Please update your browser. Why?
Type Tips

Some Thoughts On Web Fonts by Yves Peters

And this is the second of two interviews used for The Future of Web Typography, a chapter in Smashing Book 2. I submitted my answers to Vivien Anayian around the same time as Stephen. This interview too is presented in exclusivity in its integral version on The FontFeed. I would like to stress again that I am more of an observer, not a tech-savvy specialist. If you think I made factual mistakes in my answers, please feel free to comment below.

What are some of the most horrible mistakes in typography on the web have you seen so far?

Y V E S P E T E R S | You have to realise that until recently the typographic possibilities on the web were quite limited, so I looked at web typography with a completely different set of expectations than when I look at type for print. The worst mistakes are always made by people who don’t grasp the medium very well, who don’t know how to translate their typography for this fluid and fickle medium. People who don’t realise type can be enlarged, so they specify absolute measurements instead of relative proportions. Who don’t take into account the resolution of the screen and format type that doesn’t “sit” well on the pixels. Who use the wrong characters, like acute accents instead of single quotes, or guillemets for arrows. The worst though are those who are so obsessed with using a specific typeface that they convert whole chunks of reading text into images or Flash files.

What advice would you give to designers on how to choose and decide if a typeface is suitable for use on the screen? Is there an algorithm to follow?

Y V E S P E T E R S | There is a whole lot of exploring and testing involved, as there are no real hard and fast rules, except maybe the ones that also apply for print typography. For example typefaces with delicate or unusual letter forms, with small counters, with thin serifs, with closed or geometric character shapes are not suitable for text sizes. But you can’t say sans serifs are better suited than serif faces, because the typeface I personally prefer for continuous reading on screen still is Georgia.

Irrespective of the design itself the hinting of specific typefaces may be poor. Again one has to assess how the character shapes “sit” on the pixels. This last but very important aspect may become irrelevant though, because the new retina display on Apple’s iPhone 4 has a 330 ppi screen. If higher screen resolutions are accepted and become the standard, there may be no more need for screen hinting, and type designers will regain the freedom they always enjoyed when creating faces for analogue media.

Your articles on the FontFeed regarding free fonts are eye-opening. Do you think that all the legal implications that come with using fonts (free and commercial) would prevent web fonts from ever becoming mainstream and used by general public?

Y V E S P E T E R S | The type industry certainly needs to take a long, hard look at itself and ask itself some difficult questions. We should learn from the many mistakes the music industry has made when it comes to online distribution and use of digital music.

Most of the legal bindings in the End User Licence Agreements (EULA) make perfect sense, but I for one am convinced that some restrictions actually hurt only the honest users, whereas the pirates and sharers couldn’t care less. If type users are at least aware of the EULA, or even take the time to read it because they are concerned about what is allowed with the font(s), foundries should reward them by not putting too many stumbling blocks on their path. These users care and want to do the right thing, so the foundries should care for them. Whatever is specified in the EULA, pirates and sharers will always come up with some inane justification for stealing fonts. Please don’t punish the honest users for the actions of those thieves. Restricting embedding or inclusion in web documents will not stop font theft. The pirates will always find a way, no matter how hard the foundries try to prevent it.

How would the introduction and implementation of web fonts affect all the small businesses and individuals who prefer the DYI approach by installing free themes and templates or purchasing inexpensive premium themes, as well as the template and theme designers?

Y V E S P E T E R S | I think it will only affect them if fees are low enough for those small businesses and individuals to not think twice about licensing web fonts, and if it’s sufficiently easy to implement them. Licensing shouldn’t be too cheap neither, because then it would hurt the perceived value of digital type. Below a certain price point people won’t be motivated to license fonts anymore; they will pirate fonts or use whatever is available anyway. It’s not that different from regular design: there’s this threshold defining whether users continue doing things their way, or adopt new possibilities.

On a side note, I am convinced that the most difficult aspect of design and typography is not the actual design part, but convincing (potential) clients of the value of professional design and typography, and the benefits that can be gained from it. If clients can be made to see that using professional fonts or a specific typeface improves the quality of their website and helps them differentiate their business from the competition, then they will readily adopt quality web fonts. The same applies to template and theme designers.

Are there any commercial web fonts available today that are legal to use by designers for their both free and commercial themes and templates?

Y V E S P E T E R S | Yes there are – foundries like OurType for example simply allow editable embedding, while other foundries make you pay an additional licence fee. Some however don’t allow this at all. And then there are foundries like Typotheque that link the web licence to a specific domain name, so to use a theme or template on a different domain requires the end user to license the fonts. There are also free fonts whose licences allow this type of use, although I am not sure if any of them are well hinted. By the way, people complaining about the lack of decent free fonts on the internet are hypocrites. If you read between the lines what they really are complaining about is that their personal favourites are not free. Well, tough noogies…

Would love to read your thoughts on some of the concerns regarding web font delivery/hosting services and font licenses that I’ve read online: If designers/clients already paid the license to use the font for their design applications, why would they now have to pay to use that same font on their web sites?

Y V E S P E T E R S | That’s the whole format versus product discussion. You could compare it with VHS cassettes versus DVDs versus Blue Ray. Why do you need to buy a Blue Ray disc if you already owned the movie on DVD, and even on VHS before that? It is the same movie, the same intellectual property after all. Yet the DVD plays on a different machine, has an improved image and sound quality, and added functionality like language and subtitle selection, extras, and whatnot. And Blue Ray discs offer an even better quality with even more elaborate functionality. Same goes for the difference between PostScript Type 1/TrueType versus OpenType versus Web Fonts. Same typeface design, but different font formats with different functionalities.

What happens if the server where web site fonts are being hosted is down and the web site is reverted to the fallback fonts that may look very different from the original web font?

Y V E S P E T E R S | I am quite confident that web font services, just like regular web hosting services, are prepared for such eventualities, and equipped to remedy this kind of situation as quickly as humanly possible. That’s not saying that there may be a disruption of service at some point, and then users will have to accept that their website looks different for a limited period of time. But if the website CSS has been coded properly and responsibly, the difference shouldn’t be too jarring. You know, I am very down-to-earth when it comes to problems like these. I am a firm believer in the Law of Entropy. Things break, and everyone dies eventually. That’s life for you.

Furthermore this is another problem that may become irrelevant in the foreseeable future. As soon as the major web browsers Firefox, Safari, Chrome, and so on will support WOFF and EOT natively, I think web font hosting services like Typekit, Kernest, Fontdeck et al may become obsolete.

What happens if a client signs up for a type hosting account, chooses the type for the web site and sometime down the road, after the client’s online identity has been well established, that particular typeface is withdrawn from the hosting library/server? What are the implications and solutions?

Y V E S P E T E R S | As far as I know this has not happened before, and I don’t really see why a type designer or foundry would completely withdraw a particular typeface from all web font hosting libraries. In any case the contracts stipulate that even if certain fonts become unavailable for purchase they still will be hosted by the web fonts hosting service. But again this is a problem that may disappear with native support of WOFF and EOT in web browsers.

As Stephen Coles rightly mentioned in the Web FontFonts article on The FontFeed, one of the advantages of web fonts is that companies and organisations that use a certain typeface for their identity can now easily extend their brand to their online presence. However, what if the font in question is not available for download as a web font, nor is it available via web font delivery/hosting services such as Typekit? Worse, if the brand’s typeface is not suitable for the screen? How do you see that being resolved in future? Would such brands need to find/use an alternative font online?

Y V E S P E T E R S | What we are experiencing since a few years is a gradual but dramatic shift to an increased importance of online reading. Web Fonts are in their infancy, and I see them continue gaining importance in the near future. More and more typefaces will become available as web fonts, so it should be just a matter of time before this problem becomes irrelevant. The typeface not being suitable for screen will also be remedied when higher screen resolutions (like the iPhone 4 300ppi Retina display I mentioned earlier) become standard.

You also have to ask yourself how we have been doing this up to the advent of web fonts. As a graphic designer I am very pragmatic myself. If I need to choose a colour for a logo, I make sure that I don’t specify an RGB or PMS colour that cannot be approximated in CMYK for print. Similarly, when picking a typeface for branding, map out all the different possible uses in print and on screen, and don’t paint yourself in a corner. And responsible designers have always suggested alternative typeface choices for online branding, so as long as it’s needed they will continue doing that. If you’re smart you try to foresee what’s going to happen in the near future, and design accordingly using the tools that are available to you today.

Would they be able to purchase an individual license to use the web version of that font online? Of course, they can always switch their brand’s typeface or use a different one online as they do now, but I’m just curious in general terms what’s the future of a branded typography on the web?

Y V E S P E T E R S | I think most type designers or foundries would happily oblige and provide a web font version for an extra cost. Almost all the ones I know are very customer-friendly. Plus I am pretty confident they would rather have their designs out there, being used in real life applications, than (potential) customers turning to the competition. Not everything is possible, but quite a lot. And like I mentioned in the previous answer, this is only the beginning. I expect things to evolve rapidly and for the better, improving web typography in ways we can’t begin to imagine.

Lastly, do you think that the introduction of web fonts would re-start the fleet of MySpace-looking web sites with a horrid typography? If yes or now, then why?

Y V E S P E T E R S | Frankly I was never really bothered with the cringe-worthy MySpace look, just like I was never bothered with the torrent of poorly cobbled-together typefaces that appeared when grunge design exploded in the early to mid-90s. Every new technology offering new possibilities generates a wave of unbridled enthusiasm. This inevitably results in an incredible amount of amateur and poor quality work. But after this initial wave has subsided usually a new generation of creators mature, enriching the field. The grunge boom is responsible for the fact that we now have more type designers under or around thirty than ever before in the history of type design. That’s incredible, and typography is truly living a golden age. So I am really looking forward to discover what the effects of the advent of web fonts will be in five to ten years. And I am more than willing to accept all the crap design that has to come with it.

Goofing off at a Troubleman rehearsal. “Your puny rhythm section can’t touch ours.” © Yves Peters
Goofing off at a Troubleman rehearsal. “Your puny rhythm section can’t touch ours.” © Yves Peters

Source: recovered from FontFeed

* * *

FontShop Letternews

Subscribe to our newsletter to receive amazing offers, useful type tips and information about the latest font releases.