On the error, and the inspiration, of randomness

by Michael S. Kaplan, published on 2007/08/12 04:01 -04:00, original URI: http://blogs.msdn.com/b/michkap/archive/2007/08/12/4347121.aspx

This is one of many posts that was written around the time that TypeCon2007 was going on, and edited later....

One of the more fascinating of the people who presented at TypeCon2007 was Professor Edward Mendelson, whose presentation Dispatches From the Font Wars went through not just the tumultuous period of competition for the pre-eminent font standard, but also many thoughts on the casualties of the war (and especially of the subsequent peace).

I had the chance to talk to him afterward and between his presentation and especially that later conversation I really feel like I gained a few insights into the nature of the "font peace" under which we currently live, as well as a few of the flaws in that peace, many of which continue to haunt us to this day....

I found the presentation to be fascinating, from his initial W.H. Auden quote ("A professor is someone who talks in other people's sleep") to the very end, I wished that I was talking about something that would allow me the opportunity to present the same style of presentation (nothing at the moment qualifies but maybe one day....).

Who better to quote Auden than his literary executor? :-) I think he was mildly impressed that I recognized the quote though he'd probably be less impressed if he realized I did not realize it was Auden until later!

Stepping away from the lost technologies (not sure what to call their cause of death -- "friendly fire" maybe, with the emphasis on the fact that the term is in quotes since not all of it was friendly!), there were a few interesting issues that I recognized, some that I talked with him about at the time, and some that only occurred to me later.

I thought I'd talk a bit about some of those issues in a blog post or two. If that does not sound interesting then you can probably skip the rest of this post....

One of these points was inspired by an issue he mentioned in the talk -- the loss of the randomness that pre-digital typography gave the process -- from quality of the metal/wood to variation in either of them to the splash of the ink, so many of the "random" factors that gave the type and the fonts some of their character and charm were simply not present in digitized type. And although it would be possible to emulate such things using current technologies, efforts to do so within Microsoft or any of the large foundries seem at best minimal.

The randomness is not the issue I was referring to, here. Instead what occurred to me is perhaps some of the excuses for the lack of interest in reproducing or emulating it in digital type, and perhaps even some of the reasons.

What occurred to me was a (and perhaps someone say one of the) fundamental distinction between the engineer and the artisan (which is I think a very reasonable term to put on those who create and design fonts -- not all they are, but all that is relevant to this particular conversation).

To an engineer, the various factors that made the pre-digital type random were not intentional -- whether one looks at variation or wear in the metal, the extra splash of the ink, etc. And to an engineer, design is largely the product of intent, not of accident or mischance or error. Innovation, a word bandied about far too often in these contexts, is not an accident. Thus at a very fundamental level, the factors that led to randomness are easy to write off as bugs, especially given other requirements on the design like repeatability and reproducibility (one can't have document flow vary from run to run or page counts and layout could vary, for shame!), it is easy to lower the priority on the requirement.

To an artisan -- be they font designer, typographer, designer, whatever -- design is largely (at least ideally) the product of creativity. And creativity does not fear accident or mischance or even error, as these factors can often enhance a design and push it into previously unrecognized directions.

However, the artisan is in this case largely dependent on tools provided by the engineer, a factor which could easily impoverish the spirit of the artisan were the wells of the best typographers not immeasurably deep.

However, being forced to work within these tight boundaries is not a situation that will be comfortable forever. One can put as many knobs and options into the tools as they want, the Knuthian/Hofstadterian METAFONT would not be able to capture all of the possible factors that randomness can add to the mix.

I'll talk about this more soon, as I have some thoughts on some potential directions for solutions (as well as some blind alleys!).

FYI, Inspiration does not come from motivations. It's in the reverse.

-Dan Grosp
CEO
grosp.com

Well, as I said at the end of Edward's generally great talk, we know of three ways to get randomness in OpenType fonts. One is obscure and unsupported (CFF supports a random operator internally, like Type 3, but nothing supports that AFAIK). One is simple but not supported by applications (the OpenType 'rand' layout feature). The third is to do pseudo-randomness with contextual layout, which I and others have coded and demonstrated pretty extensively.

Cheers,

I was thinking about this (it inspired my words "And although it would be possible to emulate such things using current technologies..."), though it is not entirely clear (to me) whether this is the same thing (though I admit to perhaps not understanding the undrlying technologies behind all of those chioies so I may well be in error!).

Please consider a donation to keep this archive running, maintained and free of advertising.
Donate €20 or more to receive an offline copy of the whole archive including all images.