Designing punctuation - what size?

mr's picture

Are there any guidelines about designing punctuation? Especially their size.

It seems clear that it needs to be (relatively) bigger in text than display, but by how much? Or is there no way besides experiment?

Ricardo Cordoba's picture

Well, for starters I suggest looking at existing typefaces, both text and display, to see what other designers have done. That'll give you something to start on... Then you can experiment based on that.

cuttlefish's picture

This is a very good question I often struggle with myself.

if your punctuation is circular, seeing as circles read smaller than squares of the same width, I'd say your eriod should be 20-30% wider than the thickness of your thickest stem, up to a a certain point. Derive the rest from there.

Naturally I defer to the opinion of those with more knowledge and experience than I, should they choose to speak up about it.

guifa's picture

Apostrophes generally tend to fill the vertical space of an ascender, so, once you have the period, if you raise it up, and then add the curl down to the x-heigh, it will look about right. To make a comma, take the apostrophe you made, and then drop it down, and its curve will tend to be about right, unless you have a super-low x-height with a miniscular descender. For that and the periods, here's a sample I put together. Just look at other fonts in a similar way for the other marks.

http://www.elahorcado.net/random/punctuation_comparison.pdf

Edit: ack, apparently OmniGroup does use a separate engine for generating PDFsfrom their Export command than the standard PDF creation in OS X. Anyways, some of those are a little off alignment from where I put them but you get the general idea.

«El futuro es una línea tan fina que apenas nos damos cuenta de pintarla nosotros mismos». (La Luz Oscura, por Javier Guerrero)

dberlow's picture

"Are there any guidelines about designing punctuation?"

Yes there are.

I learned and continue to suggest one start with the i's dot and make a period that is apparently the same size as the i's dot in the middle of the point size range for which the font is intended to be used. (One can also start with the j's dot if one doesn't trust ones own i's. :))

The period's size may be made larger for smaller uses, and smaller for larger uses. The rest is based on that period with comma being made to look the same size despite its tail, the colon slightly smaller in its twinning and etc. But the period usually remains the largest.

If there is no lowercase, then base the period on the stem of the H and if the l.c. dots are somehow stylistically deviant, make one that isn't deviant just for this purpose and toss it when done.

Cheers!

Nick Shinn's picture

is there no way besides experiment?

I generally try to avoid following guidelines, and "work out" what looks best for a particular face by trying variants and see how they perform in different settings, before looking around to see how other designers have handled things.

As David demonstrates, guidelines can provide procedures that are useful for experiments, and using the i-dot as a benchmark is a good idea, because it connects a number of thematic elements; in size, it relates to the period, and in position to quote marks and accents.

Again, as he explains, the size of the various dots in a type varies. Consider, for instance, the dots in the question mark and bang; don't assume that these dots should be uniform with the dots in period, i, dieresis, etc.

Gary Long's picture

Punctuation marks being "road signs" aiding interpretation of the text, they need to be legible and distinct without being distracting. In some typefaces the tail of the comma is so thin and/or short that at small point sizes, it's hard to tell it from a period. A version of Times that somebody used to print out a manuscript for me a while back had periods so small that I was always confusing them with specks in the paper stock. The two elements of the double quote marks are sometimes placed too close, again causing legibility problems at small sizes.

Experimenting is the only way, paying particular attention to the smaller end of the point-size range a particular design would normally be used at. The human eyeball is often the best measurement tool.

Syndicate content Syndicate content