Well, Gill himself once made an Arabic (during a sojourn in Egypt) but: it wasn't meant to match Gill Sans; and it was somewhat freaky so was never published. A while back* there was an article about it (embarrassingly called "Gill Sands") in Print magazine (in which the dunces actually managed to put one of the illustrations upside-down).

* I just checked - it was exactly 14 years ago.

hhp

I remember that Eric Gill made a rather freaky Hebrew that was published... but this post reminded me of a Peanuts cartoon with the punchline "How can a hula hoop be hi-fi?" - in this case, the question is, how can an Arabic be Gill Sans?

Of course, for most practical purposes, putting in a generic Arabic typeface as part of a Times Roman or Arial typeface works to allow the characters to be displayed - and, thus, there is no real need to truly capture the spirit of Gill Sans in the Arabic portion of the font. Although that might still be a thing for a type designer to attempt - but such a typeface would presumably be sold separately and cost extra, as it probably should.

His Hebrew was a few notches freakier than the Arabic. It's on the order of the cra... stuff in Schonfield's book.

I think it's possible to make any sensitive non-Latin of any font - it's just that most people wouldn't have the eye to see what's really going on. That's just the nature of the beast. In fact not unlike how most people think Georgia is a big Times.

hhp

Oh? And here Ah sat ova mah cuppa and thot Georgia wuz a kyoot Modern.

And because it had oldstyle figures (but they got the 1 wrong for the sake of protecting the typographically illiterate from confusion), I thought it was some kind of a hideously deformed Garamond... or an awkward grafting of oldstyle features on a Century Expanded-like base. But those aesthetic sins could be excused as it was a passable font designed to be readable on low-resolution displays.

Hello, I created Gill Sans Arabic Version for my Company

Here it is

Would you like a critique?

hhp

Um, that would be piqued "my curiosity".

To me the "secret" of translating a typeface's idiosyncrasies (better: characteristics) across scripts is to avoid the facile migration of formal shapes (which very rarely works, and often results in cultural damage) and focus instead on figuring out what those shapes evoke in the source culture and migrate those to shapes that evoke similar things in the target culture.

hhp

Deadlines and confused clients can certainly cause one to design something not to be proud of... Barring those things, as a rule it's possible to translate any "feeling" since humans anywhere all feel similar things.

BTW there's actually a hidden risk when designing something close to Latin, like Cyrillic or Greek: you feel more justified doing a straight formal migration. But something like Trajan does not evoke in its forms the same things to a Russian or a Greek; so for example its "A" should not be identical between those scripts. The idea of "cognate shapes" (as put forth by John Hudson for examples) is a red herring. In contrast when you're doing a totally different script (like Arabic) the disjoin is so obvious that it forces you to think more deeply.

hhp

But something like Trajan does not evoke in its forms the same things to a Russian or a Greek.

It does, at least to a Russian.

<…> its “A” should not be identical between those scripts.

It should, at least between the Latin and the Cyrillic glyph sets.

Maxim: that was pretty fast. :-)
To me, since Latin users and Cyrillic users are necessarily different in terms of their "visual language background", any formal identicalness is a lost design opportunity.

I find it hard to imagine that an Arabic script Gill Sans brings up the same feelings and associations as the Latin Gill Sans does.

Well, if you do apply formal congruence between the two scripts, I certainly agree they bring up different feelings – and that's actually my point. What I'm advocating is making new shapes in the Arabic that evoke to Arabic readers what Gill Sans tends to evoke to Latin readers.

And yet it is.

Almost always, yes. But it doesn't have to be: look at the "Y" shapes in Adobe's Sava.

What should be different?

I don't have nativity in Cyrillic so it's hard for me to know offhand, but in the case of the "A" for example I would say the two main factors are width, and height of the bar. In the latter especially it should be possible to establish whether a higher bar or a lower bar evokes to Cyrillic users what Trajan evokes to Latin users.

hhp

Since Times Roman came in special versions for German and French, that capital "A" maybe should look different in Russian or Greek should not be too hard to accept.

While only native readers of those languages could really decide what is appropriate, a guess, a glimmering, might be possible from looking at what is popular in those languages. In the case of Russian, it seems that typefaces in the Modern classification (akin to Bodoni and Scotch Roman) were popular to nearly the exclusion of Transitional or Oldstyle faces, whereas in the Latin alphabet world, Transitional (like Baskerville or Times Roman) and Oldstyle (Garamond, Bembo, Caslon) were much more popular.

In the case of Hebrew, I've heard that the Frank - Ruhl face was so popular as to be Israel's equivalent to Times Roman - to me, that face looks, visually, like a Scotch Roman, while the "Siddur" type of face is more like Times Roman or some other comfortable Transitional or Oldstyle.

For Greek, sans-serif faces seem to have a greater level of popularity for text use; this is not surprising to me, since it seems to me that they better avoid the tension between capitals that are like Roman capitals and lower-case that is like script or italics (even if it isn't sloped Porson-style).

Martin Silvertant > ... if I were to design a typeface for it, it would probably feature strange deviations one might argue make the typeface terrible, while others might say I've done something refreshing they haven't thought of yet.

@ It is the strange deviations that is refreshing to me. Go Go Go!
http://typophile.com/node/120811

Arabic has 28 letters (not counting extended letters use by other languages, such as Farsi) most of which have four forms (some only two) depending on the position in the word. Some people think that Arabic is in fact even more complex than that – that it relies on complex ligation to be authentic.

if I were to design a typeface for it, it would probably feature strange deviations one might argue make the typeface terrible, while others might say I've done something refreshing they haven't thought of yet.

Indeed the non-native has an interesting and useful role to play: naïve invention. Like how vernacular lettering can end up informing formal typography. That's why when I'm hired as a consultant for Armenian type design by non-natives I prefer to let them first experiment, based on their instincts. Interesting things come out! On the other hand, when it comes time for me to apply my native expertise they have to be open to my confirmation or rejection of certain innovative features. Too often non-natives don't accept their limitations, sometimes resulting in actual cultural damage to vulnerable writing systems.

hhp

Much dismissal of native advice stems from the –admittedly essential– artistic element of any design activity. The artist in a designer can fall prey to hubris.

BTW tell your dad that most type designers are not autistic but would have been just as obsessive! :-)

hhp

Well, one can say that Arabic has 28 letters, just as one can say that Hebrew has 22 letters. Yes, they're all consonants. But the writing system isn't a full-fledged abugida, like Indic scripts or Ethiopic, since vowels are only an optional annotation.

Too often non-natives don't accept their limitations, sometimes resulting in actual cultural damage to vulnerable writing systems.

Indeed, that has happened. But not all non-Western writing systems qualify as vulnerable. Occupying powers, and missionaries giving people their first writing system have the power to cause trouble.

And the appeal of the Western/U.S. media can lead to people doing it to themselves. But we also shouldn't underestimate the resilience of traditions.

Thus, if Koreans want to do a Fraktur-ish version of their script for a heavy metal record jacket, it's not likely to stunt the growth of their writing system. They can be allowed to have fun.

Actually Arabic's three vowels are not optional when in their "long" form.

But not all non-Western writing systems qualify as vulnerable.

Certainly true.
The ideal balance between innovation and tradition depends mainly on the degree of vulnerability. Although paradoxically, there can be no cultural survival based only on tradition.

hhp

Ah, so Alif, for example, is a long vowel - not a consonant (glottal stop) as I thought.

The "a", "ee" and "oo" sounds have long forms that are always marked and short forms that are optional (although sometimes crucial for understanding).

hhp

It almost sounds easy.

Really? To me it's the biggest challenge in type design. Well, barring the final emancipation of the white.

hhp

Hrant > Arabic has 28 letters (not counting extended letters used by other languages, such as Farsi)
Quadibloc > one can say that Arabic has 28 letters, just as one can say that Hebrew has 22 letters.

@ Arabic has not more than 19 letters covering all that used by Farsi, Jawi and other languages?!
http://typophile.com/files/QB3.3.1.pdf

Martin Silvertant > Can you illustrate what kind of strange deviations you're employing there?

Ask http://www.linotype.com/6732/arabic.html why they dismissed my fonts?!
http://typophile.com/files/YS-Aziz_1.pdf
Another example for what you have said?! > ... it's frustrating when you've done years of research on a subject and you get quickly dismissed just because the other person doesn't have a similar insight to verify the truth of what you say.

> Besides, I asked what kind of deviations you're employing; not ...

@ Well, Linotype, Fontlab and other type foundaries still do not recognize the fallacy of the Arabic mark-positioning notion that need to be constantly deviated as I said here:
http://typophile.com/node/82934

> Also, I'm curious if the Arabic script has the equivalent of the colon, semicolon, dashes etc. Does it have a need for it?

@ Yes, it does?!

Aziz, I think it would be useful to answer questions more directly.

hhp

Hrant > I think it would be useful to answer questions more directly.
Martin Silvertant > can you show the difference in pictures

@ At you service with flowers

Thank you.
What are the superscript numbers with parentheses?
And what's that upside-down comma in the Latin?

hhp

Arabic has not more than 19 letters covering all that used by Farsi, Jawi and other languages?!

Most references, at least in English, correctly or incorrectly, treat ba (ب), ta (ت), and tha (ث) as three different letters, jim (ج), ha (ح), and kha (خ) as three different letters, and so on. Even though the difference between them consists only of dots and the like, they're not thought of as diacritics.

That's why I had said 28 and not 19.

Historically the dots were actually vowels applied to a base. But that's material for a museum, not living people.

hhp

I just ran into the perfect example of what NOT to do:
https://www.behance.net/gallery/4790859/Khawaja-Typeface-Arabic-Myriad-Pro

hhp

Quadibloc > Most references, at least in English, correctly or incorrectly, treat ba (ب), ta (ت), and tha (ث) as three different letters, jim (ج), ha (ح), and kha (خ) as three different letters, and so on... That's why I had said 28 and not 19.

@ That depends on how you look at it, optimistically or pessimistically?!
Arabic has 28 leters or 19 letterforms.
Latin has 26 letters or 52 letterforms.

Hrant > What are the superscript numbers with parentheses?

@ Hindi Footnotes followed by full stop and comma kerned into one single glyph.

Hrant > And what's that upside-down comma in the Latin?

@ Arabic Footnote followed by comma kerned into one single glyph.

I just ran into the perfect example of what NOT to do:

It certainly seemed like that, based "100%" on the curves of Myriad Pro in Latin. But it was designed by one Mahmoud al-Khawaja, not by Joe Smith.

A native user of the writing system might know how much borrowing from external sources he can get away with, while one without that advantage needs to hew closely to the established tradition.

Sadly, and as John Hudson once made me come to terms with, more often than not Latinization is carried out by natives!* This is because many non-Westerners have an over-riding desire to become more Western, and they're too cavalier with their culture. Couple that to a shallow understanding of how visible language really works, and you get Frankerstein fonts like this one.

* Especially those who have spent some time in the West, and have developed an emotional attachment. Note that Alkhawaja studied at Savannah Collage of Art and Design, and claims to be part of US culture as well as that of his native Jordan.

I have nativity in Arabic myself, and harbor some healthy mistrust of Western influence on other cultures. In addition for many years now I've been looking deeply at formal relationships of writing systems. To me this is a clear case of inviting cultural colonialism.

The best letter to look at there is the و (waaw) which has a spur at its top for absolutely no reason other than to look more Western at all costs.

hhp

I don't think that native users of a script can do no wrong. But if someone who was not a native user did something like that, without even a precedent to go by, for a script as different from Latin as the Arabic script is, that would be absolutely idiotic - thus, "the perfect example of what not to do". Done by a native script user, it is merely highly questionable.

I would agree... although doesn't being a native make it less excusable?

hhp

It may make it less excusable, assuming there is something to be excused.

Certainly the most startling letter in it was the waaw, as it looks 'way too much like a (single-story) lower case "g" - but then, if one was not familiar with the Latin script, that wouldn't be obtrusive for that reason.

In the case of stuff like heavy metal Korean, what's happening is that users of a non-Western script wish to quickly appropriate the wealth of display typeface options that the Latin alphabet has built up over years. Advertisers with products to introduce and ad campaigns to begin... can't wait for the natural and organic evolution of a script.

The glitz and glamor of our wealthy Western lifestyles has beguiled the rest of the world in many respects, and to expect the rest of the world to live simple lives in poverty so they can remain quaint and picturesque for our sake is unreasonable. People want to survive, to thrive, to have fun. So they jump at things that advance their personal interests, with little thought of the long-term impact on values like cultural survival.

There's most definitely something to be excused. Like an Arab teaching his kids to speak Arabic like an American.

Who's not familiar with the Latin script? And the thing is explicitly a companion to Myriad. Plus in any case it's more noise than data.

The glitz and glamor of our wealthy Western lifestyles has beguiled the rest of the world in many respects

Something I've expressed often.

to expect the rest of the world to live simple lives in poverty so they can remain quaint and picturesque for our sake is unreasonable.

Nobody should expect them to live in poverty. That's exactly why we should fight cultural impoverishment. When we become too similar, by losing ourselves, that's a loss for everybody. Except corporations...

I'm no fan of traditional calligraphic Arabic type. But let's not throw out the baby with the bathwater.

hhp

You know, I'm not familiar with the Arabic script and so I actually don't see anything wrong with the Arabic Myriad Pro. It looks somewhat forced into the geometric construct of the Latin script, but then I think that's the charm; to have a minimalist, geometric Arabic script. What do you feel should have been handled differently here? Could someone perhaps draw up a sample with a letter as you think it should have been designed as opposed to what Myriad Pro in the sample is now?

Just look at Adobe's own Myriad Arabic (a design that's not slavishly traditional) which actually came out a year before Alkhawaja's work. I guess he wasn't satisfied by a culturally authentic rendering.

hhp