Perfect Visual Repetition verses Individual Charactar

Ryan Maelhorn's picture

If you've got an 'h' you've got an 'm', an 'n' and a 'u', and sometimes a 'w'. You almost have a 'b', and once you have a 'b' you have 'd,' 'p,' and 'q.' That's about 1/3 of the lowercase alphabet from two characters.

When type design started of course this wasn't really the case, you could say you had the design for all these letters based on the h and the b, but you still had to carve the shapes out of metal by hand anyway. Today, we can simply copy, paste, rotate and flip in an illustration program or in fontlab.

It seems like a cheat though doesn't it? And anyway it seems that to make things more legible, at least to dyslexics, you want these characters to have differences. Now, not everybody is dyslexic, but I would argue that in some way, what makes things more legible to dyslexics, makes them more legible to everyone else as well.

So you make them different, because its better, and because its boring not to, and because you don't want to feel like you're cheating.

I started off trying to make all my fonts perfect. Perfection was the goal. Oh I was so enamored by the idea that type design is at the same time BOTH so right brain AND so left brain. Science! Everything as similar to everything else as possible. Measurements! The exact same bowl on all the bowled characters. The w should look as wide in the uppercase as it does in lowercase. This gives a better visual repetition. It looks better. The text becomes a more beautiful pattern. But, does it read better? More important to me, would it still possess character?

By making the letters more similar to each other, more perfect, perhaps what I was really doing is making the whole thing more boring.

The quirky bowl of the lowercase p was changed to match the bowl of the q. Now it looks more legitimate, more serious... Or does it? Maybe it is just more boring. the exact left brain nature of science and logical thinking may not have been quite as important as I would have hoped...

So then I started working to make all the glyphs as different as possible. To head in the other direction from perfection -- entropy.

Now everything is different. The set doesn't look like it belongs together. The x height is all the same, the baseline is observed religiously, but still these characters don't look like they belong together.

So I head back in the other direction, taking little bits and pieces and copy and pasting them into other glyphs, trying to make them look more related. But is it what I am really doing just making the font lose all its character?

There is no end to this story, no nice neat moral waiting for us at the end.

I hope to start a conversation about these ideas which are probably more important in type design than in any other art. I'm sure anyone who has been at it for awhile has had to deal with these issues and may like to impart some wisdom.

hrant's picture

This is a good journey.

Do contemplate the existence and relevance of the Display-Text axis, with any font falling somewhere in between the two non-existent extremes. But there's also a threshold, somewhere on the axis.

BTW many punchcutters "cheated" too, by using counterpunches.


Syndicate content Syndicate content