Bruno Maag Interview: Designing Type for Other Languages and Alphabets
Bruno Maag heads up the London-based type-design studio Dalton Maag, which works with very large clients such as Nokia and Ubuntu to design typographic systems that often cover several different languages, sometimes in entirely different writing systems. While Bruno was in Seattle for a client meeting, John D. Berry got him to talk about his company’s work and his approach to design.
Interviewer: One of the most interesting things that you’ve been doing is multilingual, multi?script interfaces. Has that always been part of your brief? Or part of the agenda?
Bruno: Not really. We’ve always wanted to do multilingual typefaces, and for the best part of the last ten or fifteen years we have done Latin, Greek, and Cyrillic. With Greek and Cyrillic today, you really have them kind of all bundled together with Latin, and it’s more or less the same. I know that some people out in the community will kill me instantly for saying this! But the reality is, the forms are ultimately familiar to one another, that’s probably what I mean by this.
Interviewer: They have common roots.
Bruno: Common roots, indeed. So from that point of view, I think you can bundle then together as a script group if you wish.
We have been wanting to do Arabic for a while, simply because I like Arabic scripts – I think they’re just very, very beautiful.
The first big job where we started getting into that arena was the fonts we designed for Ubuntu. Earlier this year we released our Arabic and Hebrew to Ubuntu, although I don’t know whether they have uploaded them yet. Hopefully they will come soon as well.
Ubuntu Hebrew and Arabic in development
Then of course Nokia hit us, in a major way. Originally the project was only supposed to be Latin, Greek, Cyrillic, Arabic, Hebrew, Thai, Devanagari…
Interviewer: Oh, only that?
Bruno: ”Only” that! You’d have to work for the best part of two years. But probably about three months into the project, all of a sudden they came to use and said, “Can we change the order? We need Ethiopic, Armenian, and Georgian. And to follow that, we need Bengali. We need this…” They just kept throwing it in. “Oh, we need Khmer, as well. Can you think about that?”
So currently with Nokia we’re at 19 writing systems released, including Chinese. We’ve finalized Malayalam and Telugu earlier this year. But of course, because Nokia is now “in a change,” if you wish, I’m not sure at the moment whether the project is going to continue. There’s certainly a desire from various groups within Nokia to continue the project. I think everyone sees the value of it, the impact. So we’ll see. But on the back of that, because we’d been managing so much publicity, we then worked with HP. They needed fairly extensive scripts, again: Latin, Greek, Cyrillic, Arabic, Hebrew, Thai, Devanagari, and CJK.
A small sample of Dalton Maag’s work for Nokia
And on the back of all those, we’ve just started working on another three major projects, including the one that I’m here in Seattle for. Which are multilingual as well.
Interviewer: How do you go about doing that? In other words, there have been many debates about “harmonizing” type designs across scripts, for and against them.
Bruno: Yeah. I do believe that, first of all, you can create a harmony, you know. But creating harmony doesn’t necessarily mean that you have to Latinize everything. You can do a harmony by respecting the traditions and the cultures of the relevant script systems.
The main thing is that you’re looking at the texture of the type. I mean, when you set two script systems against one another – and it doesn’t matter what script systems they are! – when you look at the tonal values, and they have even tonal values, that when you look at it – I know this is all very esoteric! – but you look at it, that the rhythm feels the same, and that patterns are maybe similar, as well. Even though some people will say, “How can you compare a southern Indic script to Chinese in patterns?” But there are patterns that emerge that you pick up. I think those little details, those things are the same.
You also want to look at calligraphic styles. Obviously the Latin writing system, as far as typography goes – again, I’m getting myself in very hot water now – I believe typographically it may be the most advanced. Historically speaking, the Latin world has historically had typography since Gutenberg, which has bypassed many other cultures, simply through technological developments.
All these cultures have a very, very deep and long calligraphic tradition, but at the same time, typography has bypassed them. So you can look at the various calligraphic styles, and there have been a lot of developments; just look at the Arabic, there’s at least a dozen calligraphic styles. So you can maybe see which calligraphic style does best fit your Latin design. What rhythm is most in keeping with your Latin design? Then you start adapting.
I think that’s the way to go forward. Once you have worked on those rhythms and textures, you then start looking at details. Are there curve treatments, for example, that you can start harmonizing, certain tensions in strokes? How do you deal with character terminals? That kind of thing. But also, width and height proportions, is there something that you can do there? If you have a very condensed Latin type, what can you do in a Telugu, which has a te
ndency to be all round and very broad? Can you actually tighten that up a little bit as well, so it has this more condensed feel to it? You’re looking at all these various things.
But of course, throughout the whole process, you will always work with native readers, designers. We also work with linguists. We will have a whole group of native readers on a script to make sure we get appropriate feedback. We tend to work with a variety of people, from a very conservative, almost orthodox end to a very liberal end, to get different opinions about things. From that we can distill the essences, and we believe we get a very good result.
Interviewer: Is it always designed in?house?
Bruno: With the exception of CJK, at present, we do everything in?house.
Interviewer: That’s a lot.
Bruno: It is, and it’s been challenging. I think that’s putting it mildly.
It’s been very challenging because in some places you go and there is absolutely no research material. Bengali is particularly difficult. For one thing, in this script system, you start looking at all the conjuncts, and it’s just random. Looking at Devanagari, you look at conjuncts, yet you can see what the base characters are. You can disassemble it and bring it back into the individual parts. It’s actually fairly systematic. There’s clear rules.
But with Bengali, you start looking at conjuncts, there is absolutely no relation to the base characters. When we were working on the Bengali for Nokia, it was so bad that linguists didn’t agree whether certain conjunct forms were actually acceptable or not. That’s what we were facing, and there is no precedent. The only precedent you can find is the printed examples that Monotype and Linotype did, back in the ’50s and ’40s, when they started setting this stuff. But even those samples are very limited because of the technologies that were at hand. Character sets are limited, conjunct sets are limited, etc., etc. You don’t really have anything, and in calligraphy people do whatever they want to do anyway.
It was really, really tough, for a lot of these things, just to get the research. In the beginning we vastly underestimated how big a bite we took. With Bengali we thought, “OK, we can get that done in three to four months.” Eventually it took nine months to do it. Three to four months was just the research period! At the same time, not only is it a design challenge, you then have to deal with all the technological challenges as well.
If you work for a client like Nokia, they have such a broad requirement list. Whilst you do everything with OpenType technologies today, and broadly it works, you then still have to start developing specialized solutions for certain devices, for certain environments, and you have to constantly test. It is hugely challenging. I would advise anyone not to take this on lightly. You can save yourself a lot of hassle!
Riding the storm: when the company has to grow
Also, for the company, it was extremely challenging, not only as a project but just to handle the growth. Once Nokia started throwing all this work at us, we had to grow very rapidly. That meant we hired people, tried to get a lot of people from Reading (MA in Type Design), so people who already had a decent understanding of type, and some of the type technologies. I remember one girl, Hanna Donker, the day she started we said, “OK, you’re now doing Kannada.” She had never looked at Kannada before. “This is your job now.”
We had to throw people in at the deep end. For the best part of two years, where we grew from 7 people to about 40 people…
Interviewer: In two years?
Bruno: In two years, yeah. And we still continue to grow. We are over 50 now. It was a very stressful period for everyone involved. As a business, we couldn’t develop strategies, it was just firefighting, surviving from one day to the next.
From the project point of view, it was extremely satisfactory. When do you ever get the chance to work on a fantastic project like that? But in the way the business was developing, not very satisfactory at all. It’s only been the last twelve months or so that we started to get the beast under control. We started to calm everything down.
What we’ve done now, as well, because we have managed to calm it down, we’re putting people through training internally, even retrospectively – people we hired two years ago are starting to go back into training – to make sure that everyone is following the processes, and that we can establish clear ways of doing things.
The way of doing things may be wrong, but at least everyone is doing it wrong, so you can change it, and you have consistency. That’s the main thing, consistency across everything you do.
Everyone who’s joining us now, they have to go through a four? to six-month training, depending on skill level, before we let them loose on a project. Irrespective of where you come from, irrespective of your background. “This is what you’re doing, this is the process you follow,” and so on.
It’s already paying dividends. The new recruits that joined a year ago have gone through the program and their working methods are completely different; much more effective, much more efficient, and that helps.
All round, the last two or three years have been a sensational learning curve for anyone involved.