User:Zaljohar/Critique of present Arabic to Latin transliteration systems.

Different transliteration systems of Arabic to Latin were proposed. Officially named as Romanization of Arabic Unfortunately all of them suffer from a problem with at least one of the following four important aspects demanded in any transliteration system: Precision, Flexibility, Familiarity, Phonetic approximation.

Precision: this entails that a word must have one interpretation as regards the identity of letters in it. Many of those systems use digraphs as letters, which will obviously lose their precision when transliterating from Arabic script that is skimmed of short vowels, which is the customary way of writing Arabic, so: UNGEGN, ALA-LC, Qalam, Online script, all will lose their precision on skimming.

Flexibility: this entail the ability to use common formatting on words like Bolding, italicizing, underlining, and changing case without changing the literal composition of the word itself. Many transliteration systems use case change to generate letters, so in these systems h for example may not be the same letter as H. examples: Qalam, Buckwalter, BATR. So these systems lose flexibility regarding Capitalization rendering them useless for official purposes which often demand upper casing. Also well recognized systems like DIN, ISO, ISO 233, SAS, SM, UNGEGN, ALA-LC, and ArabTeX, all have dots or lines or other diacritics below letters, thus affecting underlying of words, so h with a dot below it will appear as h when both are underlined, therefore rendering the text non flexible in this respect. other systems use sometimes italics to denote different letters, which also renders them non-flexible with italicizing.

Familiarity: basic Latin is what most people using Latin alphabet are familiar with. Familiarity would depend on the proportion of extended Latin characters to all characters in a system; the smaller it is, the more familiar is the system. In my personal opinion, the total number of characters in extended Latin should not exceed 40% of the total number of characters in a system. Unfortunately, this is not the case with DIN, ISO, ISO 233, SAS, SM, IPA, which have nearly half or even more of their characters in extended Latin.

Phonetic approximation This entails that an Arabic letter must be represented by a character in Latin that is the nearest to it in pronunciation. This is good regarding the official systems, but it is bad regarding the Online script (Arabic chat alphabet), which use numbers to represent some letters,Buckwalter, SATTS violate this principle clearly.

Therefore, as one can see from the above that: ALL systems violate one or more of the upper four desirable qualities in a transliteration system.

What is required is a system that is:

Precise: use monograph Latin character for each Arabic character.

Flexible: Bolding, italicizing, underlying and changing case will not affect the literal composition of the word.

Familiar: Extended Latin: total character ratio must be less than or equal to 40%.

Phonetically approximate Most characters reflect the nearest approximation possible to how the transliterated Arabic letter is pronounced.

To address all of the above goals, I have defined a very simple transliteration system satisfying all of them, I gave it the title of: SEHL which is an Arabic word meaning EASY.

EXPOSITION OF SEHL

Tilte: SEHL