Metrics for superscript letters

Jure Kožuh

Hello

I am trying to figure out in which way to define superscript letter (a-z) metrics ...
For super/subscript numerals I know it is best to define them as tabular (http://fontfeed.com/archives/figuring-out-numerals/), but I don't know what the best definition would be for the superscript letters - tabular (or monospaces, which is almost a non-possibility as the letters would have to be drawn in a way that would allow this) or proportional?

Also, are there any uses for subscript letters (I haven't made them & I don't think I have to ... )?

[Deleted User]

The user and all related content has been deleted.

Jure Kožuh

So as I thought, the letter sups glyphs should have proportional metrics and the numerals should have tabular. Thank you for clearing that out!
I also see that you suggest I should also make subscript letters. Well that is no problem, I just create additional glyphs as components and move them vertically.

Thank you!

James Puckett

But I draw the line at getting too wrapped up in kerning them

This is a great use for the kerning copy and transform tools in Metrics Machine.

Nick Shinn

I make proportional superior and subscript figures, but tabular numerators and denominators. I don’t bother with kerning, except for 74.

Craig Eliason

What is a use-case where tabular superscript figures would be clearly preferable?

Nick Shinn

When using them at the bottom of a page to number footnotes?
Some typographers do that, but I always used full-sized figures, as the type size for footnotes is so small that superior figures end up being way too tiny.
Besides, it’s extremely unlikely that one gets into double figures, and typographers can always tab to the text if they don’t like the alignment that proportional figures cause.

Mark Simonson

Superscripted tables?

Craig Eliason

Do superscripted figures following a figure align right of a decimal tab?

Kent Lew

I’ve never understood the reasoning for tabular width on superscript figures. I don’t really agree with Nick. It is not at all unlikely that footnote calls will run to double figures — depending upon the kind of publishing we’re talking about, I suppose.

I believe most publishing styles now prescribe full-size figures for heading the notes themselves. Superscript is only used for the in-line reference.

And with tabular subscripts, then in a formula like C₆H₁₂O₆ (carbohydrate) you run the risk that the H₁₂ looks more like H₁ ₂O.

One will argue that these are specialized cases found mostly in academic publishing. But I ask: what is the purpose being served by tabular width for super/subscripts in general use that is worth sacrificing performance in complex typesetting?

Kent Lew

Do superscripted figures following a figure align right of a decimal tab?

Craig — If I understand your question correctly: 1) Typically footnote calls in a table will be hung outside of a right-align tab alignment; and 2) typically in a table of numerical data, any specific notes relating to the data will use an alphabetic or symbolic reference, to avoid any remotely possible confusion with an exponent.

If you have a copy of The Chicago Manual of Style, 15th ed., you can see examples of the former in Tables 13.1 & 13.2, and you’ll find the guideline for the latter in §13.47.

Jure Kožuh

So we need a consensus ... Tabular or proportional? We have reasons for proportional, we still need reasons for tabular.

[Deleted User]

The user and all related content has been deleted.

Nick Shinn

Kent, you’re not really disagreeing with me, I was just suggesting the best possible answer to Craig’s query that I could think of. You’re right about the double figures, I was thinking about footnotes on a per page basis.

Jure, looks like proportional-only is quite alright—although bucking this convention will create extra work for James :-)

Deleted Account

So we need a consensus ...

When users have one.

Thomas Phinney

Besides other issues noted, I think that the thickest rounds in bdpq are a bit light. They should be a little heavier than the corresponding straight vertical strokes.