James E. Agenbroad wrote: > It seems to me that we have three separate domains to deal with: > 1. What should be keyed as input of Indic scripts, mainly Devanagari? > 2. How shall Indic scripts data be stored and exchanged? > 3. How should Indic scripts be displayed on screens and in print? > > ISCII and Unicode are not concerned with the first. They are very > concerned with the second. There may be general agreement on the third, > but a variety of output devices are involved. Unless ISCII changed from a > phonetic based approach to a graphic based one I doubt that Unicode [...]
Totally agreed: point 2 is the only real task of Unicode; point 3 is also the task of Unicode, but to a more limited degree (and it is in fact dealt with in Chapter 9 of TUS 3.0). Point 1 does not have to be part of Unicode per se. It is, however, a implementation issue of paramount importance. The users' perception of how good Unicode is for Indic languages depends almost entirely on how well screen editing is implemented. So, I agree that Unicode does not need to do any formal steps on point 1 (apart --sorry for repeating myself-- finding a way to encode REPHA RA IN ISOLATION). However, it may be a good idea to: promote discussions on this point, compare existing solution, propose new ones. It would also be great to enhance the informative material in the TUS book about the various methods by which Indic and other complex scripts interact with keyboard, mouse and display during an editing section. Good solutions in this area would help greatly to send old font encodings to retirement. _ Marco
