Why the link on the first zero in %?

Ray Larabie

What's up with that?

Hrant Հրանդ Փափազեան Papazian

?
What link?

Christopher Slye

I can’t really answer your question, but it’s always seemed like a natural connection to me. (Way better than, say, a c_t ligature.) I’ve always assume it’s an outgrowth of some handwritten source.

Hrant Հրանդ Փափազեան Papazian

Oh, he means in some fonts. Yeah must be some handwriting fetishization. Best avoided. Plus if there's a "c/o" glyph I like to ligate the "c" and "/" to make it less like a "%".

notdef

There are no zeros. The percent comes from a short form of “per cento” (per hundred), “pcº”, following the scribal tradition of abbreviating words to its first and last letter, with the latter superscripted. The ‘p’ was eventually lost.

Ray Larabie

I've noticed them on typewriters and some newer typefaces. How far back to they go?

Vasil Stanev

It is how the hand writes it out, non-mandatory, and it is a design choise by the type designer wether to include it or not.

The Shady characters blog is useful for questions related to such glyphs:
https://shadycharacters.co.uk/2015/03/percent-sign/

Chris Lozos

Way far back, Ray

Hrant Հրանդ Փափազեան Papazian

Frode Helland said:

There are no zeros. The percent comes from ....

Today, and for the foreseeable future, it's zeros. Hence the ‰ sign.

And the pilcrow used to be a "C", but –thankfully– is now its own thing.

Nick Shinn

I get bored, mix things up. Sometimes my percent roundels are little zeroes, sometimes they are circular “degree”s, sometimes I make the ligature Ray refers to.

Times has the most distinctive percent sign, IMO.
But perhaps there is something even more demonstrative?

Hrant Հրանդ Փափազեան Papazian

Hold my kebab...  (Sorry.)

Mark Simonson

The percent symbol with the link at the top seems to come from roundhand script. Note how this form has a heavier stroke on the left than the right.

K Pease

The most naïve designs of the percent sign are the ones conceived as a fraction with zeroes. Backformations such as the per mille are an interesting curiosity but not a great excuse. There are situations where actual fractions may be also present and you would want them well disambiguated. The link stroke can help affirm its identity as its own symbol.

Hrant Հրանդ Փափազեան Papazian

Since nobody sets a fraction with a zero as the denominator, what's not a great excuse is chirography... That said (as I implied above) a symbol being itself is a great thing, so we should never stop looking for a better way... but we can surely find an avenue less arbitrary than historicism.

Nick Curtis

I suspect that it's simply a holdover from cursive handwriting.

Helmut Wollmersdorfer

In some cursive handwritings the first character looks like \c connected to \t (reduced to /) and the final \o is also connected.

In old specimen catalogues they offered two variants: one with steep slant like the slant of \t in an italic font, the other more like a fraction.

John Savard

Of course, in addition to the per mill symbol, another symbol which at least appears to be related to the percent sign is the "care of" symbol.

Oh, yes, there's also the a/c symbol, which stands for "account":

And here are some more examples:

Matthijs Herzberg

K Pease said:

The most naïve designs of the percent sign are the ones conceived as a fraction with zeroes. Backformations such as the per mille are an interesting curiosity but not a great excuse. There are situations where actual fractions may be also present and you would want them well disambiguated. The link stroke can help affirm its identity as its own symbol.

Why is it that the % should be clearly differentiated from fractions?

Helmut Wollmersdorfer

John Savard said:

Oh, yes, there's also the a/c symbol, which stands for "account":

Thanks, so maybe the sign Ko in German specimen books means "Konto" (German for account).

From 1915, Bauer, Hauptprobe



Unfortunately I forgot what m/n means.

Paul Miller

I must admit on some of my fonts I have made the percent sign as a fraction, zero over zero.  But it looked OK in my opinion.  Should the percent sign be differentiated from a fraction ?

K Pease

Well, it's not a big deal, and of course it only takes a moment of examination and logic to see the difference, but that moment represents the "legibility/readability" distinction. Anything to make recognition quicker. And the ease of recognition of a symbol, especially collectively across all the forms it can take and to the most people, is, sorry Hrant, bound to its history. History is arbitrary but every alternative is more arbitrary.

Both fractions and percentages are used with numbers, and they may rarely be used together like "2½%" which is mathematically valid if not typical usage. I have seen rare cases in which the percent is constructed as a nut fraction. A passable solution for some extreme constraints, but I think you can see why that, at least, can cause trouble and should be recommended against in general.

Hrant Հրանդ Փափազեան Papazian

K Pease said:

Anything to make recognition quicker.

Well, to me not absolutely anything (even though I'm actually a huge fan of differentiation). The potential confusion is exceedingly rare here, and things like "c/o" are virtually anathema to immersive reading anyway. So I happily sacrifice such hypothetical confusion events for the sake of shedding the historical baggage of calligraphic type, which holds us back more than anything else in our field.

John Savard

Hrant H. Papazian said:

the historical baggage of calligraphic type, which holds us back more than anything else in our field.

The field has not been held back so much as to prevent Bifur, Calypso, and Optima from coming into existence.

But I can definitely see your point.

Openly calligraphic types, like Freehand, are one thing; they exist as a potential alternative that is comfortable and traditional.

But typical standard serif Roman types - Bembo, Times Roman, Century Expanded, and so on... while comfortable and readable, have in them vestigial characteristics that come from calligraphy. And it's fair to ask why.

Couldn't the characteristics that make Times Roman apparently more readable than Helvetica for books be separated out from the homages to Roman inscriptional carving and to calligraphy, so as to allow a wider variety of readable typefaces, and perhaps greater optimization?

Despite the current fashionability of slab serifs in display type, Egyptians are, thanks to having serifs, more legible in body copy than sans-serif types, but they have hardly ever been used to typeset whole books - even though sans-serif books have been set.

Since readability and legibility, though, are partly derived from familiarity, it should come as no surprise that many type designers see no percentage in being bold and innovative to that extent...

Hrant Հրանդ Փափազեան Papazian

Like with an abusive partner, you don't always need to have a plan for the future to simply know you need to shed something. But luckily we have plenty to plan, around what we read, and have always read: notan.

The role of serifs is interesting, if generally poorly rationalized (no, they're not "train-tracks for your eyes"). I believe what they do is make letters less themselves (in fact less legible) and more parts of wholes in the low-acuity parafovea, where most of immersive reading actually happens; and possibly even in the fovea, even though there's enough acuity, because recognizing a number of things as one thing is faster than recognizing those things individually and compiling them.

Matthijs Herzberg

@John Savard Said:

Couldn't the characteristics that make Times Roman apparently more readable than Helvetica for books be separated out from the homages to Roman inscriptional carving and to calligraphy, so as to allow a wider variety of readable

I would say that Times Roman is a face that already separated the senseless and sensible aspects of calligraphy, unlike humanist and old-style faces.

It retains what is useful to even texture and scaling to small sizes (stress in letters like /e and /c, transitional-style stem joints, etc), but rationalizes many 'silly' features of a Jenson-style face: an /e with diagonal crossbar and a beak, to name an obvious one.
On the other hand, it hasn't broken out of its mold as far as a modern serif, which often lack the even texture of broad-nib based faces.

And of course, familiarity is such a large part of it, which is why all of these discussions tend to be opinion-based... but it can still be interesting to speculate.

@"Hrant H. Papazian" said:

I believe what they do is make letters less themselves (in fact less legible) and more parts of wholes in the low-acuity parafovea,

Which is why this is the future of text faces

Hrant Հրանդ Փափազեան Papazian

@Matthijs Herzberg  :->  Cartouches... Maybe the Ancient Egyptians had it right all along!

Paul Miller

Hrant H. Papazian said:

Like with an abusive partner, you don't always need to have a plan for the future to simply know you need to shed something. But luckily we have plenty to plan, around what we read, and have always read: notan.

The role of serifs is interesting, if generally poorly rationalized (no, they're not "train-tracks for your eyes"). I believe what they do is make letters less themselves (in fact less legible) and more parts of wholes in the low-acuity parafovea, where most of immersive reading actually happens; and possibly even in the fovea, even though there's enough acuity, because recognizing a number of things as one thing is faster than recognizing those things individually and compiling them.

When we read we think we are recognising the shapes of the letters but it is much more about recognising the shapes of whole words and pattern matching to what we expect to see.  So serifs aid the recognition of words more than they aid the recognition of letters.

John Savard

Matthijs Herzberg said:

I would say that Times Roman is a face that already separated the senseless and sensible aspects of calligraphy, unlike humanist and old-style faces.

Personally, I am inclined to agree with you, as, while I understand Hrant's view of the influence of calligraphy on typography, I do not share it.

So the idea is that while this may be true, it could also be true that while Times Roman shed the most obvious unnecessary inheritances from calligraphy in the Roman type, there could be other, subtler ones that should also be removed as well without impairing readability.