<div dir="ltr">A while ago I wrote a letter to Donald Knuth and asked him about the copyrights of "The TeXbook" and "TeX, the program" textbooks and also asked what he thinks about the bidirectional algorithm of Omega and TeX-e-Parsi. I recieved his answer today. I would like to share his answer with you so I just write it down:<br>
<br>%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%<br>Dear Vafa,<br>
<br>
Of course I would be delighted and honored if you took the time to translate<br>
"The TeXbook" and "TeX: The Program" into Persian.<br>
<br>
You probably know from my webpage that Japanese, Russian, Polish,<br>
and French translations of "The TeXbook" already exist; thus<br>
the question is, how does one go about making another, when the<br>
work is copyrighted?<br>
<br>
The answer is that you must find a publisher who is willing to publish<br>
the translation. Then it is usually routine for your publisher to<br>
sign a contract with my publisher (Addison-Wesley), especially<br>
if your publisher has a good reputation.<br>
<br>
Nobody else has yet even floated the idea of translating "TeX: The Program"<br>
into another language. Indeed, that seems unlikely, because sales of that<br>
book are so small that no profit will be made. You would need to find a<br>
philanthropist who is willing to donate money to this cause; but<br>
really, how many people would ever read the result, who couldn't<br>
already read the English version anyway? (Most of the material<br>
is in the formal C language, not in English.) Thus, while I would<br>
be happy to see a great interest develop in this book, I would<br>
also be really surprised if it would happen, because so many other<br>
important books exist that should probably be translated first.<br>
<br>
About your other question: Many years ago I received letters from<br>
Persia about Farsi-oriented version of TeX, with some samples of<br>
the output. If my memory is correct, I heard that there were at<br>
least two such systems, in fact. Naturally I congratulated the<br>
authors on the fine examples they showed me; but I know nothing<br>
of the algorithms by which they handled bidirectional matters.<br>
Nor do I know anything about the internals of the Omega system<br>
(although I expect Omega would follow the Unicode conventions<br>
quite faithfully).<br>
<br>
So I can't answer that question. But I would imagine that a system<br>
developed in Iran would produce books of finer quality than<br>
a system developed elsewhere, even though I greatly admire<br>
the knowledge, experience, and taste of Yannis Haralambous,<br>
simply because an outsider cannot totally understand the<br>
subtle aspects of such a venerable culture.<br>
<br>
(And incidentally, I have no good ideas about how to typeset<br>
Farsi in the traditional Nastaliq style with maximum beauty!)<br>
<br>
Best wishes for 2009! -- Don <span>Knuth</span><br><br><br><br></div>