Here I tried to sketch the problem of encoding of Sumerian
and Akkadian texts.
There are several places where the Sumerian and Akkadian
texts were put into computer. If somebody will write some ancient
texts into computer, he has to choose the way of encoding. In my
opinion he should consider the following important aspects:
READABILITY
- how will the text be readable including the special
characters used in transliteration
EXCHANGE
- if a bigger amount of the texts is available, how easy can be
the texts exchanged with other collegues and via E-mail
PROGRAMS COMPATIBILITY
- how succesfully can such a method be used in several program
products or in different operating systems
PRINTING POSSIBILITIES
- how can the text be printed
On the IBM PC compatible computers with DOS I see in principle
only two possible methods to encode the texts:
I) WYSIWYG - with it is meant, the method to overwrite the upper
ASCII with special characters used in transliteration in
Sumerian and Akkadian languages, i.e. creating screen and print fonts.
Readablity:
The text is to be visible without problems on the screen and
it is possible to read it like in a true book edition.
Exchange:
The best way to exchange the texts is when both users use
the same redefinition of the ASCII table, if not there are several
programs that enable a conversion. The problem can arise if
somebody writes only bu2r with accent for second and other scholar
uses bur with lower index 2, it should be unified!
To use E-mail to exchange transliteration with upper ASCII,
there is only one solution that I see now: to uuencode the file by
the sender, then send it and by the recipient to uudecode it.
Programs compatibility:
In my experience there are no problems in programs that
use ASCII. The text in ASCII serves as a source text. That
can be easily converted and used in several programs.
Printing possibilities:
A special program must be written to prepare print fonts
of the special characters for the respective printer.
II) PSEUDOTRANSLITERATION - with it is meant, the use of lower
ASCII up to 128 in fixed combinations, to replace i.e. the s with
hacek with sh, or a replacement by other characters $ or with some
other character, the accents and lower index number are replaced
with numbering after or inside of the sign by normally positioned
number, as i.e. bur2, bur3 etc.
Readability:
The text is difficult to read, but one can get used to it.
But by the increasing number of the special characters, like
in the case of Akkadian, where in a transliteration many special
characters are indispensable, is the transliteration hardly
legible. I mean the characters like sh, zsadee, t., ch, aa,
and others, sometimes character like in d_immatu character d_
and the fact, that each special characters should be also in
small like in capitals, so in this case a text can be hardly
readable, i.e. u2-$a-ri&, $ara%&u for shara:t.u seems horrible
to me. I find unconvenient that a user needs more time then
a user with WYSIWIG encoding method.
We need also some comments written in English or other
languages, than a transliteration of s with hacek as sh is not
suitable, because you can then find e.g. after a query on "she"
also the "she" in English comments.
Exchange:
I guess, that there are no problems in sending the text
encoded this way to other collegues. By the different use of
lower ASCII it can be converted. The problem with bu2r and
bur2 can arise here also. A conversion is more difficult,
because a user should replace one sign with two signs, or
vice versa, e.g. sh for s with hacek, etc.
Programs compatibility:
You can find troubles when sorting, i.e. sha will be in
normal sorting program after sa and before si, that should
be together, and every word with sh has to be after the last
word beginning on s. I found no problems by the graphics.
Printing possibilities:
By printing the text from an application you will not get
the result that you expect, because the program does not know
that sh should be printed as s with hacek and so on. Then each
time you will print something from a source you have to converte
the signs to get a true s with hacek on the paper from sh or &.
There are also some scholars staying on the bridge between
using these two sketched methods. They use standard lower as well
as upper ASCII codes to encode the texts. For tsadee stays i.e.
French cedi, by this use is also an exchange via E-mail difficult.
Conclusion:
Each of these solutions is not satisfactory enough. As it
was indicated, we pass through advantages and disadvantages by each
available solutions.
It can be said that we need to use a standard, but there is
no ready made solution.
Query:
Do somebode on the list has any suggestion? Are there other
solutions possible?
We would like to know if there is somebody, who writes any
Sumerian and Akkadian texts or texts in other ancient
langueges, in which extent, with which encoding system, what about
programs that are in use! Hej! (Czech interjection !)
thanks in advance,
Furat Rahman, Praha