Sunday, December 8, 2013

Inventor of the first phonetic keyboard driver for Indian languages, Panini of Modern times: Prof. K P Rao


K P Rao
 Presenting to you, a special star whose birthday falls on 29th Feb, inventor of the first phonetic keyboard driver for Indian languages, ‘Nadoja’, Prof. K P Rao (Read Prof. K P Rao biography)
The Shaping
My upbringing was in a remote village near Mangalore, but it was blessed with multiplicity of language and culture. I knew Sufi music, Catholic Church service and temple bajans. In School, we had the finest teachers who taught us to learn by doing. I couldn’t afford text books till class 8. But it was not a concern at all. I did my B.Sc. in Chemistry at St. Aloysius College, Mangalore, since there was no seat available in Physics for me.  But I enjoyed chemistry. After that I took up a job in BARC. This place was blessed by two legendaries- Homi Bhabha and D D Kosambi. Those times were enchanting.  Kosambi influenced me a lot. His ideas of continuous learning laid the foundation for me. I was sent to Purdue University. Part of sabbatical away, I took electric engineering without real purpose and incidentally, I was the topper after 3 semesters!
Chemistry to Computer
I was aware of what would happen with computers. It is sort of story of 55 years of love with gadgets for me. I was in the semiconductor department in BARC, so had hands on with circuitries.  I used to enjoy the concept of programming languages and its abstraction, starting from FORTRAN to C and JAVA. My interest was in type setting program. Later, I worked for TATA press which was a forward looking organization.  I created fonts of my own, putting them in a grid and creating layouts to help print Indus Valley scripts in crude and rudimentary machines.

Then I worked as the director of an organization called Monotype during 1974 when use of laser beam to print in films was just coming up. It was difficult to make fonts with special characters and incorporate them in a keyboard and make it more logical. So I, along with my friend, Deb Banerjee, wrote a logic in Pascal to process Telugu text and use digital fonts for a leading Daily in Hyderabad. It was very simple. It was done on a MPM machine with 128K RAM. We wrote the complete editor and sorter that would finally generate output to the Monotype Laser Printer machine. We tried to print Indian languages on Dot Matrix printers for the purpose of proofing.

Sediyapu
An attempt was made to print other Indian Scripts on dot matrix printers. This was to simplify the illogical Indian language typewriters with a logical input from a computer and a reasonably good output from a dot matrix printer. The editor allowed all functions such as cut, paste, copy, mark, search and replace. The final output was to the dot matrix printer. The input used the simple logical input from the English keyboard. The computer, depending on the script collected and arranged the right character shapes or fonts. The fonts were script specific but not the input system. The system we developed was good enough to print any Indian language. At a later point of time, with a lot of rework I released Sediyapu, catering to this need. I didn’t patent it for a reason- I wanted to make it open source and wanted to see others contribute to the linguistic development. Sediyapu, the program, named after my teacher, was written in 72 hours in C. Sediayapu even today, runs in Windows perfectly. This is the tribute to my teacher. All other language processing tools also use the phonetic input system invented for use in Sediyapu, even if the layout is slightly different.

Ideas: Ideals: India
Everyone is interested in dilating a problem rather than solving it. I want to do Natural Language Processing in Indian languages, on mathematical basis and arrive at a universalized algorithm for the language. Panini/Patanjali describe- grammar as the rules of assembling/disassembling or ordering of Samjna’s to produce ‘Language’ – a mode of communication. I am trying to universalize it. I don’t know what to do when I solve it, so I hope it doesn’t yield a solution! Some mysteries with symbolism identified by Panini, 7 centuries before Christ, still remain mysteries! As I see, the efforts that we can put are insufficient for generating the knowledge required.
As a teacher, if there was no textbook, no syllabus, no references for a particular course, I would jump to take that subject up. At times, I used to meditate to think of content that could be taught. I saw mutual learning there. These challenges motivate a teacher. I taught in all kinds of institutions; post graduate and undergraduate at Manipal and elsewhere but was happy teaching at IITs.. Now I am leading a research oriented retired life. Learning never stops!
Today we have information. Mindsets are set in a way where education is projected as the climbing ladder for promotions. They don’t apply themselves to gain knowledge.

 I am still a student, I attend classes, online open courses and attend workshop. I do a course once in 2 months! Greek History to fMRI.

We were enchanted with this conversation and delighted to see his down to earth nature. Numerous sparkles we saw in his eyes promised us that there is more to come. We wished him health and when we bid adieu, he concluded by saying-

I was only incidental in the problem that I solved for Indian Language script layouts for computers. If not me, someone else would have solved it. Meanwhile, I was awarded prestigious D.Lit. by Kannada University. Once announced, believe me, I could not do much learning. I hope to get back to work soon.


No comments:

Post a Comment

Vishwa Manava Concept by Kuvempu

Search this blog