This title got pruned back too far; put the word "Free" back into it.
HN pro typosetters: some expert critiques of these faces --- particularly the ones that aren't just offering a teaser weight, and aren't display faces more appropriate to deviantART than a magazine --- would be awesome. Are free typefaces improving, starting to break out of the foundry system?
Of the fonts listed, the ones most interesting to me are:
- Lato: Complete font families are hard to come by, especially ones under an open licence, with reasonably complete character sets and the necessary kerning adjustments. This one looks pretty great.
- Crimson Text: Looks pretty promising, and has a lot of very useful features on the roadmap, but as the page says, "Crimson Text is under heavy development, and the spacing/kerning is abysmal. If you use it in public these days, you'll likely embarass yourself." I'll be keeping an eye on it, but sticking with Linux Libertine or Heuristica for the moment.
- Neuton Regular: Friendly and with a bit of personality, it's a shame there's only one weight available, but it looks pretty solid for body text.
I’d advise asking at Typophile if you want good expert critiques. I don’t know that there’s a big enough community of type experts at HN to start a useful conversation.
I’m not an expert by any stretch of the imagination, but I liked Espinosa the best of the featured faces. From a Google search, here’s the Typophile discussion of that one: http://www.typophile.com/node/74292 ... note that apparently only the regular-weight roman is available for free, though it does include small caps, ligatures, old-style numbers, &c.; the other 7 fonts in the family are $50 each, or all 8 + 4 fonts of special drop caps for $400.
I thought Phoenica looked familiar - turns out it's pretty similar to the font Canonical commissioned for Ubuntu. I think I like Ubuntu's better, actually. And of course it's free :-) http://font.ubuntu.com/
I am a developer, but I have been trying to push my boundaries. I am using a lot the @font-face css rule, but I noticed that most of the websites keep using image replacement for text. Why is that? Is bad to use the @font-face rule?
Rendering differences is the reasons. To get the site to look consistent across all major browsers and platforms Cufon is the only practical option at the moment.
The licensing issues is a problem of its own even with image replacement... even if it's a server-side image replacement like Flir. Many foundries are still scared sh#tless of the Internet and are overly protective of their work. Some are not, for example - surprise - Adobe and Dalton Maag. For other fondries however it's a business decision. For example, one cannot use any of HF&J fonts such as Gotham and Archer online in any form or fashion. They are building their own font service. Anytime you see a site sporting Gotham under Cufon, it's an explicit license violation.
So in short - if you are considering using a font with an image replacement, check the license. If it is not explicitly permitted, then in all likelihood it is disallowed. If still set on using this font, email foundry and ask.
You mean I can't automatically embed HFJ faces, right? If I painstakingly Photoshop my hedlines, I can use Archer for that, right?
Also, when implying that font-face has dealbreaking rendering quality issues, are you referring to their performance for setting body copy, or do you think it flat out doesn't work for display type either?
It's not so much that rendering a whole bunch of text takes a while (though it might, but it probably has more to do with the time it takes to download the font), but more that the letters look different depending on who looks at it.
After that, it's up to the designer to decide whether or not the difference is acceptable. I'd say it works out fine in pretty much all cases. Your site doesn't have to look the same in all browsers, people.
> Also, when implying that font-face has dealbreaking rendering quality issues, are you referring to their performance for setting body copy, or do you think it flat out doesn't work for display type either?
As of today, there seem to be very few fonts, even from the specialist services like Typekit, that render well across all browsers via @font-face. In my experience, this does apply to both body text and headline sizes, but what you might call the risk/reward ratio is different.
For typical body fonts, there is never going to be a huge difference in appearance between text set using @font-face and text set in well-established screen-friendly fonts like Verdana, Georgia and the Microsoft C-fonts, at least not until we get the kinds of resolutions now used on smartphones routinely available on full-size screens as well. IMHO, using @font-face for body text seems like asking for trouble today: poor hinting or antialiasing can literally render your text illegible, and there is hardly any potential upside anyway.
On the other hand, for headlines, pull quotes, and other text typically set a bit larger than body copy, you can get some genuine character in the font on a typical display today. There certainly are still problems -- many trendy graphic design blogs currently look terrible on my screen because they are trying to be too clever with this technology -- but the flaws are less significant relative to the overall effect, because things like antialiasing and hinting don't matter so much at larger sizes, and it's very unlikely that your text will actually be unreadable as a result. That means there is more incentive to use different fonts via whatever mechanism when you're setting larger text.
> You mean I can't automatically embed HFJ faces, right? If I painstakingly Photoshop my hedlines, I can use Archer for that, right?
Yes on both counts.
> Also, when implying that font-face has dealbreaking rendering quality issues, are you referring to their performance for setting body copy, or do you think it flat out doesn't work for display type either?
Depends on the typeface, or more specifically on whether it was explicitly tested and "optimized" for @font-face use, which typically refers to generating hinting info needed to make the font render decently on non-Apple platforms. For example, Museo will render great in all browsers, while something like Avenir will look bloody awful in smaller sizes in FF (not that it is available for online use to begin with). This narrows down the list of usable @font-face typefaces quite a bit, it also complicates testing and what not.
Two reasons.
1. IE support whilst possible doesn't give great quality results, especially at smaller point sizes.
2. Font licensing hassle. None of the foundries seem to want to license fonts and getting an online license is often too much hassle. Image replacement is ok, since you're not making the font file and metrics available to download.
Actually, point #2 is wrong. Most of the major foundries have their own font licensing system these days (or it's a work-in-progress). It's not the greatest solution since you can't have all your licensed fonts in, say, TypeKit, but it's not too bad.
Yes, that may be true but that's some foundries and some fonts. How these things normally work is that the branding and typography for a brand is done by a separate agency who specify the typefaces, then a separate digital agency have to implement it. If the chosen typeface isn't available with an online license then you have to fall back to replacement.
Maybe in another year or so things will have improved to the state where more typefaces are available for online use than not.
> Most of the major foundries have their own font licensing system these days (or it's a work-in-progress).
Perhaps, but are any of them commercially reasonable yet? Pretty much all of the ones I've seen require either ongoing payments and/or third party hosting, or an up-front fee that scales with the (anticipated?) number of page views. As far as I can see, no-one else providing a related product/service gets these kinds of terms: would you pay a monthly fee for a piece of clip art, or a higher licensing fee based on how many times you wanted to print your annual report on paper?
Wrong is too harsh. Those are all recent developments, sometimes work in progress. Licensing wasn’t easy only a few months ago and it is only now getting gradually easier. It’s – like the grandparent said – a hassle.
HN pro typosetters: some expert critiques of these faces --- particularly the ones that aren't just offering a teaser weight, and aren't display faces more appropriate to deviantART than a magazine --- would be awesome. Are free typefaces improving, starting to break out of the foundry system?