TypeTool might suit you.

Yes, Fontlab is poverful (yes, the v is a bug in Fontlab) and can be scripted with http://www.robofab.org/ in many wonderful ways. But consider http://fontforge.sourceforge.net/ instead. There you can play around with Euler curves (Spiro) and other fun things.
You could also try to give Glyphs http://schriftgestaltung.de/glyphs/about.html a spin, as it's free (for now).

Thaks to you both. No offense, but I'd like to leave the question open on the theory that one never knows what might turn up.

By the way, is Glyphs intended for Mac use only? I don't see anything that looks like a conventional Windows installer when the file is unzipped. That leaves me rather puzzled as to how to install it, open it, and use it.

Yes, Glyphs is OS X only.

Since you are using Windows you might want to try this range of font editors:
www.cr8software.net/type.html

There are three 'levels': Type light (free - basic editor), Type 2.2 (more advanced, with transformations, kerning etc) and Type 3.0 (which includes OpenType features and action scripts).

I wrote these myself, and like yourself I am from a scientific/technical rather than design background.

not interested in the demo version since I want to experience full functionality

AFAIK, the only limitation in "functionality" in the demo version is in saving and generating fonts...

@ clauses: thanks. I wondered about that, since there was nothing familiar-looking in the zip file. That kind of brands me as a PC type, I suppose.

Windows is favored by some font producers because Fontlab is less buggy in it's Windows incarnation. At least that is what they say.

I'm an engineer, and I see typography as an opportunity to apply math in a nearly-pure form

Thought this might turn more heads…

Thought this might turn more heads…

Not nearly as much as trawling for a free or low-cost non-demo version of FontLab...

>>an opportunity to apply math in a nearly-pure form
>Thought this might turn more heads…

A benign ignorance, soon corrected by experience. Almost every glyph must be modified from simple geometry to look good, or even to look geometrical. There may be formulas underlying what looks "right", but we are very, very far from finding them.

I was myself afflicted William. I probably still am!

Andrew (1985), have you drawn type?

Circles can't be circular to look circular, nor squares square, nor rounds the same dimensions as squares. Dividing in half vertically visually is not equal in measure, horizontals and vertical strokes that are equal in width don't look so, and on and on. And this is the simplest stuff.

One of the most astonishing things to me I have learned in drawing type is that what we see is not what is on the page. Our brain's interpretation of data is a *huge* contributor to what we see, and that is not well understood yet.

>>an opportunity to apply math in a nearly-pure form
> A benign ignorance, soon corrected by experience.

William, maybe the original poster is thinking of a different way of applying maths than you are?

Maths and fonts can have many relations:

It can be prescriptive like in the constructions of the renaissance – this does not work, as we know, and I guess this is what you were thinking of. This is what mathematicians call a sufficient condition for what looks right.

It can be used to describe a necessary condition, like in the RMX Harmonizer, and I personally belive this is relevant and useful.

It can be used to process shapes, like the other RMX tools or MM or Superpolator do, and again, I think it does so successfully.

Finally, maths can be used as a medium in which to describe or store shapes (i.e. Bézier curves), and since we are working on the computer, this kind of maths application is not only sensible but inevitable.

It would no doubt be very interesting for you John (1940LaSalle) to read Donald Knuth's epos on TEX and METAFONT. He came at it from the very same angle you did. Reading these books will inform you and save you a great deal of time. The collected tomes come i a handy box-set: 'Computers & Typesetting, Vol. A-E, Millennium Boxed Set". Another great book by Knuth is his 'Digital Typography' a sort of digest version of the box show here:

Should you however prefer to hear him tell the story, you can do so here: http://webofstories.com/play/17109

Tim, I completely agree, and am grateful for the work you and others have done with the mathematical side of font creation. Also as more is understood of the math, more can be incorporated, as with your RMX harmonizer—though I haven't used it yet.

I took the original poster to be saying that type design itself is "pure" mathematics. And as of yet, not that much of mathematics of the visual-processing side of reading is understood, so the eye must be the arbiter.

Reading Donald Knuth books reclining on a Mies daybed much? :-)

I still haven't opened the box. I read on my iPad :-P

Thanks to one and all for your comments. The truth of the matter is, I don't know how I'll wind up using math and font design together! I suspect a lot of it will be applications of analytic geometry, but that remains to be seen.

@ berkson: I suspect strongly that you're on target. That is, (let's say) analytic geometry will get the basic forms delineated, but there will likely be the aesthetic aspect of "that doesn't quite look right somehow" that will yield modifications. So it'll be a blend (not sure what proportions) of math, engineering and art. Sue me...:-)

And all of that, sports fans, is why I'd like to see if I could find a discounted or older used version of FontLab to give it a whirl and check out some ill-formed, hazy theories.