Keeping in line with modern developments in computer typesetting, all
our documents should use Unicode compliant fonts. (irrespective of
whether or not they require devanagari or other Indian scripts).
Perhaps some members of the group know about these things, but I still
felt compelled to write this note for the group as a whole.
Each character has a (binary) code associated with it (even characters
that we do not 'see' such as space, linebreak, tab etc). To put it
very briefly, Unicode is a new standard for assigning codes to
characters. The first 128 characters are Latin characters and other
keyboard characters and punctuation marks. Unicode declares
(reserves) codes in the range 0900–097F for devanagari explicitly.
Other Indian scripts take up other range of values. In non-unicode
scripts the mapping of the character code to the character itself is
arbitrary and leads to utter confusion when the documents are ported
from one software (or version) to another or from one OS to another.
Sticking to Unicode means assured portability (irrespective of OS,
browser, word processor for all modern systems), and ability to search
phrases and words in that script. You can search for गुलाबजामुन by
typing it as such; no need to type gulaabjaamun or any such thing.
This is not possible in non-unicode documents. You can also thereby
mix different scripts without worrying about clashes and gibberish
characters.
Note that UNICODE is only a standard for STORING characters in a
script in a standardised manner, not a way of keying-in characters of
a script in a document. For the latter, when non-Latin scripts are
involved, there are many "keyboard mappings" eg m17n, itrans,
inscript, remington, several phonetic mappings (an endless list!). It
is possible to use most of these mappings in combination with unicode
fonts. But do make sure that you actually have unicode compliant
fonts on your machine, and use only those when you create new
documents. It does not matter whether you use Win, Mac, Linux.
Unfortunately it is very hard to convert a non-unicode document to
unicode one for scripts other than Latin. Unfortunately very few of
the old fonts are unicode compliant. Many, but not all, common word
processor Latin fonts are unicode compliant by now, but please check
any way.
Another sincere request: Microsoft bundles an UGLY font called
'mangal' for devanagari with its OS. Please desist from using this
UGLY font. The alternative Arial Unicode is far better looking and is
bundled with Windows-8 and -10. Some of our recent translations have
used mangal. Apart from being inherently ugly, mangal does not respect
the commonly used styles of ligatures.
A summary of common devanagari fonts on Windows are seen here:
http://www.lipikaar.com/support/download-unicode-fonts-for-hindi-marathi-sanskrit-nepali
Gargi, Kalimati, Lohit are good choices for running text, Lohit is the
default on Linux systems that have Indian scripts enabled.
One beautiful devanagari font I use is called Ek-Mukta. The great
thing about this font is that it has matching, Latin (English!)
characters, which most other fonts sorely lack, so bilingual text
looks very nice with these. See these links for more info.
https://github.com/girish-dalvi/Ek-Mukta
http://blog.prathambooks.org/2014/07/ek-mukta-free-and-open-source-font.html
A few other Unicode options are:
CDAC-GISTYogesh
http://salrc.uchicago.edu/resources/fonts/available/hindi/yogesh.shtml
CDAC-GISTSurekh
http://salrc.uchicago.edu/resources/fonts/available/hindi/surekh.shtml
chandas
http://www.sanskritweb.net/cakram/
akshar
http://salrc.uchicago.edu/resources/fonts/available/hindi/akshar.shtml
Hope this is useful. Those who know about these things, please don't
take offence!
Thanks and regards,
Bhas
On 11/01/2018, Bhas Bapat <bhas.bapat@xxxxxxxxx> wrote:
Dear Aamod and others,
I had a look at "aapale anna -- Our Food" and this seemed fine. I
have not seen "aakaash darshan" The kind of errors you mention abound
in almost all translations, but these are rather less important.
My concern was especially regarding the chapters on light and magnet
-- these chapters read very poorly. The sentence construction is
abstruse and has a very 'Hindi' feel to it. In some cases even the
gender aspect is repeatedly incorrect. Can someone (Madhav?) please
confirm that the two chapters (Light, Magnets) that I had made
corrections on are still around? If these can be scanned and sent to
me I will know what to look for and what to avoid for the future. I
seem to remember that the [poor] feel of the Machines chapter was
similar to that of Light and Magnets.
I will be sending a mail following this on the attention that is
needed when inputting devanagari text. I had sent this earlier to
some Eklavya members and it might be irrelevant or superfluous to
some. To all such people: please excuse me for this!
Bhas
