Local language usage is on the rise and the major hindrance for adoption is of course, the keyboard.
While few players provide the transliteration service that enables one to type in local language using Qwerty keyboard, Kalibonca goes a step further and has built the Brahmi Indian Keyboards.
The product boasts of a simple interface and is quite easy to use (switch on the caps key and you have characters in local language, switch it off and you are back to English keyboard) and is built on how Indians learn (and use) the local language.
Coming to the local language market, here are some interesting stats:
- ~50% of Internet access happens from Tier-II cities and only around 11 per cent of India’s 30 million Internet users prefer to browse online in English [source]
- 67% of online newspaper readers are from small towns (source)
- Indian media is embracing local languages – right from newspapers to content portals (google news/ Orkut are available in regional languages.)
But, I see lack of an intuitive keyboard as one major hindrance in adoption. The challenge is not just limited to content creation alone, it’s about PC adoption and unless the keyboards are localized, PC adoption will still be a challenge (PC penetration is less than 3% in India).
Local Language Keyboard: Kalibonca
Kalibonca does solves a major pain point that goes beyond the digital world and will find significant adoption in schools. The keyboards are available in Hindi, Bengali, Punjabi, Gujarati, Oriya, Marathi, Telugu, Kannada, Malayalam and Tamil language;and Kalibonca also has a mobile version (Brahmi Smart Phones) which they are licensing to handset manufacturers.
Interestingly, the team got the patent almost after 6.5 years of applying! – well, this is the kind of patience that they would need to crack the Indian market.
What’s your opinion?
Related startup: Lipikaar
The claims are far away from truth.
1. This is not the first keyboard for Indian Languages. They existed for nearly 3 decades
2. The layout is not intuitive as they claim. Actually, the keyboard layout should be based on scientific study on the most frequent used keystroke and its positioning on the keyboard for fast access to the most used finger
3. The keyboard layout is based on Hindi. The Kannada layout is incomplete as it has more characters compared to Hindi
-Pavanaja
Awesome product!! Do they have a working prototype? integration with some handset mnfrers?
I dont think hindi alphabets need to follow the same style as querty keyboard – its a new learning experience and keeping it simple is what really matters.
Good show guys.
It is amazing this was not done many years ago. The keyboards thus far have been too cumbersome and take years of practice for proficiency which is only good for a government job you’ll keep for years. Pavanaja is incorrect, the Kannada layout is complete, the picture in the article is simply one of the layouts. I checked their website and found more information. http://www.kalibonca.com I have not been able to find anyone who has started selling it. I noticed the patent just issued so this should hit the market soon! Finally!
I can’t figure out how John Frank came to the conclusion about Kannada keyboard layout. From the layout given on the web-site I can conclude that it is incomplete. There is no other layout given on their web-site. No download of any version (trailware, shareware, cripleware,…) is given on the web-site. Unless John Frank has access for some working versions, he can’t conculsively say anything about the completeness of Kannada keyboard.
Nowhere on theier web-site I could find any data from any scientific study conducted on the cumbersomeness of the existing keyboard layouts, the scientificness of their layout, etc. They should publish some data if they have. Just making some claims saying that their layout is the best just amounts to marketing talk.
I think there is some weight in their assertion that using phonetic keyboard has a barrier of knowing English alphabets, but putting all the characters in a liner order on the keyboard is not the right answer in my opinion. Reason is ergonomics. For a simple example, a very easy to reach key ‘G’ corresponds to the last character of ‘?’ series. Looking at the typical Hindi text, that character would almost never be used. So it is like a wasted key right there.
There is lots of research on developing ergonomic Hindi keyboards and I very much doubt if they are the first company with Hindi/Brahmi keyboards. Haven’t Hindi typewriters been around for quite sometime?
Guys – Have this whacky idea – How about tying up with Times’ TeachIndia campaign?
Think about it.
I am Mahesh Jayachandra, neurocientist and the inventor of the Brahmi keyboard – the buck stops here.:)
First of all many thanks to Ashish and the gang at pluggdin for the story. Great job guys!
Thanks for your kind words, Rajesh. Yes, we do have a prototype – a full-fledged, out-of-the box Hindi PC with our keyboard. Works like a breeze. You said “…keeping it simple is what really matters.” *You got it!* The Brahmi keyboard was specifically designed for first-time, Indian language users.
Specifically, we tested the input method to universal approval from rural kids in India, faculty from Hindi Departments in Indian schools, Ivy League professors in the US, NRIs, parents etc. If you are in Bangalore anytime soon, I’d love to demo my stuff. (Beer on me, of course.)
Thanks for your comment, Abhaya. Here’s my take. The frequently used keys is a myth as far as keyboards are concerned. The brain is plastic and reorganizes itself within weeks when learning a new motor/language skill. Literally, any keyboard can be learned given the need and enough time.
(There has been a lot of work on plasticity in the brain – hundreds of papers in the last few decades. Good starting points are those from the labs of Michael Mezernich and Jon Kass in the US and our very own Neeraj Jain at the Natl. Brain Research Institute in Delhi.)
Consider the QWERTY keyboard. It was originally designed with exactly the opposite consideration. *Frequently used keys were kept further apart so that the typewriter keys would not stick.* And yet accomplished typists routinely type at speeds greater than 60 wpm throughout their lives and this is the standard.
Finally, extending the analogy, if keys spaced far apart were really detrimental to use, there would be no concert pianists or harmonium players.:)
Incidentally, the first guys who set up this particular scientific straw man was the BUREAU OF INDIAN STANDARDS in 1991. Annex-4 of the ISCII Standard, suggests that, “…The Indian script alphabet has a logical structure, derived from the phonetic properties. The Inscript overlay mirrors this logical structure. The overlay has also been optimized from phonetic/frequency considerations. It is divided into two parts: the vowel pad on the left hand side, and the consonant pad on the right hand side.”
This is classic “Computing by Diktat”.
None of this is backed by any scientific papers in peer-reviewed journals. Think about it – all inscript keyboards have essentially the same structure for all the Indian languages. So if these paternalistic jokers are to be believed, all Indian languages have the same “phonetic/frequency considerations” – plainly not true. On the surface of it, these considerations would probably differ between, say, Tamil and Punjabi. (The subtext is the fact that these guys have assigned the vowels to the right side of the brain and the consonants to the left. Wow! Talk about neural
conditioning by diktat.)
Cheers!
I am Thilak from Bangalore. I have been in the business of selling Indian Language software for the last 13 years. I am interested in seeing demo of your product.
Regards,
This is for the good doctor Pavanaja (DrP).
This is a longish post, please indulge me. And you thought the previous one was long.:)
First of all …Ooops! Kalibonca just re-did their website and the Kannada keyboard image was not the correct image. It’s been fixed. Thank you for pointing that out.
Now, DrP, it would be useful in such posts to explain that you are involved in pushing a competing Kannada Transliteration Keyboard to the Karnataka Government. (For the pluggd_in record, phonetic/transliteration keyboards are a dime a dozen and have been around for a while. Transliteration methods also go against the grain of every local language lover in India. Try telling Bal Thackeray that he needs to learn Chinese to type Marathi.)
>>The claims are far away from truth.>>
Fighting words, DrP! Now the question simply is, can you handle the heat in the kitchen?
You seem to specialize in setting up straw men and shooting them down – odd behavior for a scientist. Let’s get to details.
>>Example 1: This is not the first keyboard for Indian Languages. They have existed for nearly 3 decades.>>
I suggest you read the copy properly or check your eyesight. There is no suggestion that this is the first keyboard for Indian languages. Rather, this is the first *simple* keyboard for Indian languages. There, that wasn’t so difficult now, was it? Let’s move on.
>>Example 2a: The layout is not intuitive as they claim.>>
The simple response is “Why?” Care to share?
Note: This statement puts you in a very special category as every person we have tested it on finds it to be intuitive.
>>Example 2b: Actually, the keyboard layout should be based on scientific study on the most frequent used keystroke and its positioning on the keyboard for fast access to the most used finger.>>
I appreciate your scientific zeal.
So actually, I would like to ask you about *your* scientific studies on the most frequent used keystroke and its positioning on the keyboard for fast access to the most used finger as it applies to *your* keyboard.
I have been unable to come up with any keyboard/input method papers published by you in any peer-reviewed journal. I checked this morning on PubMed and Google Scholar.
What data did you use to get approval from the Karnataka Government? I assume that you used rigorous experimental design, careful matched subject groups, ideally a double-blind study topped off with sturdy statistics. We are all intrigued and would love to see your work.
Let’s take the gloves off for a bit, shall we?
Peer-reviewed papers in journals count, honey-buns, not outpourings on your web site or accolades from the Bhasha boys. I also checked on PubMed for Inscript, Devanagari and Transliteration/Phonetic keyboards. NOTHING, nada, nahi, illa, zip, zotto, zuchcum since 1966. Naughty, naughty, Dr.P.
>>Example 3: The keyboard layout is based on Hindi…>>
No DrP, the layout is based on Ashokan Brahmi. All Indic scripts are derived from it, including Hindi. And yes, it is obvious that Kannada and Telugu and Tamil have different numbers of characters. And your point is…?
>>Example 4: Nowhere on theier web-site I could find any data from any scientific study conducted on the cumbersomeness of the existing keyboard layouts, the scientificness of their layout, etc. They should publish some data if they have. Just making some claims saying that their layout is the best just amounts to marketing talk.>>
Oooooooh baby, now I’m quaking in my shoes! We never made any “scientificness” claims. Again a straw man you’ve put up and attempted to shoot down. I designed the keyboards for *school children* and *first-time* local language computer users. You are not included in this market. Nor are the 5% of Indians that are English-fluent enough to use computers (MIT ref in Resources section of http://www.kalibonca.com). The real market for my keyboards are all those people who don’t compute – 95% of one billion. Transliteration ain’t gonna cut it for these people who have been ignored until now. Get it?
As a scientist, you should be the first to know that not everything is science. Design, functionality and usability can be invoked independently of science. Think Apple.
The design philosophy behind my keyboard was captured perfectly by the good folks at CDAC, in 2006. Here’s a quote from Dr Dinesh Katre, CDAC’s point man on keypad usability, from a 2006 conference in Guwahati. While he refers to cell phone keypads, he could just as well be talking about keyboards.
“The keypad layout should be designed as per the original structure of
language without bending the rules or sacrificing the details. The existing linguistic habits and preferences of users may be considered rather than introducing new rules and techniques. The keypad design should not expect from users the prerequisite knowledge or expertise of a language. The keypad should be usable for even unskilled users. Various human factors involved in text entry process may be considered while designing the keypad layout and the input mechanism. The keypad layout for Devanagari script should be standardized.”
(CDAC ref in Resources section of http://www.kalibonca.com).
Naughty DrP! You you did not read the resources section, did you now?
*The Brahmi Challenge*
As you know, in science, you either put out or STFU especially if you say stuff like, “The claims are far away from truth.”
So, how about this?
Let us collect some experimental data. Expose subjects with a knowledge of Indic languages (but with *no prior exposure to any Indic keyboard*) to the Brahmi layout versus your transliteration/phonetic layout. And ask them which is easier to use to type their names. Any time, any place, any Indic.
Ashish could be the moderator. In fact, he is planning to do the start-up disco again – this could be a prelude to the libations.:)
After the initial data collection, if you are still have the cojones, we could strive to do some EEG recordings at NIMHANS or NCBS.
So instead of being magisterial and blabbing about “marketing talk”, why don’t you get off your toots, get your hands dirty and do some real science.
Are you feeling lucky today, DrP?
Cheers!
Mahesh Jayachandra MD, PhD
Hi Mahesh,
Thanks for the reply. There are couple of things I would like to add to the discussion.
First, if the brain is plastic enough to learn any keyboard layout, I guess there is no point comparing all these different layouts
. But what I fail to understand is how can the concept of most frequently used key be a myth. If you analyze the character counts of any text, it is very clear that some characters are more often used then others. And common sense would say that they should be easily accessible. I fail to understand how the plasticity of brain negates that. Our hands are not plastic, are they?
I know the story behind the QWERTY keyboard and that only shows that a lot of things has to be taken in account in laying out the keys of the keyboard. Same goes for Harmonium and keyboard keys, all of them have been laid out as considered best, not to make them easiest to find if you do not know how to play the instrument.
Regarding scientific evidence, I have seen lots of work about evolving the best keyboard layout, one by my next door neighbor in college. I am sure you can find them in good AI and ML journals.
But I would concede all arguments if you say that you have ground tested the product. No theory can beat hard experimental evidence. And I am sure you must have been very careful in designing these tests. However here is one thing that I would like to point out. Your layout will be very easy for people to use if they do not know Hindi typing at all or are using it for the first time. It will be easy for first few sessions and for reaching till a particular speed of typing. I hope you have conducted tests for extended periods of use and then taken a feedback. I will draw an analogy with ‘touch typing’. When we start typing, it is very easy to type using just 2 fingers. But that method can only take you so far. For good typing speeds and for making sure that only two fingers are not bearing all the load, ‘touch typing’ is the thing to learn.
Finally, let me make a case for your product
. I think it will be very useful for those people who do not have to deal with computers day in, day out but probably need to use it once in a while. It will take away the awkwardness of just standing there and trying to find the right key. It can be a useful for small children when they are starting to learn to use computer. And many other places and all these are very important usage. But I think it is not right to compare this layout with inscript or any other non-linear layout based on frequency considerations and write them off.
I rest my case.
regards,
Abhaya
http://pothi.com
Pingback: Rediff to speak Different Languages |Technology and Business Startups in India