reaction is this: It's early. Very early. Browsers supporting font linking have only been in circulation for a matter of months. The most successful brands on the web have been around only a handful of years. Even the most seasoned web designer has a bit more than a decade of professional experience. How will this all come together? How will individual web site owners and giant corporations make use of webfonts? I can't begin to guess.

One of the reasons we started working on the Typekit idea was to try a bunch of stuff and look for answers. Some of our customers like the idea of subscribing to a library of fonts and trying lots of alternatives in their designs. Other customers have commissioned fonts and are eager to finally use them online. And there's a whole group of our customers aren't even interested in choosing fonts -- they just want cooler looking blog templates.

Which one of these methods will win? Probably no single one, or perhaps something we haven't even thought of yet. But it seems like now is the time to experiment and iterate, so that's what we're trying to do.

I'm pleased the scope of the discussion continues to open up . . .

@ Jonathan, you make a logical case for finding currency in the 'value' of the the font given its 'home' . . .

@ Jeff, can you give us a sense of what's working so far? How do people feel about renting? Has anybody asked for a one time payment to use a particular font in perpetuity?

I continue to be intrigued how the consensus of the participants of this thread mostly fall in the camp of rental but agree the argument has merit . . . On some level I've always imagined designers working on a project, deciding they need to enhance their work with a font, shop for one, buy it, apply it . . .

That said, we're currently selling fonts to designers . . . However with a rental model, we're now selling fonts to site owners VIA designers so it creates a confusing licensing issue which may create confusion about ownership . . .

It's already difficult to have the discussion with a freelancer who's working for a client about who owns and should buy the font let alone renting one to them . . .

Stuart :D

Why rent when you can buy? it's not a question for users, but for foundries.

In the short term, because foundries don't find enough inherent protection to make them happy in the existing formats/setups for people hosting fonts themselves. Certainly not for raw desktop fonts....

But financially, if it's a rental, it's easier to make the price scale with the volume of the site. Foundries can potentially price to be attractive to the very low end and still charge enough at the high end.

At least, that's what I like about rental from the foundry side.

From the user side, it can be made affordable regardless of who you are.

Cheers,

T

> the consensus of the participants of this thread mostly fall in the camp of rental

But you don't have to settle for just one camp. I think those defending rental are simply saying that a subscription model is one viable option that works for some users, particularly new users on the web.

(I'm tracking this thread)

@Thomas Phinney
TP>it’s not a question for users, but for foundries.

I'm sorry, Thomas, but this is upside-down thinking.
Starting with the premise of - what are we willing to do? what's best for us? what makes us "comfortable"? - and then hoping customers will go along doesn't usually have a happy ending.

Do I have to go digging through back-issues of the Harvard Business Review and the Wall Street Journal to prove this case? I don't have to. All I have to do is wait until the results are in. (They may already be in, actually.)

Try starting with - "What do web users seem to want? What's their reality? How do they see things? What's their normal expectations? What fits into their normal workflow?"
And then maybe you'll get somewhere.

BTW - there is no rental model. Typekit, Typotheque, and Kernest don't rent in the common sense of the word. They are font-hosting and font-obfuscation services. Calling it a rental clouds the specifics. The "service" part of it and the "obfuscation" part of it might, in the long term, prove more significant and useful than the fonts themselves.

Also, what constitutes the "volume" of the site?
And why should site owners see a connection between what it costs them to use a font and the "volume" of the site?
I certainly don't. I see a connection between the bandwidth expended by the services company on my behalf but that's about all.
Are you aware of any potential customers who see it differently?

@all
To anyone's knowledge, has any formal market research been done about any of this?

Respectfully, Rich

And why should site owners see a connection between what it costs them to use a font and the “volume” of the site?

Because in a well-run business there should be a correlation between site volume and income. If a business is running a web site to make more money, using my fonts, why shouldn’t I get more money for those fonts? After all, if using the fonts of one’s choice is not relevant to running a successful online business, why can’t businesses just stick with the MS Core fonts?

You don’t see a connection between business and font costs because you’re a web development pundit. You need only worry about what is ideal for web developers. You don’t have to find a way to monetize a type family that represents months or even years of work but might only be purchased a few dozen times depending on what way the winds of fashion blow.

Are you aware of any potential customers who see it differently?

How about the tens of thousands of businesses that license photography and illustrations this way and have been doing so for several decades?

Looking specifically at site traffic (aka volume), what are logical tiers? I'm most familiar with Alexa's ranking but I'm not sure how it's calculated . . . I'm very curious what is considered 'typical' volume?

For those in the know, what is the low, median, and high end of traffic volumes? Also, how would pricing slide accordingly? Compare that to the cost of 'cloud' hosting for the same bandwidth . . .

Even if we are renting bandwidth to users for the webfonts with optimized compression, at what point is volume actually calculable? Last I looked, hosting bandwidth fees were dropping as fast as RAM prices . . . Are we upselling bandwidth then?

Stuart

"BTW - there is no rental model. Typekit, Typotheque, and Kernest don’t rent in the common sense of the word. They are font-hosting and font-obfuscation services."

Yes, it's a service. What they're renting is access, not the font files themselves. But the key point is that it's an ongoing payment instead of a one-shot up front. I agree that the obfuscation etcetera is a big part of *why* a hosted service makes sense... from the foundries' perspective. The customer couldn't care less.

TP>it’s not a question for users, but for foundries.

"I’m sorry, Thomas, but this is upside-down thinking."

Well, no. The reason it's a question for foundries is that today, users do not in fact have the option of buying for web font usage. They may very well want to, but foundries have generally been unwilling to go there (with the notable exception of Ascender).

Also, what constitutes the “volume” of the site?

There are many different ways one could measure it, and the folks renting or selling the font licenses can pick whichever suits them. TypeKit uses the bandwidth of font downloads.

"And why should site owners see a connection between what it costs them to use a font and the “volume” of the site?"

Why shouldn't they? If they're large they are paying for their bandwidth usage in other ways. Cloud computing services charge based on volume in various ways. And, what James said. :)

Of course the question of what will happen ultimately depends on *both* the buyer and the seller. I expect that we'll find that some users are willing to "rent" and that some are not. I expect that we will see both service and sales models that are successful. It's a big world out there, and I believe there's room for several vendors to make a go of it.

Cheers,

T

I think a big issue web designers have is: "How do I sell my client on this extra font cost? I have no idea. Rather than have this conversation, I could use a free font or consult Aluminum's russian buddies."

How to remove this fear:
Web designers know how to sell hosting to their clients. Make fonts part of the hosting bill and the fear is gone. Seems like it would make sense to offer embedded fonts right from the control panel, and bundle the billing in the hosting package. How big would the audience be if a typekit/typotheque style solution was offered Fantastico style? As a web designer I'm used to a fee-for-traffic solution. I expect it in that context. My clients all expect to pay $xxx monthly for web hosting, and if we go over our bandwidth, we pay. Just like a cell phone. Just like Peter's font hosting model. A great partner in this endeavor would be the web hosts themselves.

"How about the tens of thousands of businesses that license photography and illustrations this way and have been doing so for several decades?"

I don't license photos on per-views of my web site. Granted, a lot of us used to pay for photography and illustration on a per-printed-piece model in the past. Some of us still do, though that pricing model seems to be taking up a smaller and smaller part of the market.

So comparing fonts to commercial art works in that sense.

We could also compare fonts to software. Same types of analogies could be made. SOME software is licenses on a per-user basis. Some software is licensed outright regardless of the size of operation or web site you might be running on it.

Again, there's rooms for all models.

"For those in the know, what is the low, median, and high end of traffic volumes? Also, how would pricing slide accordingly? Compare that to the cost of ’cloud’ hosting for the same bandwidth . . ."

This came up in a presentation of Kernest this past weekend.

There are valid concerns with the outsourcing of hosting of files if you are a major bandwidth user. If your site is pulling enough traffic that it could bring down one of these services, that's not going to make you, nor the service very happy.

However, a good counterpoint was brought up in that if you are running a site of that size, you already are likely outsourcing a lot of your file hosting...your jQuery is on Google, your images on are on CDN, your DBs are on a farm. Etc. So that perhaps isn't as critical of an issue as one may thing at first. Plus, if you're that big of a site, you would likely want to work directly with the foundry to license your own @font-face agreement to leverage the CDN you are already using.

@Randy

I don't we web developers/designers have any fear of this. We've been doing just fine without @font-face up to this point. We'd like it. We're going to use it when it makes sense (technically and legally), but we're not losing sleep over it either.

I agree with the hosting part, though I think it's important to separate web designers/developers from web hosting options. Sometimes it's the same, very often, it is not.

But the point about hosting being a service makes a lot of sense for both the consumer (web designer) and provider (font foundry) in a lot of situations. That's pretty much what Typekit and Kernest and the like are. They are font hosting services.