Transmitting files with diacritics

[Ashok Aklujkar]

In an attachment to a recent message to Indology, the following helpful
statement was made: >The only font characters that can be relied upon to
always display correctly are the 62 simple Roman characters and the 10
digits. In spite of all the magnificent electronic technology available and
all the work done by international standards groups, the other characters,
especially the upper 128 are subject to a wide range of "interpretation" by
virtually all available software. The font as it appears on your screen
will be influenced by the version of your word processor, the version and
type of operating system, the country code pages in use, and various other
factors.<

Being loath to correcting someone's retyping of my computer-printed
(usually complex) research articles, I have been using the system given
below for the last few months to send my articles for publication. It does
not solve every problem, but it does reduce the work on the clerical side
of publication considerably and, when I get the proofs of my piece as a
hard copy, I have to do very little correcting, if any, unless my original
contained mistakes.

Sequence:
(1) You type your journal contribution in your own font with all the
diacritics using the keyboard layout you are used to.
(2) After your text has reached the final form you change the characters
with diacritical marks globally and matching the case (upper case, lower
case etc.) to 'letter + number' pairs according to the scheme given below.
(3) The file is sent in this changed form either on a disk or as an e-mail
attachment. If the person doing the reconversion on the other side is
unfamiliar with the language, a copy of your printout is also sent to guide
him or her.
(4) The other side replaces the 'letter plus number' pairs according to its
keyboard layout for characters with diacritical marks, globally and
matching the upper and lower cases.
(5) It does whatever minor reformatting, checking or editing that may be
necessary and sends you a printout as proofs.

I know nothing about computer programming. I developed the scheme simply as
a practical solution that could be used until a universally accepted ideal
solution is developed. (If you believe that an ideal universal solution
will be developed soon, you will also believe that all car manufacturers
will soon standardize car sizes and sizes of all spare parts in the
interest of eliminating polution that is caused models and spare parts
discarded because of minor differences in sizes.) It is not terribly
inadequate and saves much work as far as indological writing with
diacritical marks is concerned, although for special purposes (e.g. texts
with accents or files that contain profuse use of letters followed by
numbers) a few 'buffers' may have to be introduced to block global changes
from producing unwanted results. One can easily do that by inserting
periods, commas etc. between a letter and a number following it. However,
even without such devices the present scheme saves much work for the
publisher and the author.

-- ashok aklujkar


Ashok Aklujkar's system for indicating diacritics through the use of numbers.

The same principles apply to noncapital (or small) letters and the capital
letters.

Replace globally as far as possible.

While replacing, remember to tick "match case" in the "Change" dialogue box
(your computer may have different wordings for these.)

a1      =  long a; similarly for "i1", "u1," etc. Thus, number 1  stands
for the macron or bar on top of a vowel and indicates greater length.

r2      = vowel r (usually pronounced as ri or ru). Thus, 2 indicates that
a dot should appear below the preceding letter . Similarly, "l2",
"h2(visarga)", "t2" ,  "d2," "n2,"  "s2" (retroflex sounds).
For the aspirate retroflexes this will not pose a problem, since "t", "d"
etc. will be changed first.

Note that the long vowel/vocalic r will look "r21". Therefore, its
conversion to r carrying a diacritical mark should be carried out first by
using "21". Since its occurrence is very rare, this may be done
non-globally.

m3      = anusvara
n3      = nasal consonant at the end of the ka-class/series
s3      = palatal s, the one which is usually indicated by an acute accent
on top of s.
Thus, 3 indicates most top marks such as dot and acute accent used to
indicate nasal off-glide, nasality and palatal nature.

n4      = nasal consonant at the end of the ca-class/series, tilde

Q5      = single quotation mark that appears on the right of the quoted
part. And APOSTROPHE (the latter used for avagraha). The latter situation
will not be common. So, change it first non-globally, if necessary.
Q6      = double quotation mark that appears on the right of the quoted part.