Spring 2000 (8.1)

Standardizing Azeri Fonts

by Betty Blair and Ed Lake

Azerbaijan desperately needs standardized computer fonts. There are none even now, nearly 10 years after the new Latin alphabet has been adopted.

Photo: Azerbaijani kindergartners celebrate "Alphabet Holiday", showing that they have mastered all 32 letters of the Azeri Latin alphabet. Their hats spell "A-L-I-F-B-A", the Azeri word for "alphabet".

Why can't I read some Az Latin or Az Cyrillic documents that are typed on other computers? Why do these letters always come up looking like garble on my computer?

Simply, the Azeri font character assignments are not compatible with the character assignments that were used in preparing the document.

Computers are designed so that every text character, tab mark, paragraph mark, punctuation mark has its own number equivalent that all computers understand. For example, your "A" might be assigned Number 65. Their "A" might be 128.

At a basic level, these codes are called ASCII (pronounced ASK-key) which is an acronym for American Standard Code of Information Interchange, a universal system of numbering characters. All computers support at least the original 128 character standard ASCII set; and many computers use an extended ASCII set of 256-characters or other character mapping techniques.

When a character assignment is established for Azeri, you won't have any problems reading other people's files. That's why it's critically important for the Azerbaijani government to establish a standard. Eventually, standardization will occur by default, but it could happen much faster and with much less wasted energy if the government would define a standard as quickly as possible.

Why is my "upside-down e" in a different place on my keyboard than on other keyboards?
Keyboard placement of characters and the character assignment of font sets are two different issues and totally separate from each other. It is possible to have "upside-down e" on any key - wherever you wish, it doesn't matter. The operating systems today have the capability of conveniently remapping characters on the keyboard. See Chart: "Where's the upside-down e?"

But if keyboard assignment is also standardized, this will facilitate what is called "touch typing", meaning a typist's ability to learn where the keys are placed, which obviously facilitates speed.

I grew up typing Azeri on a Cyrillic typewriter. It confuses me when I have to type Azeri on American computer keyboards.
It is possible to create two official standard keyboard layouts, one for those who prefer to use the Russian standard and another for those who prefer the American. More important than layout is the character assignment which must be identical no matter where you place keys to create a letter. On the American keyboard layout, you might find "the upside-down e" ()on "w" whereas on the Russian layout, you might find it on "period". The critical point is that both use the same character codes. Then, no matter which computer you use, they both print the same letter.

I have a document that I created in Azeri Cyrillic but now I want to print it out in Azeri Latin. Is that possible?
Yes, if the character assignment is the same for both Cyrillic and Latin, then you won't even have to type a single word again, just exchange the font. It's as simple as converting font styles from Helvetica to Geneva. Very simple.

I've been using our own set of Azeri Cyrillic and Latin fonts for many years now. If a new set is standardized, then all my old documents will have to be retyped.
Don't be afraid of standardization. Computers can handle those changes with what is called a translation driver, which will also become available for all the various fonts that are out there. In other words if your "" has an assignment of 78 and the standardized font gives it a new assignment of 89, the software program can reassign that letter and put it in the standard location for you along with any other changes necessary. So it won't be necessary to retype all those documents that you've spent so many hours and weeks creating. The computer can convert them. Once standards are developed, translation programs can be developed.

In English there are thousands of font styles - Helvetica, Times, Arial, etc. Why are there so few in Azeri?
Once the character font assignment has been standardized for Azeri, font styles will proliferate all over the place. But the font creators realize that there is no market yet because so many font sets exist. When all potential customers use the same font assignment, then the market will increase immensely, making it worth all the effort to create different styles.

When I type a list in English, it's easy for me to alphabetize automatically, but not when I type in Azeri Latin. Why not?
Once again, this is a problem that can be solved when the font becomes standardized. The alphabet sequence in Azeri Latin is similar, but not exact, to English. One of the most obvious differences is that Azeri "x" is not at the end of the alphabet as in English but follows the letter "k" as Azeri "x" is pronounced as "kh".

The alphabet sequence in Azeri Cyrillic was different because it was based on the Russian alphabet sequence. Once again, if the fonts were standardized, programs could be created to make sorting automatic in both Azeri Latin and Cyrillic.

I'm used to the position of letters on my layout on my Azeri keyboard, I really don't want standardization or I'll have to relearn all these positions again.
For the sake of the future, it really is important that a standard layout be established as it will facilitate typing speed. However, there are specific software programs like KEYGO that will enable reassignment of any character on the keyboard so that even if you adopt a standard font, you can personalize your keyboard layout to suit your own preferences. The most important thing is that all fonts have the same character assignment.

I've heard that in the future, UNICODE will make it possible to print all the languages in the world on a single computer. How will that affect Azeri?
UNICODE is a universal character encoding scheme for written characters and texts. It is a standard, not a technology. As of 2000, Version 3.0 of the UNICODE Catalog is available. Instead of 256 character positions that are available in the ASCII system, UNICODE allows for 65,536 code elements. Already 49,194 characters have been assigned as of February 2000.

This coding system was created by those who understood the internationalization of business. Think of UNICODE as a huge catalog - a museum of the world's alphabets and symbols where all characters are assigned a name and a code, including those with enormous character sets such as Chinese and Japanese, as well as archaic and obsolete scripts like hieroglyphic Egyptian, Old Persian cuneiform and Greek Linear B. It can handle right-to-left scripts such as Arabic and Persian, and scripts of Asia of vertical alignment. It allows for scripts to use diacritical marks such as Azeri with umlaut ( and ), circumflex ( and ). For more about UNICODE, visit: www.unicode.org.

Already the Azeri alphabet is listed with UNICODE according to Elmir Valizade, who heads the Computer Department in the office of Azerbaijan's President. In the future, look for Azeri in UNICODE to be accessible through universal Web browsers such as Netscape and Internet Explorer.

For UNICODE to work requires sophisticated computer systems such as Windows 2000 for IBM, updated software applications and drivers - obviously an expensive proposal, especially for Azerbaijanis, many of whom cannot even afford to buy a personal computer. So UNICODE is the long-term solution to the font standardization problem. In the meantime, standardization of extended code assignments is the cheaper interim solution.

I've been looking for Web sites that use the official Azeri Latin script, but can't find any. Several places allow me to download fonts, but make nothing available to read in Azeri. Some few sites use older versions of the alphabet. I'd like to read Azeri on the WEB.
Yes, you're right. Despite the fact that numerous sites exist that relate to Azerbaijan (see www. resources. az.net), few of them are based on Azeri, many are in English or Russian. Once again, we predict that once the Azeri fonts are standardized, Web sites in Azeri will begin to proliferate.

In the meantime, Azerbaijan International has just launched a new Web site totally dedicated to the Azeri language. It includes articles in Azeri in two scripts - Latin and Arabic (familiar to Azerbaijanis living in Iran). Currently, there are no items available in Azeri Cyrillic (which is no longer official).

Also you will find language learning material and articles in English about the Azeri language itself. Visit AZERI.org or AZARI.org. - mirror sites with the same material. And let us know what kind of material you'd like to start reading in Azeri on the Web. If fonts become standardized, you'll see how quickly many individuals start to create sites using the Azeri language.

Our institution has many old documents that were published in Azeri Cyrillic. Is it possible to scan them and print them out in Azeri Latin?
Yes, once the fonts have been standardized, this will be an easy task. It requires scanning the documents, running an Optical Character Recognition (OCR) Program, mapping the document into a standard Azeri Cyrillic font, then changing the font assignment to a consistent Azeri-Latin font. You'll need to have both Az Cyrillic and Az Latin fonts installed that have the same character assignment.

Furthermore, such documents can be shared both easily and economically by posting them on WEB sites. In fact, this process is likely to become the greatest solutions to the problem of converting Azeri Cyrillic texts to Latin. It means you won't have to retype every book and document that deserves to be available in the new script. And if they are posted on the WEB, you won't have to go to the enormous costs of printing them on paper.

Deep appreciation goes to Ed Lake for his enormous contribution to the discussion of fonts with our staff since 1997. Lake opened Amoco's office in late 1991. A few months later, a contract had to be signed but there were no Azeri Latin fonts. Lake kept company that weekend with "White Out", black pen, scissors and a copy machine, and the contract was ready by deadline. Azeri officials were amazed.
Lake went on to create one of the first Azeri Latin fonts in 1992. It was meant to be a temporary fix, but "Azeri 5" fonts are still circulating today. And Lake, who now works for BP Amoco in Wyoming hasn't given up on Azerbaijan getting standardized fonts - hopefully, some day soon.


From Azerbaijan International (8.1) Spring 2000.
© Azerbaijan International 2000. All rights reserved.

Home | About Azeri | Learn Azeri | Contact us