Questioning ICT Myths

Posted on April 24, 2006  /  219 Comments

At an ICT policy consultation meeting in Dhaka, organized by the APC (, two participants prompted a much needed reality check for a room full ICT4D professionals. Mridul Chowdhury, a research affiliate at the IT Group in the Berkman Centre, Harvard University also a director of D.Net, kick-started the discussion with a presentation that questioned some of the key assumptions that that form the premise to much work in the ICT4D space. For instance, a lack of community information can be resolved by building a telecenter; poor governance can be solved by introducing ICT systems into governments; a lack of market information can be solved by building e-commerce portals. Chowdhury stressed that certain underlying factors had to be taken care of first; the information gaps that we’re really trying to solve need to be identified; the governance process needs to be reformed before throwing ICTs at it, etc. ICTs cannot solve all the world’s problems, and before one tries to throw ICTs at it, one should really assess whether ICTs are really the optimal solution. Nalaka Gunewardena, Director at TVE Asia Pacific continued the discussion in his aptly titled presentation, ‘Rhetoric vs. reality.’ He raised a pressing question: do good policies necessarily imply good implementation? His response to this was not necessarily, and often, not. The overarching issue is that unless certain conditions are fulfilled and critical issues resolved (for instance infrastructure bottlenecks), ICTs alone will not solve the problem at hand (even if it is at all the optimal solution) and even good policies will not achieve the desired outcomes. It therefore appears that many of the common assumptions, which are the premise of many ICT4D projects, to which much hope and funds are pinned, need to be scrutinized if development goals are to be reached.


  1. A debate in the open?!

    Ruvan, you said almost everything that needs to be said. Donald, with all due respect, Sir, I have to agree with Ruvan that you are influenced by the CJK ideographic scripts. The Japanese wear their own blinkers about this thing. However, contrary to what most people say here and elsewhere, you CAN make a font that is also UNICODE COMPLIANT adhering to the principle that every character should have its own glyph — 1820 plus as you say. It would be just a tad heavy, but still practical.

    Sinhala is correctly considered as one of several Unicode scripts that belong to the Indic group of languages. I agree with the Unicode technology. Without Unicode standards, we cannot make a font for Sinhala that shows all the characters that Mr. Gaminitilleke’s book shows. Again, Donald, thank you for sending it to me. It is gratefully appreciated. That book should be with every font maker.

    I plead with much humility for you all wonderful people to download the file at the following location (I replaced the right-leaning slashes with pipe character to avoid the message lopping off at slashes. So, please replace them with the slashes when typing in the address bar):
    Once downloaded, read the advice Unicode gives Tamil and Malayalam font makers. It does not have much for Sinhalese because as we know, we have been an enthusiastic bunch. That’s okay. Our sister scripts are the Dravidian ones. (Remember we got it from Pallava’s?).

    Donald’s kayanu-bænði-muurdhaja-sayanna is shown in it. Not only it, it also shows how to make the ‘kxo’ (that is romanized Sinhala. Why should I avoid it. I have great respect for the two and half years of work gone into it. x=muurdhaja-sayanna).

    If we followed the advice that Unicode is taking such pains to describe, this is what should happen:

    I type kxo (or its equivalent on a Sinhala-only keyboard), and then the following shows up on the screen:
    Then I say Oops and backspace once. The display changes to:
    kayanu-bænði-muurdhaja-sayanna (=kx).
    Notice that I did not type kombuva at the beginning. The font did it for me automatically. Unicode says this in so many places. This is why Unicode is repeating the word ORTHOGRAPHY. The font has the intelligence to make display the characters the way it ought to be written.

    I think we were too attached to the Wijesekera keyboard and did not seriously look at the INPUT method. The fault with the Wije keyboard is that it contradicts the Unicode standard and gives marks (e.g. kombuva), the same status as base characters. So we proudly used all those esoteric devices like ZWJ. (Make Pali fonts separately). I watched the Linux developer group talking these things till they started insulting each other. Then I proposed them to use romanized Sinhala to insulate the nice language from the rest of the world. I think they threw me out. Sinhala Unicode group too did the same to me, but worse to Donald.

    As Sinhalese, we should learn to listen to criticism and even gobbledygook for the sake of civility and openness in discussion AND especially because we are talking destiny for the Sinhalese. If you guys are going to have a discussion, (I dread a debate), please include the Sinhala users and Pali users too. Maybe tape ahead. The composition of the panel should be like the wages Boards: Employers and employees (I am talking what I knew 30 years back, sorry for being such a fossil). In this case, the big wigs that talk in English about Sinhala and the down trodden Sinhala-only unfortunates.

    May I also remind you that there are two user groups (did we ever consider user groups) that you did not take into consideration: students studying abroad and Pali users world over who often call Sinhala script the Pali language.

    In the meanwhile, I’ll try to make a font that follows Unicode and fits on the Latin-1 pages.

  2. Hey, Dharma!

    Nice to hear from you.

    You know, I think I can do a font for you that will at least work in Windows XP without any extra installations. No Sinhala keyboard, no that service pack, this service pack. Maybe you’d have to extend the keyboard to European style — no big deal. maybe you need to pay me (though far less than what you would be required to pay Microsoft to use the font developed by our guys.)

    I think you won’t reply.

  3. Dear Ruwan

    Quote”a cintec committee comprising of sinhala scholars of various persuations sat down to set some standards for terminology” unquote

    Please read the article by “Computer imbroglio in keeping with Sinhala language
    by Aelian de Silva” it is on web on pdf format at

    Quote”since most work was done on microsoft operating systems which kept changing drastically their input and rendering methods (from dos to win 3.x to win 95…).” unquote

    This was happenening all over the world but you were not aware of typology and typography of Sinhala. All you know was a Sinhala typewriter and could not think byond this instrument.
    This was the case for all indic languages.
    gihan did try to show him how the system works

    This happend at the ICTA in the presence of Mr Manju Hatthotuwa and the President of Sri Lanka Association of PRinters. Dr Gihan got very angry and walked away from the meeting.This was a very undiplomatic way of expression. Mr Manju will confirm this incident.

    sinhala unicode (SLS 1134) rather than unicode

    Unfortunately I cannot paste images into this site. Please check the unicode chart for Sinhala and SLSI 1134 chart for Sinhala

    sinhala just hadn’t shown up on the world scripts map as yet as far as electronic support was concerned – pure economics.

    No economics behind SInhala but the main problem was — Sri Lanka had no National Standard for Sinhala characters. Nobody ever knew how many sinhala characters were in sinhala hodiya.

    i argued sincerely trying to convince him,

    “”Some code points in UCS have been assigned to combining characters. These are similar to the non-spacing accent keys on a typewriter.””

    I can understand the typewriter technology but Computer is not a typewriter

    Combining characters follow the character which they modify. For example, the German umlaut character Ä (“Latin capital letter A with diaeresis”) can either be represented by the precomposed UCS code U 00C4, or alternatively by the combination of a normal “Latin capital letter A” followed by a “combining diaeresis”: U 0041 U 0308.

    Donalds Comment:

    What ever the combination is used the German umlaut character Ä is listed in unicode as one character. Defined as ISO 10646 Table 2 row 00 Latin -1 Supplement
    DEC 196 as = latin capital letter A with diaeresis

    Like wise you have to specify all Sinhala characters in the unicode = SLSI

    You have given the combination but you have not given the end location of “DU” …..etc etc

    Because the end location is not given various font makers keep the “DU” in different locations thus the data is not compatible. We are using a computer not a typewriter.

    All this is beacuae you have the typewriter concept within you and cannot think byond

    – making the fatal mistake of aligning it with the japanese (or CJK) –

    You mean to say that Sri Lanka is a develop country with a higher GNP and GDP that of Japan China or Korea. Also you think that Sri Lankas’ computer literacy is higher than that of Japan China or Korea. They have solved this problem having thousand of characters in thier language.

    This is because thay had proper educated people who had national feelings and made a proper National standard of thier own languages without bending the heads to UNICODE.

    for sinhala since all indic languages (as all european ones) are allocated just a single code page

    This where you went wrong

    Has UCS been adopted as a national standard?

    Yes, a number of countries have published national adoptions of ISO 10646, sometimes after adding additional annexes with cross-references to older national standards, implementation guidelines, and specifications of various national implementation subsets:

    * China: GB 13000.1-93
    * Japan: JIS X 0221-1:2001
    * Korea: KS X 1005-1:1995 (includes ISO 10646-1:1993 amendments 1-7)
    * Vietnam: TCVN 6909:2001
    (This “16-bit Coded Vietnamese Character Set” is a small UCS subset and to be implemented for data interchange with and within government agencies as of 2002-07-01.)
    * Iran: ISIRI 6219:2002, Information Technology — Persian Information Interchange and Display Mechanism, using Unicode. (This is not a version or subset of ISO 10646, but a separate document that provides additional national guidance and clarification on handling the Persian language and the Arabic script in Unicode.)

    donald to honestly admit that his interest is no longer in language but in a patent

    Why worry about my patent if you have a proper system.

    Since there are enough problems in your system (SLSI1134 =r Sinhala unicode)

    I am the only person giving a solution to these problems.

    The Intellectial Property Act No 36 of 2003, Part IV Chaper XI

    “invention” means an idea of an inventor which permits in practice the solution to a specific problem in the field of Technology”

    By default without any negative writings my patent should have being approved but yet pending.

    Remember I have the 100% copyrights over my publication ISBN 955-98975-0-0 and its industrial acceptibility.

    Donald Gaminitillake

  4. JC,

    As a typical Sri Lankan I prefer the software in my PC to come free of charge.:-)

    But, yes, I do not paying a reasonable amount if you can provide me a solution that I can use and that will make it possible for me to communicate with others, they too not facing the hassle of downloading and installing. Let me study your previous posts and come back to you.

    But not the other hand I feel cheated. (No, not by you but by the government of Sri Lanka.) Under the e-Sri Lanka programme the country has spent handsomely to do whatever you say you can do for me now. The amount spent can by no means be small. I am sure at least half a million to one million dollars or even more would have been spent on it for all these years. (I know several people from ICTA and UCSC worked on it.) All these were our money. The money of the public. We did not pay them to come up with another fonts set (which they have done so far) but to make standards and put Sinhala into major OSes.

    So after spending so much by the country, don’t you think it would be fair for me to expect Sinhala free of charge in my PC rather than paying another developer?

  5. Quote
    The amount spent can by no means be small. I am sure at least half a million to one million dollars or even more would have been spent on it

    You are correct. After Sri Lanka spending all this they were unable to give a proper product.
    It is corect for you and the general public to expect correct Sinhala in computer free of charge.
    Who takes the resposibility of the incorrect product!!

    Donald Gaminitillake

  6. Dharma,

    Thank you for taking me on. Why I like you to come here more often is because of this:
    You bring out the MOST FUNDAMENTAL questions: Why do we do this at all and whose paying for all this? And you remind that it is because you want to type Sinhala with the least hassle.

    Then you say you have already paid too much and someone else (JC) fixes the problem but you don’t want to pay him.

    Here’s a story:
    My car developed a severe oil leak. So I took it to the garage. The garage owner told that they have to lift off the engine and replace the gaskets. The price was as much the price of the car. After three weeks in the garage, I paid and took the car back. The leak remained and the garage owner gave technical reasons why it still leaks (part of it being the station wagon is 11 years old). I went to another garage. That mechanic put it up the hoist, showed me the leaking oil and said that I had to take one of the wheels off and replace some parts there because the oil is shooting from the side into the engine and politely hinted that the other garage did not diagnose correctly. Back home, I got my girl to help me figure out where the oil comes out of. So, I got in the car and and pumped the pedals and turned the steering wheel etc. while she peeked into the engine. We found that the leak is from the hole developed on a tube that runs automatic steering fluid. Our handy man who had come to do some fixing in the house was looking on and offered to help. Though not not a mechanic, he read the repair manual, got dirty all over and fixed the promblem in 45 minutes. By now I had paid out a lot of money and was angry. Did I pay the handy man too?

    Ok. Let’s get back to business and bargain.
    Here’s the description of the font:

    The font is Sinhala script, of course.
    Input is through US-International keyboard layout. (It’s what European do. It is set up in the Control Panel — no loading of software)
    It is Unicode compliant:
    Partly that means, when you type b followed by a, it shows bayanna. In other words, you type the way you speak. The font displays the way it should be written. Sinhala orthography is hard to program. it will come in stages. But the first version would be pretty much okay.

    Installation: You copy the file into the Fonts directory. I’ll send a batch file that automates it too, if you want — Do’t get too carried away. Think of the mighty green bucks that I charge, like Microsoft. (Major difference: I am a Sinhalese living abroad and wear the sarong whenever possible. Similarity: We both are American businesses, I can speak English AND Sinhala, not much Sanskrit that you guys write nowadays).
    The characters are not the greatest looking but would be better than most you see. I’ll improve it as I get paid. You get upgrades free (Microsoft charges for upgrades.)

    Condition: You pay for each copy of the font file. (your friend pays for their copy). Every user should register with me. And I’ll sue anyone that is caught using it without registering it with me. (US is serious about software piracy. I am on the same boat with Microsoft on this.)

    Here’s a hint: Think like a business person. The Sinhalese had it for thousands of years until the British broke its bones. Recall that Anuradhapura was named after a merchant, not some coconut politician (Coconut means white inside, brown outside). So, if you think my efforts are worth money, then form a company over there to be the sole distributor (Sri lanka allows monopolies, if I am correct). Then loby the government to buy from us (you and me) than upgrading all those government owned computers to Windows Vista, whatever it is going to be called when out. Get the other developers like Microimage as your partners. They are pretty reasonable and intelligent.

    So, the bottom line:
    How much would you pay for one copy of Sinhala font that will run natively on Windows XP that is also Unicode compliant? let’s negotiate. If you want to privately discuss, write to me at

  7. Are you kidding?

    Half a million to one million Us$ was spend on this?

    No doubt both sons of the Agency’s loku mahattaya are studying in US!

  8. þoppi velenða,

    I am not kidding. You are talking about money. I am talking serious work and business. Come out of hiding. Are you scared? Of whom, may I ask?

  9. þoppi,

    This thing about sending rich Sri Lankas to study in the US.

    It is amazing the amout of money they have. They do things that upper-middle-class Americans cannot dream of. For instance, they get apartments for the children to go to school from and pay out-of-state tuition rates for college. Semthing like $60,000 a year. I am just lucky that my daughter became a National Merit Scholarship finalist. A University offrered full tuition and boarding for her. I was seriously thinking of finding a high-powered job in Sri Lanka, just to be able to pay for her college. I know those are cushy jobs too. You just sit and talk big.

  10. As far as I understand, this whole Sinhala in computers issue has become a big political game over the years.

    For all those years, Prof. Sam has been smartly exploiting this to achieve his own personal goals. He uses this as a ‘hingannage thuwalaya’.

    The day beggar finds his wound is healed, he knows his means of income is over.

    Similarly Prof. Sam knows the day Sinhala is introduce in the major OSes, he does not have a project, and he loses his means of (a) income and (b) power. So he tries his best to delay introducing a Sinhala standard.

    Till Sinhala is introduced in the major OSes, Prof. Sam can always make his political masters happy by making offerings in piecemeal. (As we all know his political master knows nothing about this issue and to him Unicode is perhaps the name of a fancy restaurant in Hambantota. So he does not care.)

    So Prof. Sam makes his offerings time to time and little by little. Not everything at once. One day he introduces SMS in Sinhala. One day he introduces a key board in Sinhala. One day he introduces Sinhala e-mail etc. We see this from state television, and these simple things are presented as great achievements. In reality. we know these are nothing but tamashas. For example we had a standard Sinhala key board for at least ten years. It is not something one developed based on Mahinda Chinthana.

    Anyway, Prof. Sam’s political master is happy with this game because it looks like he actually does some work, Prof. Sam is happy because this gives him the two things he ever wants in life; namely (a) money and (b) power. State television is happy because they get a story to run. ICTA is happy because it has a story to publish in its news letter.

    The only party who are not happy is the users, who are continued to be cheated by Prof. Sam and clan.

    For how long Prof. Sam continues to wash rice for his grand mother like this?

  11. Dear Harsha

    “So even hypothetically if donald is right do you’ll think all these players will agree to implement his method? Then what will happen to O/S support to all the applicaitons which are supposed to emerge with Sinhala?”

    I was trying to find a answer for this.
    At last I found the answer.
    I found it in the following book
    You can purchase it at “Gandara”
    I thank Mr Dian Gomes for introducing such valuble book to Sri Lanka.

    The name of the book is

    ” Who Moved My Cheese” by Dr Spencer Johnson

    also you can visit

    Donald Gaminitillake

  12. It takes a long time for this blog to download now. Can the Admin pepare the blog to access date/week wise from now on.

  13. Donald,
    Great find. I think most of us need this training, specially those who are clinging to the products that they made with half-baked ideas.

    Keep wishing for a free freedom — it isn’t there. Stop blaming people and act!

  14. Thanks JC for the comment

    After the “Chesse” the blog got stagnant was wondering that now everybody got scared that their “Cheese” has been moved are now running in the maze to find more better “Cheese”

    In another thread Tamil Language has just commenced.
    For Sinhala there is an incomplete set but these guys never did anything for Tamil Language.
    I was the only person publicly voiced equal usage of both languages Sinhala and Tamil using a computer. Both languages do have a problems when using a computer.

    I have the Solution but these people are scared even to give it a try because their “Chesse” will get moved.

    Donald Gamnitillake

  15. JC,


    Keep wishing for a free freedom — it isn’t there. Stop blaming people and act!


    Who wants free lunches? At least I do not. All I said was we have spent so much for this so called standards development and it is gross ridiculous if someone suggests that I still have to pay for using Sinhala in my computer.

    Why should I need to pay twice (once to Sri Lanka government in tax and then to you) for the simple task of using Sinhala in my computer?

    Please enlighten me what actions should I take.

  16. Dharma,

    OK. You want a font that works and free to use? Write to me. I will register you as a tester. Anybody here wants to be a tester may also do the same.

    My efforts are since 2002. I am not a Government agency that has unlimited resources. I buy software with my humble earnings. When I devote time for my native country and language, I rob from my livelihood. In this project, I am the Business Decision maker, the Business Negotiator (with Microsoft and Unicode), Technical Expert, Language Researcher, Impact and Feasibility Study provider, System Analyst, Native Language user, Alternative System (Sinhala Unicode) Investigator, Consultant, Programmer, Font Designer, Typography Student, Typographer etc. and the Risk taker. It is my sleep I sacrifice. I neglect my family. I neglect my clients. No one pays compensation to me.

    I lived in Lanka for 37 years and perfectly understand the mentality of socialism — expect the government to provide and then blame it for failure.

    You are trapped in the notion that Unicode means Sinhala Unicode page. That is only a concept by Unicode Inc. that furthers the business interests of it’s members and especially the directors. They are not forcing you to accept their idea. They only think that it is a good idea. For me, it is NOT a good idea. The reason is that it isolates Sinhala from the inner circle of Unicode users. Chinese, Arabic and even other Indic languages can afford to use their own Unicode pages because they have large user bases that can provide a useful network of communication — independent internets. Small countries like us should not adopt isolated Code pages. (None of the European nations did. Fraktur and Gaelic are different scripts but are still based on the first two Unicode code pages.) Choosing a code page is OUR choice — individuals as well as the government. Making fonts is our responsibility too. This decision is too grave to be entrusted to a handful of bureaucrats. It’s people’s work. They own and use the language.

    Here is a quote by a pioneering mathematician and linguist that is watching my progress (it was obviously an email message with some spelling/language errors):

    “In the years up to say, 1960, one studying science at a university would also study latin and prehaps greek, because much of the older stuff was written in these languages. Newton wrote as much in Latin as English, and indeed, the notion of papers written in English, or french or german, was unheard of. Still, times have change[d], and because the english, and later americans, took an early lead in technology, much of the rest of the world came to use these languages [English versions].

    India was for a while colonised by the british, and the retention of english there as an official language, is more a case of not trying to put either of a dravidian or h[i]ndu language over the other, and so they created this foreign language as a “lingua franca”.

    It’s the same in europe, too. Much more is written in english, even for germans in germany, than in german. It is more because english has a larger vocabulary, and much of the language stuff has been nutted out. English is the new Latin, in this regard.”

    We cannot isolate the Sinhala natives from the rest of he world because of a script. While preserving Sinhala, they need to know English too. Basing the Sinhala script on roman character set would facilitate learning of English by exposure. We have found that over the past years English is only the domain of the elite. I know it is a language that can be learned by self study and effort. That’s how I learned it when I (along with others in my 8th grade class) was force promoted to an English-only GCE class. Free access to the Internet coupled with Sinhala based on first two Unicode pages is what I think would hold the Sinhala user withing the greater community of the Internet.

  17. The discussion can continue in a new thread. The current thread is taking too long to load for many people because of the length, hence it is being closed.

    Please click on link below to view or contribute to the discussion on Standardizing Sinhala for IT applications.