John Savard

Hrant H. Papazian said:

The best "flavor" suitable for an Italic is simply to mark emphasis, not to appear more informal, organic, fluid, etc.

That is a valid point.

But it's also true that because people are used to the kind of italics we have with Roman typefaces, a sloped one instead will convey a message that isn't wanted; it will remind them of certain badly drawn typefaces that happen to have such italics.

But I think we're getting into a completely different debate here: whether the function of the type designer is to serve Art and educate the public, or whether it is to do what his customers want and put food on the table. I would like to submit that there's only one answer to that question, and that's to do at least a little of both; if a balance is not found, either one fails to serve Art because what one does is forgotten in obscurity, or one turns a creative endeavor into drudgery and even prostitution.

It's also interesting to note that a technique analogous to ink traps, called Optical Proximity Correction, has long been used in integrated circuit fabrication to compensate for the fact that features on chips were no longer many times larger than the wavelength of the light used to print them. (Now, they're smaller, and so additional techniques, such as multiple patterning, are also required.)

Basically, in the Latin alphabet, having both uppercase and lowercase was a recent invention - it came about in the Carolingian script, which combined Roman capitals with uncial writing for the lower case. So the lowercase letters had developed independently of the uppercase letters over a long period of time.

Thank you for this interesting historical information.

I searched for historical information on Van Krimpen's objections, and I see that it was his preference to design directly for the unit system, even though the usual practice at Monotype was for designers to make their drawings without reference to it, with the letterforms then being adjusted by other hands.

This suggests that Van Krimpen's more serious objection was to the reworking of his designs in a significant respect by others, but I have not yet read his comments on this subject.

As to Monotype's practice, this is quite understandable given historical precedent. At ATF, the widths of characters also fit into a grid, but quite a different type of grid than used by Monotype. There, the width of each character was a multiple of 1/4 point (this is actually an oversimplification; in larger type sizes, quantization to a unit of 1/2 point or 1 point would be used if it seemed reasonable, and in the other direction, 1/8 point may have been possible).

This meant that for a given size of type, the widths of the characters had more flexibility than in Monotype's system.

At Monotype and ATF, when a typeface was produced in different sizes, the smaller sizes were optically adjusted; they were slightly wider in proportion to height, and also usually slightly bolder. In the case of Monotype, the width adjustment had to go in steps of 1/4 point of "set width".

But in the case of ATF, this meant that the relative widths of the characters had to be altered, admittedly only slightly, for every type size.

Given this precedent, Monotype may have felt that the aspect of design workflow noted above, where someone other than the typeface designer would adjust the face to fit within the constraints of the unit system, would be workable and acceptable.

Ofir Shavit said:

I have to disagree Ori, the optical issue should not be ignored in Hebrew and in your typeface. If you like to create a contrast in your typeface, you create it, and balance the letter-forms accordingly.

You are absolutely right that the issue of the apparent stroke widths of horizontal and vertical strokes being altered as an optical illusion... can't be ignored in Hebrew any more than it can in Latin.

However, that in no way invalidates Ori Ben-Dor's typeface. It may be that the desired effect is created by using equal and horizontal stroke widths without any compensation - since it is likely to be acceptable when the apparent stroke widths are slightly in the direction of the traditional ones for the script, instead of being visually uniform, but it is usually not acceptable when the apparent stroke widths vary in the opposing direction, as would be the case for Latin.

That isn't at all the same as saying that optical compensation doesn't matter for Hebrew, just that for a particular choice of intended visual effect, taking explicit action may not be required. The apparent contrast that an optical illusion creates... is also a possible design choice.