Perhaps I exaggerated in the description...

I've done one book--a dissertation in anthropology--using Chaparral, and find it a very useful, readable face (though the r is a little odd). Some might find it too colorful for a natural-scientific text, I guess. The size variants ("opticals"), which I didn't have, would make it even more useful for academic texts, with notes etc.

I really have no well-founded opinion on Avenir. Does it have text numerals?

Rotis is not one of my favorites, though the serif version is bearable.

I really didn't get this colorful impression on Chaparral... Do you have any other in mind?
About Avenir, my version, at least, doens't have text numerals. Yeah, that's a problem, I don't think I have any geometric with text numerals...

Actually, the first font that comes to my mind when thinking of a scientific text is Trump Mediaeval (Trump was German!), but this may be just because at university, my textbook in neuroscience was set in that face.

I think you need a pseudo-random font. (Given that true randomization is impossible.)
For instance, take TheSans and TheSerif, and make a new typeface from them, which uses OpenType contextual alternate coding to randomize which of the two "typefaces" any particular character is set in.
(Dead cats have no serifs.)
Thomas Phinney knows how to do this kind of thing.

Oh, man... I think that goes beyond my current typabilities... But thank you for the tip, I'll keep an eye on Thomas Phinney's blog (I didn't know him)

Nick, that would be a perfect thing for the Uncertainty Principle!

"The more precisely the position is determined, the less precisely the momentum is known in this instant, and vice versa." Heisenberg

ChrisL

Chris, I don't know if you're joking, but that argument really appealed on me. Problem is I don't have enough time to learn the how to do that and execute it properly. I'm in a work experience at ZKM, a Media Center of Art and Technology in Germany. I'll be here for 2 more weeks only and I'm regarding this book as a good opportunity, so I just want to do it as correct as I can. Oh, I'll send the cover! (why didn't I do that before????)
Marco

Marco,
I was only 10% joking and the rest was serious. It just strck me as a perfect fit--even if it would be a quantum leap for you to be able to do it in time!

ChrisL

All right, I'll give it a shot. Who needs to sleep, anyway :P
But what about the inside? Chaparral is looking very good to me, and I'm thinking of meta to the side notes and legends.
Thanks you

Nice cover! Superimposed cats!

I think you need a pseudo-random font. (Given that true randomization is impossible.)

If you're buying what these guys are selling, perhaps true randomization is not just possible, but inevitable? You could always hook up part of the back end of the typesetter (or rasterizer) to http://random.org/

Perhaps you could have the viewer interact with the cover? (Two possibilties that jumped to my mind: 1. lenticular printing; 2. a binding / wrap that requires the purchaser to tear the book open in a way such that some cat lives or dies.) Perhaps have a jacket that interacts with the cover itself? (Through transparency or cut-outs; a pair of moire images could allow small motions to induce all sorts of effects.)

Another form of interaction would be back cover/front cover, like the t-shirt that says "the cat is dead" and "the cat is alive" on front and back.
I know interaction is supposed to supercede mere depiction of an idea, but if you're familiar with the concept in a classical way, interaction is a trite hassle. It's like having to mouseover a CS logo in the OS X dock and they all look the same even after what is it, three years?! when having it spelled out would work better. So the illustration of superimposed cats works very nicely.

When I say that true randomization is impossible, I mean that the OpenType feature is not supported by any typesetting applications.

Reading is a quantum phenomenon. You decide what text means by reading it. That's why Helvetica is so stupid -- all that entaglement with billions of other texts clouds the meaning of any particular text it's used for.

Dear Nick,

I think you are having a Milli Vanilli moment. This idea has already been done by Tal Lemming in Christian Schwartz's LOCAL GOTHIC. You have seen this before. I hope your failure to provide attribution is simply an oversight.

Lemming told me over lunch that most of the information on his website is simplistic smoke and mirrors but the basic ideas are deeply rooted in a quantum physics. You see, the problem with producing a "random" effect in OpenType is that the GSUB table is static. There is no way to, as Just van Rossum and Erik van Blokland did in BEOWOLF, link to a pseudo-random number generation library. This prevents the generation of a random seed which in turn causes the utter failure to launch a randomization routine. Most of the attempts at randomization, and I believe that Phiney's attempts fall into this category, use simple glyph rotation schemes or very simple backtracking to trigger a single substitution. What Lemming stumbled onto was a way to simulate a random seed generator. From there it was a fairly simple set of steps for him to implement a basic superposition algorithm. I shouldn't give away any more secrets as he explained all of this in confidence.

Will you be in Lisbon? We should have a beer together.

BTW, do you know Professor Andreas Von Shinn of the Hochschüle of Fachhochschule Darmstadt?

SIncerely,
Randolph T. Burke

I hope your failure to provide attribution is simply an oversight.

I don't believe I claimed to have invented the idea of random variation from a set of alternates. (Making it work in a digital font is another matter.)
The first time I came across it was Kozmik the Flipper font, over ten years ago.
I have no idea how truly random Tal's "random seed generator" is, but it sounds way kewl.
John Butler has done something for the OT version of LettError's Kozmik.
I didn't mention either (and there are probably more) -- I didn't think it was an issue.
In fact, I published a typeface, Handsome Pro, in 2004 that has many contextual alternates, triggered by a variety of cues, that gives a good random effect -- there are quite a few script fonts that do this.
The only ostensibly random OT technique I have looked at is Mr Phinney's.
Sorry, not going to Lisbon. But a beer is always a good idea.

sorry, double post

Alright, I love quantum physics too, but I need some help with fonts! So... Meta ins't working as caption font... Suggestions? Something that goes with chaparral?
Thanks

How about Process Type Foundry's Bryant?
As a rounded face, it looks kind of scientific/technical, but with classic lettershapes and small x-height it matches the tone of Chaparral. Also, I think that a rounded face has a good contrast with a slab -- it's difficult to choose a contrast face for a slab, as the monoline quality is close to sans.

Which program are you using for typesetting the book? Does it include lots of math?

Nick, I finally stayed with myriad and chaparral. Bryant is very nice, though.
Kon, I'm using InDesign CS2. No, not that much math, or I'd have to learn TeX, or something like that, right? :)
Well, people, thank you very much for your help, I'm finishing the book this weekend. This forum is a beautiful thing.