Discuss: Real Web Type in Real Web Context
by Tim Brown
- Editorial Comments
2 So glad...
That we aren’t limited to boring old Sans Serif or Arial…
Long live font choices!
posted at 11:14 am on November 17, 2009 by Orlando Website Design
3 Common Sense, What A Concept!
@tim
Thanks for this. Nice, nice work. Now I’m doubly glad I added Nice Web Type to the Linkworthy blogroll at Readable Web (where the current post is, apropos, a heartfelt review/recommendation of DWWS3 )
So far, retail font vendors have shown nothing but reluctance to solicit web designers as customers. This is unlikely to change. They are not eager to move from marketing to print professionals willing to pay premium prices to a more volume based model with reduced prices likely to lead – as they see it – to a reduction in the price they can demand of print professionals, as well.
In other words, don’t hold your breath waiting for the retail font industry to get involved and start solving web-related problems. There is little motivation, and time, to do so. All efforts so far, vis a vis web fonts, have been directed not towards opening up a market for web fonts but in mitigating the possible damage to the prices that can be demanded from their traditional base of print-oriented customers.
It’s called self-interest. And it’s just the way it is.
See: An Open Letter To Retail Font Vendors and comments for a take on how web designers are currently not being served in a way that is useful to them.
For a take on how the web design community is viewed in comparison to those in the print community by one of the type community’s leading voices – and someone who has been given a generous forum here on ALA – check out this video interview:
Web Fonts Week: An Interview With David Berlow Of The Font Bureau
We face an unbridgeable divide, I fear. But I’d like to be proved wrong.
cheers, rich
posted at 12:49 pm on November 17, 2009 by Richard Fink
4 Great stuff but how apt
this is going to be a fantastic revolution! And a great tool in helping with it but already the site highlights that you should be wary of cross browser issues; as the main body text looks awful in firefox but renders well in ie.
Good luck with the project though…love your work
posted at 01:02 pm on November 17, 2009 by bangers
5 The retail font industry
Thanks, all!
Rich, I saw your Open Letter, and although I share many of your sentiments I don’t think type sellers are as unwilling as it may seem. I’ll be interested to know what it would take for a foundry to implement something like Web Font Specimen. My guess is that the issues are exactly the same as those of licensing fonts for use with @font-face: there is yet no way to protect font files from being stolen outright. If a foundry were to participate in a type delivery service or brew their own a la Typotheque, that’d be a different story. Otherwise, I would not be surprised to hear that many type sellers are already preparing ways to show their wares in web context … as soon as there is 1) greater browser support for WOFF/CWT, and 2) a way to scan for illegal use.
posted at 03:11 pm on November 17, 2009 by Tim Brown
6 What do you mean by 'OTF'?
The glyph is exported as part of a font, in either OTF or TTF format (in this context, I am ignoring all other formats—let me know if I shouldn’t). TrueType hints are ditched if the font is OTF.
Given the context, I think you’re confusing ‘OTF’ and ‘TFF’ as a format and .otf and .ttf as file extensions. Some types of OpenType font can have either a .ttf or an .otf extension, but still be the same font file inside.
A file with an .otf file extension is an OpenType font. A .ttf file is either an OpenType font or a plain TrueType font. Which do you mean in the above context?
This you may know already: OpenType fonts come in two ‘flavours’. The OpenType format can contain the font data of two separate formats (TrueType and Postscript) and provides additional common properties and functionality. So, an OpenType font can be either, 1) Postscript-flavoured (more correctly called Compact Font Format, CFF-OT), containing Postscript curves and Postscript Type-1 hinting instructions. Or, 2) TrueType-flavoured, containing TrueType curves and TrueType hinting instructions [1].
CFF-OT fonts usually have an .otf extension, and the TrueType-flavour usually has a .ttf extension, but not always. A TrueType-flavoured OpenType font can have an .otf extension [2].
So, “The glyph is exported as part of a font, in either OTF or TTF format“ is confusing as it could refer to either the OpenType Format font generally (including both CFF-OT and TTF-OT) and plain TrueType fonts, or is referring to just OpenType fonts, CFF-OT and TTF-OT, but not other formats (like plain TrueType). Which did you mean?
To say “TrueType hints are ditched if the font is OTF“, is either confused or wrong. If you mean ‘TrueType hints will be ditched if your font is CFF-OT’, this isn’t quite correct. If you’ve developed a Postscript-flavoured OpenType project, with Postscript hinting instructions, but then export to a TTF-OT font, then your Postscript hints would be lost in the conversion. But there would be no TrueType hints to be ditched. If your project is TrueType-flavoured, and you meant the TrueType hints are ditched if the font is TTF-OT, then this is just wrong; TrueType hints are not lost in a TrueType-flavoured OpenType font.
[1] The FontLab manual has a summary of the technical differences between Type 1 and TrueType hinting: http://www.fontlab.com/fontlab-font-editors/fontlab-studio-8212-professional-font-editor/download-fontlab-studio.html
[2] http://www.microsoft.com/opentype/otspec/recom.htm
posted at 04:31 pm on November 17, 2009 by Brook
7 Re: What do you mean by 'OTF'?
Brook, thank you! This is a wonderful clarification. (And actually, typophiles may have attempted to explain the same things to me back when I offered up the list of steps in a forum thread … I said at the time that I’d need to digest it some, and again here I need to digest what you’ve shared.
Your explanations and links are extremely helpful.
posted at 06:13 pm on November 17, 2009 by Tim Brown
8 Love your project
Yesterday I published Azbuka CSS Typographical Library and I run into same problem, how to test the library. So I assembled HTML for Lorem Ipsum Crash Testing . Web Font Specimen can be great for testing my next project. Thanks!
posted at 10:37 pm on November 17, 2009 by Vladimir Carrer
9 Several Responces
Erwin & Richard
Yes, this transition to allowing font use on the web is a arduous move, but it is happening. It should not be difficult to understand that foundries simply don’t want to put their software completely open and unprotected on a site – and just because browsers now support @font-face doesn’t mean it is appropriate to use immediately… If you want to discuss DRM to help make the foundries feel a bit better, there are no options outside EOT (which no one outside of IE wants to support). Seeing as how customers don’t want DRM anyway, there needs to be a new solution and compromise. Luckily new options are coming just around the corner (& some are already here), new methods to host and render fonts, and of course the new formats – which will finally make most foundries comfortable with web fonts.
“We face an unbridgeable divide” – Richard Fink
This is just melodramatic, silly, and doesn’t help the cause.
Our foundry (for one) allows our fonts to be used on the web. Right now this means .ttf and .eot files – eventually we will move to .woff or .cwt when there is support. We offer a special “WEB Fonts” package that comes with pre-made subsetted fonts in several versions to help with download times. It is fairly “low-tech” compared with something like Typekit or Typotheque’s options, but the advantage is that you have the fonts, they can be stored on your server, and you also receive the “regular print versions” of the fonts to use on anything else offline.
@Tim
Nice job overall. Brook is correct with most of his comments but I think they are somewhat overly critical. He has a point with what he says, but the crux of the whole issue I believe is the misnaming of OTF when you are really meaning CFF-OT. OTF is the general term for both CFF and TTF based outlines, they are both in the OpenType wrapper. It is all just confusing because of TTF, OTF, Postscript all having multiple meanings and implications.
Much of what you say should be more acceptable if you simply replace most of the “OTF“s with “CFF-OT”.
Specifically regarding the comment “TrueType hints are ditched if the font is OTF”: Brook calls that confused or wrong. The way he interprets and explains it, that is true. However, I believe you mean to say that if you convert a hinted TTF font to a Postscript CFF-OT then the TTF hinting would be lost – and that is correct. However, it is an odd comment; why would you do that in the first place? Even if we say you were converting a TTF to CFF-OT, you would just run the auto hinter before generating the fonts, same as if you were making a CFF-OT from scratch. No designers are drawing fonts with TrueType outlines, they are converted to TT later. Therefore there is no real native TT development or reason to convert back to PS.
Cheers,
Rob
posted at 09:28 am on November 18, 2009 by Rob Keller
10 Finally
It’s amazing to finally be able to use alternative typefaces to the ‘web-safe’ fonts we’ve always been forced to use.
I second the suggestion to force browsers to use typography standards, a unified paradigm on type would be fantastic.
posted at 06:52 am on November 19, 2009 by traxor
Got something to say?
Discuss this article. We reserve the right to delete flames, trolls, and wood nymphs.
Create a new account or sign in below if you’d like to leave a comment.
Subscribe to this article's comments: RSS (what’s this?)
1 The biggest hurdle
The biggest hurdle right now are the foundries themselves who, in similar fashion to the music industry, are still caught up in yesterday’s licensing models. I see a huge opportunity here for smaller, independent font foundries to get their fonts online, with Typotheque as an excellent example.
posted at 07:52 am on November 17, 2009 by Erwin Heiser