i am currently investigating possibilities to randomizing handwritten fonts to enable the placements of variable glyphs. i am new to this subject, so any input would be highly appreciated. thanks.
The only folks I know who have done work with truly randomizing glyph selection are LettError, with FF Kozmik. http://www.letterror.com/foundry/kosmik/index.html
However, FF Kozmik is not a handwriting font, but hand-printing.
Several handwriting fonts have been developed that use elaborate contextual rules which produce a variable effect similar to randomness, though in fact it is deterministic. One is Linotype's Zapfino, in the AAT TrueType version of the font that ships with Mac OS X. I am having difficulty finding a good link to something that explains its contextual goodies, however.
Two other contextual-letterform type families are Caflisch Script Pro (Robert Slimbach after Max Caflisch) and Bickham Script (Richard Lipton after George bickham). These two are in OpenType format, and can display their functionality in applications such as Illustrator CS and InDesign. By way of example, Caflisch Script Pro has 12 different forms for the lowercase "p," and picks one depending on the context.
For more information on OpenType, see http://www.adobe.com/type/opentype. See particularly the OpenType User Guide PDF (linked on right).
after an email conversation with luc devroye i am left with the knowledge that he and the guys of letterror seam to be the one of the only people who have worked on random fonts. is that true? can you think of any other examples? am i wrong thinking that the random placement of variations of glyphs is an important step in the development of typography?
i love this forum!
constantin
ParaType in Russia has worked on fonts with random elements. You might need to contact Emile at ParaType directly as details may no longer be on the Website.
I think Hrant is correct WRT mistakes - the direct-mail industry, which uses handwriting fonts from vletter and others use the 'crossed-out-words' method. vletter include back-spacing crossed out glyphs in their fonts, I beleive.
Cheers, Si
One point of concern I'd have is that should you design something with a random font and save it our output at a service bureau, they may open up a different setting of the font and the really nice random combination you saw on your computer will look totally different when you send it to press assuming you don't convert the text to outlines first.
Beyond that, making an intelligent OT version of the font will preserve the text setting you intended to reproduce.
Stuart :D
OpenType as a format has this capability. However, AFAIK it is not yet supported in any application, or any font. Probably the last two are a chicken-and-egg situation.
Personally, I don't think that random glyphs are an important step in the development of typography. Interesting, yes. Important, I don't think so.
Cheers,
T
Sure. Assuming you want it to work in some existing applications, you have an unexciting choice insofar as each approach is very limited in how/where it works:
- OpenType (contextual, WYSIWYG, currently only supported by Adobe's apps, any output device, easiest to code but that's not saying a lot, use FontLab or maybe DTL FontMaster to make it)
- AAT (contextual, WYSIWYG, Mac only, currently only supported by a handful of apps, any output device, hard to code contextual stuff, use Apple's command-line tools to compile the features into a pre-existing font)
- Type 1 (true random, non-WYSIWYG, will work in many apps but not most recent Adobe apps, will only work with PostScript output devices, you'd probably have to be a programmer to code it)
I hope that helps clarify your choices,
T
> I don't think that random glyphs are an important step in the development of typography.
I agree. However, carefully implemented alternates can be a HUGE development.
> produce a font that automatiacally changes glyphs so that i can create a different rythm
Yes and no. You can certainly produce fonts that vary the patterns, but "rhythm" simply does not exist at the letterform (or even bouma) level.
hhp
> Probably not "huge"
Have you considered the power of contextual alternates to aid readability?
Rhythm: some of the disagreement stems from a simple terminological difference, but some stems from a differing understanding of readability. Anyway, show me how Flow (using a decent definition, please) exists in text type design,and I'll change my mind.
hhp
if i understand correctly FF Kosmik produces a "random effect" by cycling through a series of three related fonts with alternate glyphs. Kosmik uses postscript programming. I played around one weekend and tried to create the same effect using open type programming, but i'm no programming whiz, so i haven't figgered it out yet. I was opperating on the following premise (which may be faulty as well since logic is not my forte either) have three sets of glyphs for each character: class A, class B & class C. Define the OT substitution rules so that AA becomes AB BB becomes BC CC becomes CA, ad nauseum (infinitum, whatev!) Supposing i could get the programming right would this work??? I've been wondering about this for a while. Here's another string similar to this one in case anyone's interested.
>I don't think that random glyphs are an important step in the development of typography
Oh no, randomness offers amazing opportunities!
I can think of two avenues worth exploring:
Firstly, it is necessary to introduce an element of randomness into a font that truly emulates handwriting -- not contolled, professional script, but real writing which has the rough creative imperfection of humanity, where whole-word "ligatures" are created on the fly. Moving beyond, this process, once formalized, could be used to design fonts which are not mimics of handwriting, but something else again.
And secondly: If one wanted to mimic the inconsistent quality of letterpress type, which is admirable when subtle, giving it life, rather than the dull predictability of high-res digital type, then one could arrange a "type tray" of glyphs, where each character has variants that are "worn" differently, or "print" differently (simulating arterfacts in paper or printing), and selected at random.
And is there not an element of randomness (effected by a lack of total control) in the design variants that occur between different type sizes in foundry type?
kris, thats exactly what i am interested in. it seems so strange, that so few typedesigners are interested in this randomness (most of you are thinking about major differences in the various glyphs), but i am actually really interested in minor differences, that make the font live more - a subtle randomness. and thats what i mean with rythm as well, or flow, hrant, type coming out of the computer seems to be overperfect and sometimes even disturbing to me.
constantin
Constantin, I would explore randomness if the tools I use made it possible, or if I was a programmer. But like most type designers, that's not the case.
The OpenType format, with its ability to contain mutliple glyphs for a single character, can be used to give the appearance of the randomness you seek, by assigning different glyph variants of the same character according to the adjacent character (i.e. with "invisible ligatures"). However, it is only since last year that the Mac version of Fontlab that does OpenType has been introduced, so give it a little more time.
Constantin, I for one am actually interested in the "over-perfect" problem you mention. But I worry about implementing randomness consciously. Like Johnny von Neumann (the father of computing) once said: "Anyone who considers arithmetical methods of producing random digits is, of course, in a state of sin."
hhp
What Hrant meant is (I guess): consciously implemented randomness is not true randomness. In practice, for type stuff, I don't think it makes much difference, but it's an important distinction anyway.
Not directly related, but here is something interesting about randomness.
thanks a lot for the link to randomness. it was very helpful. having read that article i must realise that i was not using the right word. looking at the definition of randomness i must actually use the word variation. i sometimes dont find the right terminology because english is not my first language. nonetheless i am very happy to have gotten such great inputs from you guys and would appreciate even more.
Captured from web.archive.org on 23 Jan 2015. view this capture