INFITT is an international organization which connects, Tamil Scholars, Government, IT Professionals and Public.
Every year it conducts “Tamil Internet Conference”. One time in India and Next time in any other country. This year “Tamil Internet Conference 2014” has been conducted in Pondicherry on Sep 19,20,21 2014.
This was my first participation to a INFITT conference.
100 papers were presented from the scholars from 9 countries.
It was a great place to meet most of the Scholars in Tamil.
Around 50 scholars came from Malaysia for this conference.
So happy to meet my Malaysian friends after a year.
I presented a paper on “Open-Tamil” a python library for processing Tamil Text.
Here is the paper
Here is the presentation
Sibi from fsftn gave a talk on “Introduction to OCR using Tesseract”
My friends BalaVignesh and Arthi BalaVignesh are researching on OCR using Tesseract.
They are building a web application for training Tesseract for Tamil Text. They gave a talk on their research.
There were many talks in various topics like Fonts conversion, Text to speech, mobile application development, Spell Checker and more.
ElanTamil from Malaysia explained their work on Tamil SpellChecker using hunspell and Grammar Checker using LanguageTool.
Most of the talks were pure academic and there were not much demonstration on practical implementations.
There are tons of research happening on Tamil Computing, Linguistic areas. But the sad part is no one is ready to share their works for public.
Many Universities run funded research on various topics, but they are not ready to share their works.
OCR, Text to Speech, Annotated Corpus, Speech to Text, Spell Checker, Grammar Checker are the highly required softwares. People are asking for them for more than 10 years.
There are many academicians did funded research by universities on these areas and created some working products with the help of their research students. After they retired they package their products and selling them.
As they see that not many people are interested in buying their products, they expect government to buy their software and distribute to all public for free.
I had a discussion with the participants asking for releasing their software as Free/Open Source Software.
But, most of them are not ready for this. They had huge fears on this. If they open source their works, they fear that some big company will take their works, sell and see huge gains.
They really had huge research and created few working software. If I have to create similar software, I have to invest more then 10 years of research, which is impossible.
If they opened their research result and their working software, many people can jump in the Tamil Linguistic area and improve their software.
There are many open source developers are ready to contribute for Tamil. But as we don’t know from where to start, we stand still on the starting point itself.
The existing software sellers, ex professors are not ready to share their works.
They keep on telling that “I have spent 20 years of research on this. Why I have to give it for free? Why I have to open source it? I have to take back the huge revenue for my works.”
They all forget that they got paid for their research works by universities, i.e by public. It is their duty to release their works for public.
I agree that if a company invests huge money and creates some software for tamil, it can sell it and expect the ROI. Even it can sell the closed source software. If the software is really useful and working perfectly, people will buy it for sure.
But these Ex Professors build their products based on their universities fund. The universities should own these software and release to public as Free/Open Source Software. But, these universities are not aware of this truth and these professors sell their works.
This is the great loss for Tamil Computing and Tamil People.
English and other languages are having great software as most of the linguistic research by their universities are released as open source.
Thats why English has so many software available.
I dont know how many decades it may take for Universities to release their tamil research works as open source.
Till then, let us leave these ex professors worrying and wondering on why their software are not, selling.
I dont know what will happen to their hard work and software, after their lifetime.
It is a happy news that few young open source enthusiasts started working on Tamil Software.
There is open-tamil python library for processing tamil text. It can convert 25 types of Tamil Encoding to Unicode. It has tamil to IPA conversion, which is a base for Text to Speech conversion.
Tesseract is being used for Tamil OCR development. Libreoffice got spellchecker and grammar checker.
I hope we can get more contributors for these projects. If they grow well, Tamil will get great open source software.
Apart from these thoughts,
Good stuff about this conference:
- Met many good contributors for Tamil Computing.
- Many papers gave new ideas for new open source tamil software development.
- Co-ordination was good for the talks.
- Food was nice.
- The Dinner Treat given by CM was awesome.
Things to improve:
- Make the Conference free for the audience. So that interested people around the city can participant. The current models enables only paid members to talk and hear the talks.
- When there are three tracks, place the notice boards and banners to show, the track, talk, and time details.
- Add the Table of Contents in the Conference book.
- Release the conference book in creative commons license.
- Do something more than yearly conference.
- To increase membership, explain the benefits of members in the website.
Thanks sir for the recognition. This reminds me that I have to do more and continue these projects. These projects are being driven by great volunteers around the globe. I dedicate all the praise and prize to all the volunteers.
The next conference will be in Singapore.
Hope we can create more open source software for tamil to talk in next conference.