Good question. Same problem with apostrophes, left and right side.

I kerned slashes and apostrophes, and I think that Font Bureau’s fonts generally have these kerned.

I think the answer to that is at least partly historical.

If you go through an old type foundry specimen book, you will find that a standard font consisted of upper and lower case letters, with the standard punctuation marks and the ampersand. Even parentheses, and the dollar sign and the symbol for pounds sterling were considered “typographic accessories”. Later on, with Monotype and Linotype, a somewhat greater variety of characters would be made specific to each typeface, including (, ), $, and /.

For a given typeface, there would often be available matching small capitals, accented letters, and fractions.

But characters like +, =, *, @, # and % were sorts. There would be one or two generic versions sold by point size, not typeface.

James — Which nobodies are you looking at?

For example: MillerText-Roman has 42 kerning pairs involving slash or backslash; Whitman-RomanLF has 74 such pairs; ZócaloText-Regular has 121.

James — Which nobodies are you looking at?

Myriad Pro and Helvetica Neue don’t appear to have slash kerns (I’m just check in Indesign, not opening stuff in Fontlab). Whitney has some for slash but not backslash, same for Calibri. Maybe part of the problem is I am not checking good text faces.

Just checked a plain sans: Univers 55 (Com OT) has 49 kern pairs both for slash and backslash.

All the fonts I create or rework get kernpairs for the slashes and quotes/apostrophes where needed - it isn’t THAT much extra work, and they might be appreciated by the users... ;)

Personally, I think kerning of the slashes, particularly the backslash, is a lower priority than than the quotes. Quotes, because of their position, greatly benefit from kerning, and and are used very often in text.

Always kern slash to itself and to “w” in this day and age.

http://w

ChrisL

Oh, dear, I wasn’t reading this thread carefully enough. This was about kerning slashes... not putting serifs on the ends of them.

>And these aren’t being drawn with sidebearings that can balance them out [...]

There are a lot of multi-case characters that can’t possibly be spaced once for all glyphs. So we do and have kerned slashes and quotes since kern 1, (1989). I have noticed that quite a few founders only concern themselves with kerning the cases and have always assumed that they simply expect the user to kern the odd combination that gets in the way.

We feel our fonts should just work seamlessly unless the user tracks a lot... or something horrid.

Cheers!

I kern slashes but not backslashes. When using my fonts to describe DOS commands, I’m afraid you’ll have to hand kern.

When using my fonts to describe DOS commands, I’m afraid you’ll have to hand kern.

Good point. I forgot that backslashes have become irrelevant.

backslashes are still necessary for rocking out in ascii

\m/, ,\m/

And after kerning slash, what comes next? Asciicircum? Mu? Integral?

>And after kerning slash, what comes next?

I kern the back slash next. . . I never kern asciicircum, mu or integral. Is that a mistake?

Cheers!

I have on occasion kerned asciitilde so that it becomes an “add on” swash...

William Berkson sez:

Personally, I think kerning of the slashes, particularly the backslash, is a lower priority than than the quotes. Quotes, because of their position, greatly benefit from kerning, and and are used very often in text.

I’m always surprised at the number of otherwise well-kerned fonts that don’t include the combinations of single-quote/double-quote. In English, at least, this combination occurs quite frequently, whenever there’s a quotation within a quotation (whichever way around you prefer to use them — singles first [UK] or doubles first [US]). I’ve gotten in the habit of globally inserting a thin space between them, so they don’t run together into what looks like three single quotes together; but whether that amount of space will work really depends on the typeface.

“Amen”

ChrisL