CHAPTER II

THE EVOLUTION OF THE ARABIC KEYBOARD

The first Arabic typewriter was introduced in 1914 in Egypt by Phillip Waked and Saleem Haddad[1]. The design (see figure 2-1) had to designate different keys for different variations of each letter. All Arabic alphabetic characters can have up to four display representations depending on their relative position in a word: initial, final, medial, or isolated. The Taa’ (ت) for example has three forms on a typewriter, and one must learn to hit one of three keys, depending on whether the letter is in the beginning of the word, beginning, middle, or end (تـ, ـتـ, ت). The typewriter layout did not contain diacritical marks, nor did it contain the letter Yaa’ in its final or isolated form (Yaa’ is written the same as alif maqSuura in the Egyptian school of Arabic spelling). Therefore, it had ample space on the Shift keys for the letter variations.

With the advent of the computer, Arabic typing was radically simplified by having only a single key for each letter. The particular shape which an Arabic character takes is based on its left and right neighbors. The algorithm determining which of the four shapes to use is called contextual analysis[2]. After the key is struck, the system software does a context-analysis and determines which of the three forms of the letter is appropriate; it then inserts it.

The Arabic typewriter seems to fairly consider the frequency of letters in the design. Seven of the ten most frequent letters in the Arabic language are on the home row compared to three of the ten most frequent letters in the English language in the QWERTY keyboard.

Very few resources exist as to how the Arabic typewriter evolved into the Arabic keyboard. But by observation, it is clear that the basic letters were left in their locations to simplify the transition of trained typists to the new layout. The letters on the number row, however are different. Those characters are: ـهـ ـعـ ـغـ د ذ ط ظ . The last three characters where removed from the keyboard since contextual analysis eliminated their necessity. But the other four letters where put in odd locations, which created a problematic situation particularly for the frequent letters د and ذ. The key designated for the two-letter combination لا was maintained in the keyboard layout. This ligature was originally designated in the Arabic typewriter because of the absence of contextual analysis. Since computer contextual analysis redraws these two letters to look like (لا) rather than (لـا) the purpose of this key is now obsolete. The period and comma were relocated to the keys corresponding to > and K respectively on QWERTY, which are also problematic for reasons discussed in Chapter 1. The character alif maqSuura was relocated to the key corresponding to “n” on QWERTY, a positive change since it is frequently accessed. Other characters added were أ, إ, ؤ and ئ (with the typewriter, the Arabic typist had to go back one space and insert a Hamza to create these letters).

The first alternative keyboard layout that this paper aims to propose does not seek to accommodate the existing layouts but only considers character frequency as a standard for character positioning. The second proposed layout seeks to conform as much as possible to the current design since the existing layout is to a large degree built on character frequency, but attempts to relocate particular problematic characters.

[1] Altrikah alhdithah fi ta'lim aldhrb a'la ala'lah alkatbah. Al-Jeel Publications, No date.
الطريقة الحديثة في تعليم الضرب على الآلة الكاتبة, pg 5.

[2] Contextual analysis is the process by which Arabic letters take different forms according to their position in the word, and according to the surrounding letters. For example, in the word (عمرة) the letter ain is initial, the meem medial, the rah final and the ta’ marbuta isolated.