Testing, Spacing, Kerning and Hinting Arabic Fonts

[Vladimir Tamari]

Thanks Saad! Your refreshing reminder that calligraphy and font design are (or should be) open to change, improvement and innovation is welcome. Over the centuries Arabic calligraphy has done very well for itself and the results are the fantastic 'polished' styles of classical calligraphy, with all their in-built rules and traditions. Enter printing, and now the calligraphers feel threatened but the font-designers gradually realize that here was a new opportunity to innovate. Because of technical issues like font metrics, the visibility of displayed and printed words font designers have to explore new issues, for example spacing, and find solutions to them.

By the way the tight spacing Aziz advocates is not a necessarily bad thing. In Chinese and Japanese characters the density of strokes is very high in a given space, but it seems that the eye can read such character-words at one glance. The Arabic equivelant to these characters would be, for example, the basmala ligature. Perhaps with such software as Tom Milo's Tasmeem where such overlapping of glyphs is possible and easy to control, old and new styles of 'tight' Arabic can be displayed.

[AzizMostafa]

By the way, just have a look at what the Self-appointing prophets have done:
http://www.winsoft.eu/www/eng/products_solutions/WinSoft-Tasmeem-arabics...
Letter Spacing : Too loose.
Kerning: No
Mark Positioning: Not too bad
Ligatures: Up to the AC-practicing eye+hand of the user?!
Correct me if I am wrong.

Evaluation to what I have done is highly appreciated with Flowers
https://typography.guru/forums/topic/29348-forwarding

[Vladimir Tamari]

Since there are already several typophile threads dealing with Arabic perhaps we should ask Miss Tiffany to kindly have them moved with forwarding to this Special Interest Group.

In the thread referred to by Aziz there is an image MRALLAH1 of his font. I am not an expert of Koranic script, but I liked the elegance and smooth connectivity of the font. In some places it is too tight for example in the word كفار the kaf touches the feh. However in the word كزرع the ra and zain and kasratan fit beautifully within the curves of the ain. In fact the marks everywhere fit very nicely within the outlines. Can you tell us more about this font family - its name, number of ligatures, kerning, hinting and other features. How long did it take you to develop? Congratulations with flowers.

[AzizMostafa]

Thanks Vladimir, To make the fonty flash one-mega watt smile, I did
I. make more than 1100 ligatures
2. place more than 99 H+V marks for the 1+0 (wide+narrow) parts.
3. write AC-juggling word-macros.
4. screw the tight parts at hard spaces.
Please allow 2-3 weeks for further details + new samples.

[Thomas Milo]

>By the way, just have a look at what the Self-appointing prophets have done:
>http://www.winsoft.eu/www/eng/products_solutions/WinSoft-Tasmeem-arabics...

Wots a self-appointing prophet? And, why the sneer? Are you against innovation? Or are the French unwelcome in Arabic typography? Wots the point?

>Letter Spacing : Too loose.

This is a parameter, it can be set loose or tight by the user. The designer provides the default

>Kerning: No

This is also a parameter, but there are no kerning pairs, because the font is dynamic.

>Mark Positioning: Not too bad

1. Designer sets defaults, 2. ACE looks for best implementation, 3. user can adjust the position.

>Ligatures: Up to the AC-practicing eye+hand of the user?!

There are no ligatures at all. The font is less than 400 glyphs. What you see are letter block fusions according to naskh script grammar. Wherever this grammar provides more than one solution, the user can choose.

>Correct me if I am wrong."

You were wrong :-)

Tulip bulbs,

Thomas Milo
DecoType
www.decotype.com

[AzizMostafa]

Happy to see Thomas back among typophiles. Welcome with Flowers:

> Wots a self-appointing prophet?
On July 16, 2007 at 10:01 am, and in reply to my never-answered-to-the point questions, here:
http://29letters.wordpress.com/2007/01/15/arabic-calligraphy-written-by-...
Thomas Milo wrote:
...IT’S SMART FONT TECHNOLOGY ON TOP OF OPEN TYPE IN THE POST SCRIPT FLAVOUR - THE BEST OF ALL WORLDS.

Emphasizing your words "THE BEST OF ALL WORLDS.",
If this is not self-appointing prophecy, what is it?
Owing Saad for his touching words+wording...

> And, why the sneer?
"Sneer of what?" Here is a scientific forum, not anything else?!
Nothing in my heart against you, and you do agree with me that
What a man does is more important than who/what is he/she,
regardless of his/her race, religion and color?!

> Are you against innovation?
Clicking my avatar says it all?!

> Or are the French unwelcome in Arabic typography?
When it comes to typography+Calligraphy, I do acknowledge that the French are among the most fonty...

> Wots the point?
The point is made clear by your comments on the 4-mentioned parameters that are inter-phrased with " the user can choose."

>Letter Spacing : .... it can be set loose or tight by the user..
A gallant failure even if thing goes word by word?!

>Kerning: ... there are no kerning pairs, because the font is dynamic?!
A bit like a man saying to his goldfish: " I do not make love with you, because you are fonty"

>Mark Positioning: ... ACE looks for best implementation, ... user can adjust the position.
If ACE looks for best implementation, then what is the user supposed to adjust?!
One or more mispositioned mark like the zamma here?
http://www.winsoft.eu/products_solutions/WinSoft-Tasmeem-NashkFont.php,
Then what if the letter-spacings is changed or an alternative font is used after the repositioing?!

>Ligatures: ... What you see are letter block fusions according to naskh script grammar. Wherever this grammar provides more than one solution, the user can choose.
Of course not regardless of the 3 more parameters?

Your answer does emphasize that is" Up to the AC-practicing eye+hand of the user?"
Nevertheless you say: " You are wrong."

Then tell me what a fonty user do you target that:
1. must have Arab-eyes, and
2. can change 4 or more correlated/dependent parameters,
that if a change is made to one, then he/she has to adjust back and again the others?!

Pity the hopeless user?! Correct me if I am wrong with Flowers

[Thomas Milo]

Oh.

Thomas Milo
DecoType
www.decotype.com

[Saad Abulhab]

Aziz wrote:

Evaluation to what I have done is highly appreciated with Flowers
https://typography.guru/forums/topic/29348-forwarding

Aziz, I am sure any fair person would give you high mark on your font (lets call it wurud - flowers-:) but I have only one problem with it: it is not yet a font that I can buy, and select from the font menu! In other word: I agree with the question put by Vladimir, where is the beef? Tell us more.

As for Tasmeem:

In defense of Milo's fascinating work, he has produced a *working* Arabic engine that can work as a desktop publishing software to print involved texts like historical works, complete with few involved and high end fonts. Tasmeem is not a common commercial application like MS-Word targeting the general audience ... at least not yet. One can agree with some of your points and disagree with others, but you have to keep in mind what Tasmeem is really targeting. Personally, I like to see Open Type flourishing for Arabic. I do not like to think of Decotype format as a replacement for it.

-Saad

[Vladimir Tamari]

I agree with Saad that there is room for both Tom's Tasmeem and Aziz' Open Type naskh. Once the user has access to both, either can be used according to availability, the level of control needed and platform used. As in Latin fonts there is always room for excellence, and healthy competition can benefit typography as a whole.

As I understand it Aziz has made an incredible labor of love to create a unique font family (Wurud??) TTF (or OT?) with artistry and precision. Let us call it a full-course naskh meal by a master chef. We look forward to the feast.

Tom's Tasmeem work is an incredible labor of love consisting of the scholarly analysis of the roots of what he calls the grammar (the visual language) of Arabic writing systems and naskh in particular. This has been ingeniously implemented by the decotype team in software where such writing can be re-created naturally on the computer. In Tasmeem's naskh the user can control spacing and choose the details of various letters and markings. This flexibility applies to any other font adapted for the system, where it can be fine-tuned according to need or level of design experience. One might say Tom makes available a gleaming futuristic kitchen; and (as soon as any techno-phobia has had a chance to wear off) guides us to effortlessly create a superb meal to taste.

Aziz gives flowers. Tom gives magic tulip bulbs to instantly grow your own.

[Thomas Milo]

Vladimir,

Thanks for clarifying this is such nice culinary metaphors. I love cooking, you could guess as you caught me the other day baking a loaf of bread and one day I hope to present you my version of felafel!

But after some soul searching, I had to come to the conclusion that I am more of model builder where Arabic is concerned. Models you normally make from fascinating objects like intriguing cars or famous airplanes. Grammars are models of languages - mental exercises driven by fascination and curiosity, attempting structurally to understand and reproduce what can be observed. The object of analysis itself, language, is is taught by examples and learned by imitating. I felt that Arabic script should be approached from the same point of view: not as a capricious art-form irreverently called AC, but as a communication system where senders and receivers over a period of many centuries successfully shared the code. In this approach historical Arabic script is understood as a tool for text manufacture with not so much learnability, but legibility as its primary characteristic.

Azizi Aziz,

Lemme try to clarify what that implies.

Naskh is a historical script style. We made a computer model of it. That's all.

DecoType ACE is technology to make this possible (which incidentally was the proof of concept for OpenType). It is not naskh, but a programming environment we (for I am not doing this alone) developed over the years for projects like a naskh model - the ultimate challenge.

Tasmeem is Adobe In Design with ACE built into it by WinSoft, who, together with DecoType, designed a User Interface around ACE. With ACE, Tasmeem can handle any Arabic script style.

ACE was first conceived as early as 1982-83 to enable the creation of a ruq`a replica. As said, the result was not so much a font, but a computer model of ruq' a. It was licensed by Microsoft around 1993. It was Microsoft who asked us to do a naskh with the same method in 1995. But not after they had first asked us in 1991-92 to provide a realistic naskh solution within their very simple design template - but I suppose that's more a topic for the other thread.

InDesign is a typesetter's tool with enormous precision and many specialized controls - for Latin-based scripts. Some people just want it like that. With Tasmeem, we try to introduce this kind of precision into the field of Arabic script.

Against this background, Tasmeem provides the user with tools to adjust the positioning of points and vowels. And, it provides specialized kerning and tracking controls based on the structure of Arabic - which is I believe the subject of this thread.

Tom's Tasmeem (nice alliteration, but it is really WinSoft's) and Aziz’ Open Type naskh are sooner complementary than mutually exclusive.

Each is in it's own league.

Bulbs,

t

Thomas Milo
DecoType
www.decotype.com

[Saad Abulhab]

Tom wrote:

>>In this approach historical Arabic script is understood as a tool for text manufacture with not so much learnability, but legibility as its primary characteristic.

The historical calligraphic script was not primarily about Legibility or learnability. It was about speed and harmony. Read what one of the fathers of Caligraphy, Barmaki, said 1000 years ago:

الخط صورة روحها البيان، ويدها السرعة، وقدمها التسوية، وجوارحها معرفة الفصول

Translation: Calligraphy is a living picture, whose soul is representation, hand is speed, foot is harmony, and strength is the knowledge of derived styles.

>>With ACE, Tasmeem can handle any Arabic script style.

My font Mehdi, which uses both Tatweel (kashida) (not today's common dash-like imitation) and Irsaal (final form tatweel) was implemented across 5 fonts with more that 2500 glyphs and plenty of glyph substitutions. In Tasmeem, less than 500 glyphs were needed. Plus, users do not need to keep changing fonts and they have additional power and flexibilty. This was accomplished with a simple template on Fontlab with no letter chopping.

Also, Tasmeem can keep your flowers fresh: Users can not steal fonts since they can not work outside the environment.

-Saad

[AzizMostafa]

@ Tasmeem is Adobe In Design with ACE built into it by WinSoft, who, together with DecoType, designed a User Interface around ACE. With ACE, Tasmeem can handle any Arabic script style.

... I can understand that some companies lack the expertise in certain fields, so they initially rely on external providers....

Twardoch's comment on 5.Dec.2007 at the end of this topic:
https://typography.guru/forums/topic/46706-forwarding

[Vladimir Tamari]

>>>The historical calligraphic script was not primarily about Legibility or learnability. It was about speed and harmony.>>>

Saad- surely legibility was an unwritten law? A beautiful harmonious but unreadable style would have been discarded as a script evolved.

>>>ACE was first conceived as early as 1982-83 to enable the creation of a ruq‘a replica>>>

Tom- nice alliteration here too ruq‘a replica! Was there just one stroke-like curve for each glyph (not a complete outline)and the program creates an outline from this curve according to a 'nib' (qalam tip) length and angle? Some reference graphics would be nice to see. Is r-r included in Tasmeem? I guess it is time to start a Tasmeem thread here.

[AzizMostafa]

> I guess it is time to start a Tasmeem thread here.
Vladimir! Already started?!
https://typography.guru/forums/topic/50979-forwarding

As you asker earlier:
>... we should ask Miss Tiffany to kindly have them moved with forwarding to this Special Interest Group.
Also looking forward to seeing it moved along with all the related topics that appear on my page.
Hope the Miss wont miss this node with my endless Flower greetings to her.

[Thomas Milo]

>>@ Tasmeem is Adobe In Design with ACE built into it by WinSoft, who, together with
>> DecoType, designed a User Interface around ACE. With ACE, Tasmeem can handle any
>>Arabic script style.

> ... I can understand that some companies lack the expertise in certain fields, so they
> initially rely on external providers....

We do not need to rely on external providers. WinSoft and DecoType have all the expertise in house.

Thomas Milo
DecoType
www.decotype.com

[Thomas Milo]

> My font Mehdi, which uses both Tatweel (kashida) (not today’s common dash-like
> imitation) and Irsaal (final form tatweel) was implemented across 5 fonts with more
> that 2500 glyphs and plenty of glyph substitutions. In Tasmeem, less than 500 glyphs
> were needed. Plus, users do not need to keep changing fonts and they have additional
> power and flexibilty. This was accomplished with a simple template on Fontlab with no
> letter chopping.

Saad,

It's less. The total glyph complement of Mehdi is now 235 including punctuation and static Arabic characters. The ACE-driven set, that replaces the original 5000 with partial Unicode support is only 161 for full support (excluding Koranic tajweed).

Thomas Milo
DecoType
www.decotype.com

[Thomas Milo]

> Is there any Arab among you?

Ha. I must have entered the wrong forum. I thought this is about Arabic Type. But now I understand it is about Arab Type - not my type of type. Sorry to have intruded.

BTW, why aren't the Italians more prominent in Roman type discussions :-)

Thomas Milo
DecoType
www.decotype.com

[Thomas Milo]

Man yacrafu yucallimu man laa yacrafu!
من يعرف يكلم من لا يعرف

c=cayn:

مَنْ یَعْرَفُ یُعَلِّمُ مَنْ لَا یَعْرَفُ

Thomas Milo
DecoType
www.decotype.com

[AzizMostafa]

> WinSoft and DecoType have all the expertise in house.

Ya! By now, I know you and my Turkish friend:
http://rtsoft.dilafilm.com.tr/
Is there any Arab among you?
Man yacrafu yucallimu man laa yacrafu!
من يعرف يعلم من لا يعرف
c=cayn: appears in all words.
While editing my post, my little daughter has come up to kiss me on my forehead.
So I have to respond to something more important than Type?!
Correcting your vocalization, you have to replace the Fatha of r of yacrafu with Kasra!
Flowers to All the expertise in house

[Thomas Milo]

Dear Aziz,

You are confusing grammar with script structure are typography.

Anyway, since when can a Dutchman not make a minor error in Arabic, while all foreigners are happily screwing all over English?

Bulbs,

Thomas Milo
DecoType
www.decotype.com

[Vladimir Tamari]

Thank you Tom for the promise of falafel فلافل ! Aziz a daughter's love is worth the whole world هنيئا لك.

[Saad Abulhab]

Aziz wrote:

>>Is there any Arab among you?

As a matter of fact I think the managers of Winsoft are Arabs, but this is completely irrelevant since Arabic is an international script, like English. Crucial expertise and contribution of non Arabs is well documented from the days of the first "melting pot" city on earth "Baghdad". Sibowayh?!

>>While editing my post, my little daughter has come up to kiss me on my forehead.
So I have to respond to something more important than Type?!

Why not naming your next font after your little daughter, this way you can instill your love of Arabic type in her instead of dividing it :-)

-Saad

[Saad Abulhab]

Vladimir wrote:

>>Saad- surely legibility was an unwritten law? A beautiful harmonious but unreadable style would have been discarded as a script evolved.

Only to some degree. Calligraphers were more busy with harmony and visual effect than with readers inability to read the text. They felt free to mold the letters leaving the reader to solve the puzzle. As a matter of fact, complicating the Arabic script for the sake of calligraphy contributed to high illiteracy rates then and is contributing even now.

-Saad

[Vladimir Tamari]

>>>>Is there any Arab among you?>>
Me. Also Tom's dedication to Arabic has practically made him an Arab.

>>>>Calligraphers were more busy with harmony and visual effect than with readers inability to read the text. They felt free to mold the letters leaving the reader to solve the puzzle. As a matter of fact, complicating the Arabic script for the sake of calligraphy>>>
It is true that under Islam geometrical patterns and scripts were prized as a sophisticated abstract art form. This led to illegible calligraphic styles that were enjoyed as decorative puzzles. At the same time it was a period of intensive scholarship when books on science, philosophy and theology were written by skilled scribes. For those books legibility would have been important. This lead to the highly legible naskh. Later for official Ottoman government documents the highly legible ruq'a was adopted.

[AzizMostafa]

I feel reminded of a previous discussions about the goods and bads of DecoType/WinSoft’s InDesign plug-in Tasmeem, a layout engine for typesetting Arabic which uses a special font format. Thomas Milo usually presents its capabilities with help of a high-end typeface of a special style. (By the analysis of this style he arrived at the technology, but in the end the technology is suited for any other style too.) To my amusement or regret (depends on my mood) this provoked some critics to blame him for propagating this particular style which they consider outdated — and disregard the layout engine because of the style of the typeface used for presentations! A layout engine of course is “mere” technology and is ignorant of anything like designers’ sentiments.
Same is true for iKern. If you object to a Fell being kerned at all, then I had preferred if there were no “automated” in “applying automated spacing to pre-designed types”. ;-)
_____________________________________ k.l. on 12.Jun.2008 1.48pm
Read more...
http://www.typophile.com/node/46301