Preliminary Thoughts On My Font Project Brief

Next up is Week 4. By the end of the week I have to hand in a draft of my font project brief, which means making some decisions.

One of the cornerstones of information design is putting users first. Who are the users of free software fonts such as the one I’ll produce for my project?

Graphic designers are the typical users of font software. But most professional graphic designers are not yet using free software, so they are not my primary audience, despite being the largest. I want to do something that benefits free software users primarily, and people using proprietary software is incidental - although if the fonts are buggy for proprietary software, that would be neglecting the largest user group, but to concentrate on them isn’t useful for the free software movement.

What kinds of free software users are there?

There is a continuum along software-freedom lines, which is something like this (and most people blur across the roles, of course):

  1. At one end, we have the most ardent Apple/Microsoft/Adobe/FontLab fan, who finds it inconceivable that software not produced by a company and licensed under proprietary terms is of any value at all, and would never share programs with friends because “piracy funds terrorists.” They’ll dismiss any “free font” as poor quality, and avoid engaging with the software-freedom concept. Maybe 1% of users are like this, although there seem to be a lot of graphic designers who are more susceptible to the kind of Cult of Mac mentality.

  2. Then we have people who use Mac OS X or Windows but are interested in tools, often being something-developers themselves, although not neccessarily programmers. Web developers and such. They will use the “best tool for the job,” and don’t mind about who controls the tools they use, and don’t mind agreeing not to share with their friends. Of course, if a program that is “the best” on merely practical values is, by chance, free software, they’ll use that. Being developers, sometimes having freedom makes it better, but freedom isn’t a primary value - so if they develop something functional, they don’t think that their users deserve freedom. Probably about 5% of computer users, I guess. Many businesses are like this, because they are not able to break the agreements that restrict the use and sharing of software.

  3. Then there are people who use Mac OS X or Windows, and will use the tools they heard about without too much consideration, and don’t mind about who controls the tools. They enjoy sharing with their friends, even if that means breaking agreements, because they don’t think at all about the issues. This is the majority of computer users. As the anti-user anti-sharing programs like Vista and iTunes become more draconian, the issue of freedom will become more visible, and since they don’t need “the best” functonality, they are more likely to switch to GNU/Linux than the previous set.

  4. Similar to them are the people who use GNU/Linux, and prefer software freedom, but not strongly; they will use the “best tool for the job,” and don’t mind too much about who controls the tools, but they have experience of an ethical and sustainable way of life. They enjoy sharing with their friends, even if that means breaking agreements not to though. The typical Ubuntu user, installing Google Earth and (probably an unauthorised copy of) Photoshop under WINE, for example. 8% of users here.

  5. Then, people who use GNU/Linux, and strongly prefer free software over “the best” because they do mind about who controls the tools. However, not where it really inconveniences them, like with wireless network drivers. To legally share software with their friends is a minimum, though. These people are often tool-makers, and software they write respects users freedom. The leaders of the Ubuntu project, that includes proprietary graphics card drivers and Adobe Flash Player, for example. This is maybe 10% of GNU/Linux users.

  6. Finally, there are ardent software freedom fans, who find it inconceivable that software not sharable or under the control of its users is of any value at all, other than to provide models for their free replacements. This is maybe 1% of GNU/Linux users.

That is very broad and not specifically about roles related to type design. A friend suggested a user-developer-support model of such roles, which I interpreted as:

  1. Users who use free fonts in free applications on a free operating system. This is my core audience - any GNU/Linux user. A wider audience is anyone who uses free fonts, but if they are using a proprietary OS or application, I’m not too fussed about them.

  2. Developers who make free fonts. They might use proprietary software to do that (ie, Gentium) but since all font formats are well supported in FontForge, that is not too problematic.

  3. Supporters who want free tools to make fonts, but not to make free fonts. When they speak about “piracy” I will be pissed off, and probably when I speak about how all functional works should respect users freedom, they will be pissed off. Hopefully not so much that they don’t want anything to do with free tools to make fonts; I think despite our differences, we can still work together on the tools :-)

The supporters issue is tricky; I’m not going to be quiet about the issue of software freedom. If people choose to use free software because it is better on shallow criteria, like it is faster at common tasks or crashes less or has convenient features, that is good. But it is not great, because such users will choose a proprietary tool that is even faster, crashes even less and has even more convenient features if one comes along. It is only by people learning to value their freedom that they will keep it, and it is far easier to let it go than to get it back.That learning can only come about through carefully thinking about the issues.

So, what is the user-scenario of my font project?

Since the MATD course has a focus on print, perhaps documents typeset in Scribus and TeX will be the primary one to concentrate on.

However, I feel on-screen reading (in Firefox) is very important. Webpages are the way people read text and see graphic design more and more, and personally, I hardly read anything on paper any more - although that has just changed dramatically with all these typeface design books at the foot of my bed, haa

I forgot where, but in the last few weeks I heard something like, the web is now 10% of the general population’s advertising exposure in the UK. And that is going up and up, eating into TV and print publications, while billboards and radio ads are steady, since we are exposed to them in our cars. The upcoming CSS3 web font downloading mechanism is going to be a kind of “uranium” that will power the font-freedom movement and businesses too, and isn’t tied to any location :-)

However, the big complication is the complex script. We’ve been asked to consider early-on which complex script will complement our latin designs, and I am thinking of, maybe, Malayalam. This is the script used in Kerala, the most literate area of India, and there is a lot of free software happening down there. Plus my friend Kaveh Bazargan of River Valley runs his company in that region, and I hope to visit that area in the next few years with him.

Free software is a big growth area in computing, and I think that theres a lot of money to be made from helping people switch to free software. Perhaps that growth is less likely to be so stellar in “center” country design firms where the overall economy isn’t growing so much, and proprietary software is entrenched, compared to the “periphery counties”.

I think that software freedom is perhaps more relevant for the periphery too; I can’t change FontForge myself - but I can hire anyone in the world to do the things I want it to do. In a peripheral country, this means hiring people in the periphery instead of the center, so that the periphery becomes its own center. After all, the issue is freedom, not price.

I’m interested in the growth areas - as in, people who are not already using computers in the way they are used in the central countries. “The future is here, its just not evenly distributed yet.” But really, I have no idea how computers are used outside the center. To really get a handle on it, I need to travel to some peripheral areas and see what its like on the ground. LEDCs are also an option, with SIL and OLPC (and my friends at aidworld) already in the space. However, in the central countries, businesses already paying for proprietary licenses have a cash-saving motivation for switching, offset against their investment in staff training and existing data that makes overall switching costs higher. Its tricky, so I emailed a dozen or so free software mailing lists for suggestions, and I’ll make a separate post about that later.

The central challenge for me will be keeping a focus on the typographic problems, I think: I conceive of fonts as a specialised kind of software, and am interested in the overall and somewhat abstract design process. I have little to say about typographic problems, because I have no depth of experience at resolving them. I’m more interested in investigating the software aspects, but, I can do that on my own sweet time. The MATD course presents a unique opportunity to dig into complex scripts and see how far FontForge’s OpenType feature support can be pushed.

So, this week I have to do a TONNE of sketching of my own latin forms, and play with some Malayalam shapes, and decide roughly what kinds of shapes I’d like to work with. I want to do something that is serifed, designed to work primarily at 12pt on screen for text and 10pt on laser prints from Scribus. I’d also like it to work well at 48pt+ in more artistic graphic design.

I also need to sort out a digital camera this week, as I’m not yet able to post images online. Perhaps I can borrow a camera from the Dept though.

(The center-periphery conceptualisation of national economic development is from top design theorist Gui Bonsiepe, whose work I was introduced to by my good friend Gustavo Ferreira :-)

Creative Commons License
The Preliminary Thoughts On My Font Project Brief by David Crossland, except the quotations and unless otherwise expressly stated, is licensed under a Creative Commons Attribution-Share Alike 3.0 Unported License.

Comments

3 Responses to “Preliminary Thoughts On My Font Project Brief”

  1. Gustavo Ferreira on October 29th, 2007 16:47

    hello dave,

    The center-periphery conceptualisation of national economic development is from top Brasilian designer Gui Bonsiepe, whose work I was introduced to by my good friend Gustavo Ferreira

    no, no, no! :-)

    gui bonsiepe is german, he was a student at the hfg ulm under tomás maldonado, and later a teacher there too. both are among the most important theorists of design. bonsiepe’s ideas brought him to work and teach in latin america in the 60s, but he has also taught in germany (at the kisd in cologne).

    you should definitely read more about the hfg ulm, bonsiepe and maldonado, their ideas will help you structure your discourse about free & open fonts.

    there are many books about the hfg ulm. i just looked at wikipedia, and there are no entries in english about the school or about bonsiepe and maldonado! the entry about the hfg ulm in german is very good though, try babelfishing that. (this is a good example of what bonsiepe calls language universes… ;-)

    esdi, the school where i studied (and also the first design school in latin america), was founded by former hfg ulmers. and the hfg schwäbisch gmünd, where i did my master’s degree, is a direct descendant of the hfg ulm.

  2. David Crossland on October 29th, 2007 18:04

    Ah, okay - I’ll update the post accordingly. I thought Gui was a Brasilian who studied in Germany, like yourself.

  3. Gustavo Ferreira on October 30th, 2007 09:53

    hello dave,

    this is long. are you ready?

    i find some of your ideas a bit confusing, which is normal at this stage. i hope i can help.

    i think i know where the source of the main confusion is: you seem not to be clear about the distinction between typefaces and fonts. (?)

    “A typeface is a design, a font is the particular means through which that design is accessed and utilised. So, classically, a font of type was a tray of metal sorts in a particular type design at a particular size. Later a font was a photographic source on film or glass. Now a font is a computer file or collection of files, depending on the particular format.” – John Hudson somewhere on Typophile (sorry, i’ve lost the url.)

    One of the cornerstones of information design is putting users first. Who are the users of free software fonts such as the one I’ll produce for my project?

    information design –> type
    free software –> fonts

    information design refers to visual communication, so here we are talking about type. whether the software which carries the typeforms is free or not doesn’t matter from the point of view of information design; the lettershapes matter.

    Graphic designers are the typical users of font software.

    graphic designers use type in their work, together with colors, shapes, illustrations, photos etc.

    the pieces created by graphic designers are read by the intended target audience(s) (readers/viewers).

    today, graphic designers use type in form of font software. (yesterday they used drawing/photcopy/letraset to make layouts, and photocomposition to generate the final artwork. and before that, metal types.)

    but fonts are also used by computers to display text on screen so humans can read it.

    if we talk about the shapes of the letters on the screen, then we are talking about screen type. if we talk about screen fonts then we mean truetype fonts, bitmap fonts etc.

    But most professional graphic designers are not yet using free software, so they are not my primary audience, despite being the largest. I want to do something that benefits free software users primarily, and people using proprietary software is incidental - although if the fonts are buggy for proprietary software, that would be neglecting the largest user group, but to concentrate on them isn’t useful for the free software movement.

    the font formats in use today (postscript type 1, truetype, opentype) are all open and well documented. (although postscript was a closed format until 1990, and the truetype interpreter is patented.)

    font licenses don’t have any influence in the support of fonts by graphic design tools – font format has. (but the license has influence in other aspects of the use of fonts, such as redistribution, modification, embedding etc.)

    if a font is buggy in one application and not in another, this is probably because each application is interpretating the specification differently. (this is why specifications should not be ambiguous.)

    What kinds of free software users are there?

    as a type-designer i think you should be asking: “what kinds of use(r) s for typefaces are there?”

    There is a continuum along software-freedom lines, which is something like this (and most people blur across the roles, of course): (…)

    i won’t comment on this part because i think it goes away from the topic type-design.

    but regarding free software adoption in the peripheral countries: have you read this?

    http://www.linux.com/articles/59637?tid=51&tid=11

    Since the MATD course has a focus on print, perhaps documents typeset in Scribus and TeX will be the primary one to concentrate on.

    i don’t see how the fact that the fonts will be used in scribus or tex would influence the shapes of the letters (type) or the font data.

    btw have you already seen this?

    http://www.typefacedesign.org/resources/essay/JeremieHornus_Essay.pdf

    also, vincent connare is close by, no? doesn’t he work at dalton&maag?

    However, I feel on-screen reading (in Firefox) is very important. Webpages are the way people read text and see graphic design more and more, and personally, I hardly read anything on paper any more - although that has just changed dramatically with all these typeface design books at the foot of my bed, haa

    i agree with you, and this is why i chose to design an outline screen typeface at t&m (a truetype hinted font).

    but designing type for the screen is very complicated, there are too many variables: operating system – windows, macosx, gnu+linux; rasterization mode – b&w, anti-alias, sub-pixel rendering; rasterization engine – every os has one (or more than one!), some applications have their own; screen resolution – it varies a lot, specially if you consider handheld devices.

    if you want to cover all possible scenarios you will probably have to make a truetype hinted font – and truetype hinting is not something you can learn in a month. or six. (ok, maybe you can, but at the expense of learning how to design type. and i think that, while at reading, you should prioritize learning type-design.)

    and with the new version of cleartype in vista things got even more confusing. the new cleartype supports sub-pixel positioning; to make use of this, some horizontal instructions are ignored by the rasterizer. which ones? it’s not documented! the ms cleartype fonts have a special bit telling the rasterizer that they have been hinted for cleartype; that’s not documented either! (this was mentioned in a discussion on the atypi list. or typophile, i don’t remember.)

    do you get the picture? it’s insane.

    if your target audience is gnu-linux, then your primary rasterizer is freetype. afaik freetype has a autohinter which gives quite good results – so one idea could be to design a font optimised for tt autohinting.

    i think you should avoid truetype hinting. it’s an amazing technology, but it’s too much for one year and i don’t see much future for it. macosx completely ignores tt intructions(1), and microsoft is playing dick dastardly(2) (hahaha, just found out his name in english) about it by not documenting the changes under vista.

    (1) it’s possible to turn it on with a hack, but it looks really ugly in the context of the all-smooth cocoa gui

    (2) http://en.wikipedia.org/wiki/Dick_Dastardly :-D

    there were some very rich discussions on screen type and screen fonts in typophile while i was working on my final project at t&m. i have made a summary, you can find it in the link below:

    http://www.hipertipo.net/tiki-index.php?page=TypophileClearTypeDiscussions

    and then there is another possibility: david berlow has presented an alternative technique, which consists in abandoning the idea of a single scalable truetype-hinted outline font for a series of single gridfitted-by-design fonts – like my Elementar bitmap font system, but with continuous shapes, so it can use the extra tonal resolution of anti-alias and sub-pixel-redering environments (no matter which rasterization technology).

    http://www.rogerblack.com/blog/screen_fonts_history

    The upcoming CSS3 web font downloading mechanism is going to be a kind of “uranium” that will power the font-freedom movement and businesses too, and isn’t tied to any location :-)

    uranium is very powerful, but very dangerous too… :-)

    http://typographica.org/001112.php

    The central challenge for me will be keeping a focus on the typographic problems, I think: I conceive of fonts as a specialised kind of software, and am interested in the overall and somewhat abstract design process. I have little to say about typographic problems, because I have no depth of experience at resolving them. I’m more interested in investigating the software aspects, but, I can do that on my own sweet time. The MATD course presents a unique opportunity to dig into complex scripts and see how far FontForge’s OpenType feature support can be pushed.

    yes – make this your mantra! :-)

    So, this week I have to do a TONNE of sketching of my own latin forms, and play with some Malayalam shapes, and decide roughly what kinds of shapes I’d like to work with.

    sketching is very important to train the eye and the mental image of the shapes. and by that i mean real sketching – with paper, black&white ink, transparent paper, photocopy etc.

    this is what i use in my own work, and in the workshops i gave in the beginning of the year:

    http://www.hipertipo.net/tiki-index.php?page=MaterialDeDesenho

    i usually start drawing in small sizes, in “low resolution”, then use photocopy to make it bigger and add refine the details.

    in the link below is a study i’ve done last week, based on a print of an existing font – do you recognise it? ;-)

    http://www.hipertipo.org/stuff/estudo_01.png

    leaving your laptop at home might help you to focus ni drawing. i’m serious.

    I want to do something that is serifed, designed to work primarily at 12pt on screen for text and 10pt on laser prints from Scribus. I’d also like it to work well at 48pt+ in more artistic graphic design.

    for screen type it makes more sense to think about size in terms of ppems (pixels per em), not points.

    that’s it for today! i hope you find it useful.

    good work, - gustavo.

Leave a Reply