Two New Trends in Arabic Typography

[behnam]

Saad wrote
>>one should just look at what the Arabs did to their Language in the Basrah Gulf area!

Oh yes! Basrah Gulf area. That was fourteen hundred years ago and my memory is a little bit sketchy!
Nonetheless, the Chinese still speak Chinese not Arabic don't they?!

I do understand your fear. Actually Reza Shah was very tempted to implement Ataturk idea. With his authoritarian power, he might have been the first and most probably the last person to actually be able to do it. But once he thought of what it might do to our literary treasures, he changed his mind (although he was illiterate, he was brilliant). But I do fear more the culture behind this matchmaking concept than the consequences of a total script change.

[AzizMostafa]

http://NoqtahCalligraphy.blogspot.com/2008/06/khat-naskhi.html

Tracing the Origins

[Pentapus]

Is it possible to get a visual example of how an "authentic" Arabic typeface looks next to a typeface developed by one of the originally mentioned teams (one Latin master one Arabic designer)?

I'm interested to see the "damage" if any was done to the forms. In the illustration we see top sections cut off. I doubt any great typographer would be so careless as to warrant the comparison.

Bonus-
Here is an interview with Nadine Chahine, the Typographer that worked with Adrian Frutiger to create Frutiger Arabic.
http://ilovetypography.com/2008/05/01/face-to-face-an-interview-with-nad...

[Pentapus]

If you read the interview you'll notice she cares very much about Arabic legibility and she considered working with Frutiger and Zapf (separately) 2 of 3 of her proudest moments.

[Saad Abulhab]

>>Oh yes! Basrah Gulf area. That was fourteen hundred years ago and my memory is a little bit sketchy!

Fourteen hundred years ago Gulf of Adan, Gulf of Uman, Gulf of Bangal, and Gulf of Basrah were all called that way in the maps (yes there were maps then), why would I have to call the Gulf of Basrah, now, the Arab Gulf, the Persian Gulf, or even the Muslim Gulf! Wasn't it the western colonialists and their local short-sighted allies and servants who eventually roped the Gulf of Basrah its name, let alone its oil!?
Basrah (not Dubai of Bandar Abbas!) was referred to historically as the Pearl of the Gulf.

>>Nonetheless, the Chinese still speak Chinese not Arabic don’t they?!

The Chinese never abandoned their language or script. The Muslims of west China, are distinct people, they chose Arabic script, and still do fiercely.

Dear Behnam, I was raised in Karbala, were I was *thankfully* exposed to a lot of Asian cultures, especially Persian. As a boy I struggled reading that Naskh Taleeq style for a little while! The way I see it, Persians, Turks, Afganis .. etc own what is now called the Arabic script, as much as the Arabs, and may be more. After all, in Iraq, we owe it to the Persian clergymen that Arabic had survived under the Utuman Turks (The grandfathers of today's Iraqi Persian clergymen who sold out to "Born Again Christian" Bush!)

Matchmaking may be an irritating and passing away "suck up", or may be a way to advance and enrich the Arabetic typography, it all depends on intention and manner of execution. Total script change is a "total sin", under any pretext! It is as much an anti Persian sin as it is anti Arabic. I would rather see Persia go back to its Aramaic Pehlavi or Avista scripts than Latin!

-Saad

[AzizMostafa]

Q: What do you enjoy most and least about designing type?
A: ... I hate kerning Arabic.

In response to Nadine_Chahine+ Pentapus, I had 2 postings here:
http://www.typophile.com/node/28634

[nadine_chahine]

Very interesting discussion. Just some random thoughts:

. Designing a harmonious Arabic and Latin is an exercise which does *not* necessarily mean that the Arabic is subservient to Latin, or that it looses any of its qualities

. The Matchmaking project is actually 5 different approaches to this problem, and should not be bundled in one lump

. There are many different ways to look at this exercise and many different ways to solve the problem. For example, compare my typefaces: Koufiya with Frutiger Arabic and Palatino Arabic

Koufiya: Both Latin and Arabic were designed at the same time; both were treated in a way that they work together but they do not sacrifice their integrity. The Arabic relates to Early Kufi.

Frutiger Arabic: the Arabic was designed to work as a signage font that can sit on the same signage system as the Latin. As it is meant to be seen in large sizes, the drawing treatment is similar to the Latin, and the style is a mix of Kufi and Ruqaa.

Palatino Arabic: the Arabic was designed to function as a bookface and is Naskh in style. The actual drawing style does not follow the Latin, but the optical size and weight are a match. This approach is on a more ideological level.

. As we stand today, we have a tradition of complexity in our manuscripts as well as a much simplified version in our traditional typefaces. The simplification was a result of technological limitations that barely exist today. However, the legacy of this simplified style exists, so do we keep it on even though the reasons for it have disappeared? Is there any typographic value in Simplified Naskh?

If you're interested in these questions, please wait till I finish my PhD in legibility studies! I'm testing the effect of complexity on the legibility of Arabic and I'm hoping to get some answers... I'm sorry I can't give more details or participate more often here as I really need to study!!

[behnam]

The visibility of Arabic script is a matter of vertical space. The line spacing in the font.
Beside the obvious physical law, that anything in smaller size is less visible than in larger size, there is absolutely no visibility problem with Arabic script... unless it is viewed through the Roman eyes.
But looking at it through the Roman eyes is completely unnecessary. Because Roman fonts do not need Arabic.
If there is a special project which requires materials in Arabic and Roman, tastefully selecting an Arabic font on one hand, and choosing a Roman font on the other, can do the job eloquently.
But Roman in Arabic fonts is needed for everyday use and for such font, Roman characters have ample room to position themselves properly beside Arabic characters.

There is a couple of exceptions however, that Roman fonts do need to contain Arabic. Fonts that provide some technical support to a device for its functionality, or technical support for functionality of communication between devices. Fixed width fonts and OS system fonts are of that nature and reasonably they should be considered as Roman fonts containing Arabic.
The issue of visibility only arise in such context. Meaning when Arabic script is deployed in a font fundamentally Roman.
For fixed width fonts, there is practically no expectation for aesthetics and Courier New for example, is quite successful to provide an Arabic support for what it intends to do. There is no typographic expectation.
For system fonts, the situation is complicated. They need to include Arabic for technical support but they can't provide the typographic needs of the script. Yet their presence is highly noticeable.
The abomination of Microsoft's Tahoma for Arabic script (which has now apparently become a school of thought!) was actually a creative approach to address the visibility issue of Arabic script within a Roman system font. For Arabic script in a Roman font, the choice is between shrinking the characters to fit the Roman line space (therefore visibility problem), or chopping off the script and maintaining the same magnification as the Roman characters. Tahoma chose the second and successfully so, for what it intended to do. But what it intended to do, by no means was to create an Arabic font.
Apple in Mac OS 10 initially had the same approach as Tahoma, with its system font Lucida Grande. But after the first or second update, this approach was abandoned to address the problem of noticeable presence of system font in Arabic rendering.
The system font of Mac OS 10 now allocate Arabic rendering to a specifically designed Arabic font called Geeza Pro. This is a very well rounded simplified Naskh font and although it hasn't resolved yet the spacing and somewhat the magnification issue, but it does address the noticeability issue and the Arabic rendering at system level on a Mac is quite good.
I think this is the approach that should be considered when for technical reasons a fundamentally Roman font needs to provide Arabic support too.

So the issue is not visibility. It's the space. Zapfino is not less visible than Times. It only needs more space. And this is *not* a deficiency. For the same magnification, Arabic script needs more space. This also is not a 'deficiency' to be 'fixed'. The difference is that for Roman script Zapfino is a design choice. For Arabic it is inherent to the script.

When the space is inherent to the script, designer can only play within and around that space. And why should you do it otherwise if you are designing an Arabic font? Of-course the creativity can go anywhere it wants. It can increase or decrease the vertical space of the characters. But it is very important to emphasize that this creativity should be through Arabic eyes, not Roman.
I'm drafting this text with an Arabic font that I made with this concept. It provides perfectly balanced visibility and magnification for both Arabic and Roman. This balance is necessary for Arabic use in which Roman is vastly present. And although this Arabic font is not intended for Roman only text, the text I'm drafting right now is quite acceptable for non professional use. This Arabic font does not cut off head and legs of Roman characters. All it does is that it puts a little more line spacing that is a tad excessive for Roman only use.
Unless there is a technical purpose, Roman font makers need not include Arabic to their fonts.

[Thomas Milo]

The event which Vladimir caught in his cartoon is being posted here:

http://river-valley.tv/conferences/arabic_typography_2008/

My talk is already available, the ones by Tarek and Titus will follow suit.

Again, I was afflicted by airconditionitis after a week of unfresh air in Dubai. Sorry.

Thomas Milo
DecoType
www.decotype.com

[k.l.]

I'm testing the effect of complexity on the legibility of Arabic and I'm hoping to get some answers ...

This sounds really interesting!
Maybe the test could take into account earlier discussions about what is actually measured, what results are supposed to tell, and thus how such tests be set up. (Esp. Peter Enneson's comments here. Is it enough to measure speed?)
The crucial question is, what will the test prove? For example, whatever will turn out to be more legible by whatever definition (to be defined) and under whatever test set-up (to be documented) -- will the result indicate that this or that kind of design is more legible per se? Or does it acknowledge that readers (participants) are used to, and thus trained in, reading it, and read it more "fluently"?

Two aspects: Make sure participants are not just students but cover all ages and type. And ask if and what they usually read -- after the test. So, have academic vs non-academic, young vs old people, which at the same time may help bring together people who read books, or newspapers, or online (TV captions?), or not at all. (Maybe even ask what kind of books and which newspaper(s), where possible check in which typeface these are set.) That way, one would not get some "abstract" information about typeface's legibility but can put them into context, and find out if -- or not -- test results and reading habits correlate.
Maybe it would be interesting to perform the same test in different countries?

My wishlist applies to Latin-script reading tests too ...

Best wishes, Karsten

[AzizMostafa]

1. Thanks Karsten for bringing about the effect of complexity on the legibility of Arabic and for these 2 links:
http://www.typophile.com/node/50834
https://typography.guru/forums/topic/51365-forwarding

2. So, have academic vs non-academic, young vs old people...

... males vs females and fonty vs unfonty ...

[k.l.]

Thanks for digging out more links!
Yes -- as diverse as possible.
One exception maybe. If possibly, avoid fonty ones, they know too much to serve as unbiased test participants. They may judge more than "feel".  :)

Oh, I should add, the idea was brought up by Nadine Chahine above (end of her post), and I just cited here in the first two italicised lines.

[AzizMostafa]

Have just recalled this one:
http://www.typophile.com/node/28777#comment-169753

@ Oh, I should add, the idea was brought up by Nadine Chahine above...

> Oh, I should blame the Kangaroo + recall the Zebra:
https://typography.guru/forums/topic/39362-forwarding#comment-169428

Thanks Karsten for reminding me with flowers

[behnam]

Legibility is a study in design. I'm sure Nadine reflection on this issue will be very valuable.
I am not a designer. I'm an amateur font maker that put the already designed characters together. Albeit, I modify them extensively. But this is not a practice in design. It is a practice in making a font that works for me.
That's why my point was about 'visibility'. The physical part of the issue, which I strongly believe has confused many designers.

[finedesign]

@Saad
Sorry, I meant to reply sooner regarding your post. I was waiting on a response from someone regarding this forum. They never replied, so I’m just moving forward.

See, I told you I would be put in my place. I am truly asking these questions as a learner. I looked up the two typefaces (if we can call them that?) and in a cursory search only found info on Musnad which I do indeed recognize. It’s written on the old Marib Dam from the Sabaen days. They also used it in Ethiopia, since Saba extended there too. The other example, Jazm I couldn’t find.

Maybe I shouldn’t have said all disconnected letters = Latin. I guess I just didn’t understand the reason the characters in your fonts were disconnected, and it reminded me of Basic Arabic, which seems to be along the same lines as Matchmaking, only with less integrity. So I assumed it was a bow to Latin or western perspective. But thank you for correcting me.

However, your argument leads me to support Matchmaking even more. Arabic is even more flexible than I imagined!

I also appreciate your online friendliness, Saad. It is refreshing.

@behnam
To answer your question as to “What’s the point?” Because it’s so drop–dead gorgeous to see the two together! It’s gestalt at it’s best. That’s an opinion from a jaded, western mindset, yes. But I think it’s progressive and very functional.

I have a suggestion for everybody…
All this talk can be a waste of time if we don’t use visual examples and discuss it on a case–by–case basis. I don’t think anyone, Huda herself, is trying to say Matchmaking is the cure–all. But in some settings, I think it works best. But perhaps you will convince me otherwise.

Behnam mentions he developed an Arabic font. Several others here I’m sure have too. Why don’t we put several cases side–by–side? I took an example of one of the Matchmaking typefaces (Nadine’s Frutiger Arabic), and put it next to a couple “correct” (I assume?) typefaces. This is only a starting point, and I am not suggesting my alternates are improvements. But I have included the source files for the below so that you can modify the logo according to your preference (using your own fonts) so that it conforms to your ideas of correctness. I would like to see them all side–by–side.

The examples follow. Sorry I couldn’t find a better example…I'm not too keen on this one. Please click the image to view the larger format, or download the source files and typeset it with your “true” Arabic fonts.

http://l.paulwreid.com/imgs/typophile/baraka_2.jpg
http://l.paulwreid.com/imgs/typophile/baraka_2_big.jpg
http://l.paulwreid.com/imgs/typophile/baraka_5.jpg
http://l.paulwreid.com/imgs/typophile/baraka_5_big.jpg

By the way, code and img tags do not seem to work here, contrary to the “formatting options” stated below. How do I post images inline if I can't use html?

Click here to download the source files. [1.6 MB]

If somebody has a better idea than this, I’m game.

salaam—
paul

[Saad Abulhab]

Salam Paul

>> Arabic is even more flexible than I imagined!

This is exactly the point: Arabic is very adaptive and flexible, let us not limit it by rigid rules or styles. I am finishing up a family of fonts for Jazm , the pre-Islamic ancestor of modern Arabic script, just to illustrate how the history of the script is a crucial part of its present and future. I will share sample of if soon.

>>I have a suggestion for everybody…

Good suggestion.

-Saad

[Vladimir Tamari]

Tom, I am now discovering various aspects of Tasmeem. With its use, one is given various options for adjusting word spacing, dot and vowel positioning, and choosing variant shapes for a word, for example a regular beh or an elongated one. Its very nice. While Tasmeem also has a 'mind of its own' after all it is an intelligent-font technology, it allows the user to interact with it to produce the desired effect. OT type is a 'take-it-or-leave-it' matter- the inputted text is exactly as the designer has made it, no more and no less.

Paul, your suggestion and example of posting graphic samples of what we are talking about is spot on. To post an image online use the 'Insert Image' link. I am not sure what inline means! Keep up your interesting and perceptive feedback. Of the three examples you gave for بركه للصلب Nadine's Frutiger works best because it matches the logo and the Latin in style. It is well designed and belongs to the same genre of Arabic 'sans-serif' styles that now has many examples, including my forthcoming AlQuds font, also with round dots.

In the image below, I will try to illustrate some responses related to the wonderful discussions above. In the early 1960's I read Eric Gill's book on typography and designed my AlQuds lettering to make an Arabic font along the same lines as his Gill sans-serif. Matchmaking! Yes I must admit it could be described as such. More than that Gill himself lived and worked in Jerusalem, so I missed meeting him only by a couple of decades and may well have consulted him had I done so! But the way I went about it was to keep Gill's designs in the back of my mind while I researched early Koranic Arabic calligraphy, newspaper fonts, the handwriting of children, modern calligraphy and many other influences to arrive at what I felt was the simple generic shape of Arabic letters. The modern embodiment of AlQuds is monoline, while Gill's glyphs have the typical Roman variation of thickness.

Saad, I recently designed a full Latin to match AlQuds's style and its six weights, so that would be an example of matchmaking in the other 'unsucking' direction to use your terminologhy :) having the Latin match the Arabic. In the image it is easily seen that Gill's is the superior font, but my AlQuds Latin makes up for its shortcomings by matching the Arabic in overall height, thickness and the orthogonal line endings. I also designed the Indic and Arabic numerals to match the corresponding languages in my font.

In the colored text I illustrate the point that first made me react against Huda's Matchmaking project. In the images khtt published, Latin text's x-height had horizontal guidelines within which the Arabic letters were fitted. Arabic has a potential in-built superiority over Latin in that visually each Arabic word is different from the other in its word-form. Latin is boxed-in within basically rectangular boxes that look much alike. The colored outlines show how the top part and bottom parts of the Arabic have a jagged outline indicating some hints of which letters are there, while the Latin outlines are just a uniform un-individualized pattern except in a few instances. I wrote about this word-shape aspect in 1974 in this article http://www.khtt.net/article-2603-en.html .

In the Arabic matched to Frutiger almost each letter tries to fit within a horizontal line even lower than the Latin x-height; for example the dal is smaller than usual and the initial tooth of the seen reaches upwards to the same height. Let us not throw away the inherent legibility advantage of connected Arabic words in the name of modernization, Westernization and globalization.

Nadine I hope you will not take my views amiss; and I really look forward to seeing the results of your Arabic legibility research. Good luck.

Lastly I must admit that my cartoon reflects the envy of an old man who had to proofread text in downtown Beirut amid the smoke and smell of molten lead cast by Linotype machines, and who had to use Rotring pens and white ink on paper to painstakingly draw his letters and correct them, and to wait months to wait for an answer to a renown Japanese font designer on how the challenges of modern printing were met there. An envy of Huda's generation with access to Fontlab, Illustrator and allied font-making software that makes the creation of a 'new' font and its publication and sale a matter of a few mouse-clicks! This is wonderful and opens up a world of possibilities, but I sincerely feel that we must be the more critical and discriminating in designing, judging and using Arabic fonts.

Benham, I appreciate your perceptive comments on Arabic-related fonts, but must you even mention the Turkish disaster that overtook the Arabic script there?! :)

Aziz, with flowers
_________________
Vladimir Tamari
Homepage: www.ne.jp/asahi/tamari/vladimir/
Arabic lettering: www.khtt.net/person-2306-en.htmlari/vladimir

[Vladimir Tamari]

If the reason for typefaces to be kept “simple” is restrictions of the technology, then something went terribly wrong. Technology should be gentle and allow for whatever may make sense. And whether designers prefer simplified or complex designs should be entirely driven by their design visions, not by technology or by their ability or inability to master it. With OpenType, I fear that often it is indeed the latter than determines design.

Karsten - This is very important. As an example, one reason I simplified AlQuds is to abridge the number of metal type shapes needed for Arabic. With computers this does not matter any more. However, perhaps due to the influence of Latin Sans-Serifs, simple Arabic is in fashion again, and it has an important place in education, for example to help wipe out the shameful illiteracy in many parts of the Arab world and beyond. I say 'again' because square Kufi is such a simplification, centuries old! Can you please explain the difference between ACE and Tasmeem in a bit of detail?

Benham can you please provide a sample image of the Latin and Arabic you referred to in your interesting discussion on vertical spacing?

[Thomas Milo]

Best wishes

Thomas Milo
DecoType
www.decotype.com
www.decotype.com/aycaadakum_mubaarakatu-n.pdf

[AtoZ]

"How do I post images inline if I can’t use html?"

Here's one work around that I've found for "inline" images:

When you add an image (using the "Insert Image" button), by default the image is placed at the end of your post. However, if you click on the "Preview comment" button, and then scroll down to the editable text window, you can cut the link to the image and then paste it (again in the editable text window) where ever you wish it to appear in your comment.

It's a bit of a nuisance to have to do this, but it's better than nothing.
 
         ~~~~~~~~~~~~~~~~~
         When going from A to Z,
         I often end up At Oz.

[behnam]

>>Benham can you please provide a sample image of the Latin and Arabic you referred to in your interesting discussion on vertical spacing?

Sorry Vladimir. I just discovered this thread had a second page!
You can go to http://wiki.irmug.org/index.php/X_Series_2 and pick any font. They all follow the same concept of magnification between Arabic and Roman. Some with more success than the others, in terms of visual harmony.
As you know, I'm not a designer and visual harmony is something I'm improving over time. But the visual magnification of all of them follows the same concept I described.
The font I was referring to in my discussion is XB Zar which to me it has the best 'standard look' for a Persian text. (the site above has XB Zar as default font) But it was one of my first fonts and the harmony is not as good as some others. But the picture below is of my latest font, XB Kayhan. As I said I'm not a designer. My main concern about these fonts is compatibility with Mac platform and language coverage. For Mac platform, I have to add AAT. I also add characters to support languages of Iran and neighbouring. I thought Afghani people will be too busy for a while to have time to make fonts for themselves. So I cover quite a few languages.
In the picture of Kayhan (with its different faces) I mixed a news from BBC in Persian and English. I think that should give you an idea.

[AzizMostafa]

The end of Arabic Typography + Calligraphy:
https://typography.guru/xmodules/typophile/files/Rayhan.pdf
https://typography.guru/xmodules/typophile/files/BetterArabicAdobe.pdf
Sketch it and MaryamSoft.com will re-do it, however complex.
Thanks Behnam with Flowers
https://typography.guru/forums/topic/58495-forwarding

[Vladimir Tamari]

That is very nice behnam. I looked at some glyphs in Kayhan and it is very solidly and elegantly constructed. The Latin-Arabic balance in size and style is excellent. This is so in the outline and shaded display fonts more than the small text font I wonder why - probably because the added textures and stylization add an extra level of uniformity. It is very generous to provide these fonts as freeware! I notice that the spacing is very tight; as Aziz once stated 'the tighter the better'.

Aziz - the sample you provided is beautiful - how was it made from a technical point of view?

[AzizMostafa]

> ... how was it made from a technical point of view?

Impossibly easy way! This might help:
http://fontforge.sourceforge.net/source-build.html

> ... as Aziz once stated ’the tighter the better’.

Personally, I:

1. always try to be accurate when quoting:
https://typography.guru/forums/topic/56166-forwarding#comment-283452

2. never forget to acknowledge others works:
https://typography.guru/forums/topic/56166-forwarding?page=2#comment-292120

3. like to exchange flowers with hard-working Typophiles

[Vladimir Tamari]

OK Aziz the quote should have been "The tighter, the nicer?!". I have installed FontForge on Linux Ubuntu on a PC, but the screen keeps flickering at an uncorrectable 60 Hz so I have not used it. How do you compare FF with FLS?
おはな　と