Typekit is a funny concept to me. While I can respect and admire their ability to build a successful company, I think the blog post shows just how rose-tinted are the spectacles worn by these font services:
Second, we could innovate on the business side as well.
We could sell fonts as a service, and use a subscription
model to eliminate Byzantine licensing and usage issues.
I wonder how much money they would have made if they'd sold fonts with a simple, one-term commercial licence fee, just like stock photography, music, icons, etc.
I know there's no way I could ever use their services with any of the clients I've had, because you simply can't factor in an ongoing cost that is unrestricted and controlled by a third party when giving a fixed price quote for most clients. However, I've spent significant amounts of my own and various companies' money to buy good quality fonts for other uses, and would surely have done so for web fonts as well if anyone was willing to take my money on that basis.
Few sites used web fonts when we got started;
today, new sites seldom launch without them.
Typekit now serves nearly three billion fonts
per month on over one million different sites,
including some of the most recognized brands
on the web.
Well, good for them, but since Netcraft reckon there are currently over 400 million Web sites and there are only a handful of font services, that suggests to me that perhaps 1% of the Web is actually using these services.
From the start, our vision has been to make
the web more beautiful, readable, and fast.
Unfortunately, what they've actually done is cause millions of pages to look terrible, indeed sometimes outright illegible, because most of their screen fonts simply aren't as good at body text sizes as the tried and tested Georgia, Verdana, etc. And there is no way that downloading a font from a third party service, even one with a great CDN, is faster than using a native one that's already installed locally.
Herein lies the fundamental problem with the whole web-font-as-service concept: at body sizes, there is rarely enough difference at typical screen resolutions to justify a change from the old favourites (and such changes are usually ill-advised anyway), while for one-off uses like headings and pull quotes where distinctive fonts can make a worthwhile impact, the web design community was managing just fine already.
As higher resolution screens become the norm, perhaps this will change, just as tiny pixel-drawn icons are giving way to scalable vector-based artwork (but it really shows on smaller or lower-resolution screens where you do still want a 16x16 or 32x32 icon and the vectors haven't been carefully crafted to fit pixel-perfect at that kind of scale). Even then, it's hard to see how you can justify paying a substantial amount of cash every month to use fonts on web sites, when no other on-line stock resources work that way and fonts for other uses don't work that way either.
Still, I wish them well, if only for the benefit of those users who can fit in with their business model and do find it worthwhile. I don't share the pessimism of some here about the Adobe takeover, because one of the few things Adobe has pretty much always done well both technically and in terms of management/legal stuff is fonts.
[Disclaimer: My business licenses fonts via Typekit]
Blaming Typekit for bad design is an interesting choice. Might as well blame browsers for giving designers access to millions of terrible color choices via CSS while you're at it. Blame bad designers for not knowing how to use fonts.
The community was not doing a good job with custom fonts for headings, pull quotes, etc. @font-face works pretty well, but foundries clearly weren't willing to sell licenses for it the same way they do for print/commercial use. It's not a difficult thing to explain to clients - it's a recurring fee just like hosting, SSL certificates, and domain registration. It handles all cross-browser issues, saves bandwidth, and gives you, the designer, easy access to a rather large library of fonts at no cost to you. Clients that can't afford the $2-8/month probably can't afford good design to begin with.
If web fonts are too slow for you in all circumstances, how is Typekit any worse than a license from a foundry? Native fonts are great, and should be used extensively, but they really limit your options.
I'm not blaming Typekit for designers who don't know how to use fonts. I'm blaming Typekit for serving fonts that look terrible on many systems that aren't running the latest and greatest browsers and operating systems (i.e., most of the machines used to browse the web). Many of them don't look that great even on those newer systems either, certainly not rendering with the quality that you would expect from a pro font displayed in a native application running locally on the machine.
For a while shortly after these font-hosting services started going live, every trendy design blog out there started to look worse than it did before, and I couldn't figure out why. For a short time, I actually assumed I'd inadvertently messed up my ClearType settings, or downloaded a dubious update from Microsoft, or found another rendering bug in Firefox, or something along those lines. Then I started to notice these little "T" badges in the corner of the pages that didn't render properly, and discovered the hard way that hardly any web fonts were properly hinted for on-screen use, which is why they all looked horrible compared to things like Verdana and Arial/Helvetica.
To be fair, a few of them are better now. Still, Typekit's own site looks pretty bad on this computer (I've just checked) and the irony is that it's their custom fonts that render like something from the Windows 95 days, while the body text in Georgia looks as clean as usual. It's not exactly a great sales pitch when the part of your site using your technology looks obviously visibly worse than the part of your site using the tried and tested workhorse you're asking us to give up.
Notice that even at display size, the rendering of the heading text in Proxima Nova looks awful. Meanwhile, trusty Georgia even renders with decent italics at body text size.
And it looks that way on several systems here, from Firefox on Windows XP to Chrome on Windows 7, so it's nothing to do with ClearType rendering differences, the Firefox rendering problems with HarfBuzz, or any other such excuse.]
As I've said many times, the reality is that at today's typical screen resolutions, at body text sizes, there are only so many ways you can turn on the pixels that are going to work. We have had a few brilliantly designed fonts created specifically for that usage available to us for a long time. Anything that is supposed to replace them needs to look different enough to be worth the hassle and render at least as clearly, and most/all of the hosted web fonts currently fail to meet one or both of those criteria. I'm sorry if this is some sort of downvote-worthy regressive heresy, but I go by results and the screenshot does not lie.
I know there's no way I could ever use their services with any of the clients I've had, because you simply can't factor in an ongoing cost that is unrestricted and controlled by a third party when giving a fixed price quote for most clients. However, I've spent significant amounts of my own and various companies' money to buy good quality fonts for other uses, and would surely have done so for web fonts as well if anyone was willing to take my money on that basis.
Well, good for them, but since Netcraft reckon there are currently over 400 million Web sites and there are only a handful of font services, that suggests to me that perhaps 1% of the Web is actually using these services. Unfortunately, what they've actually done is cause millions of pages to look terrible, indeed sometimes outright illegible, because most of their screen fonts simply aren't as good at body text sizes as the tried and tested Georgia, Verdana, etc. And there is no way that downloading a font from a third party service, even one with a great CDN, is faster than using a native one that's already installed locally.Herein lies the fundamental problem with the whole web-font-as-service concept: at body sizes, there is rarely enough difference at typical screen resolutions to justify a change from the old favourites (and such changes are usually ill-advised anyway), while for one-off uses like headings and pull quotes where distinctive fonts can make a worthwhile impact, the web design community was managing just fine already.
As higher resolution screens become the norm, perhaps this will change, just as tiny pixel-drawn icons are giving way to scalable vector-based artwork (but it really shows on smaller or lower-resolution screens where you do still want a 16x16 or 32x32 icon and the vectors haven't been carefully crafted to fit pixel-perfect at that kind of scale). Even then, it's hard to see how you can justify paying a substantial amount of cash every month to use fonts on web sites, when no other on-line stock resources work that way and fonts for other uses don't work that way either.
Still, I wish them well, if only for the benefit of those users who can fit in with their business model and do find it worthwhile. I don't share the pessimism of some here about the Adobe takeover, because one of the few things Adobe has pretty much always done well both technically and in terms of management/legal stuff is fonts.