Saturday, May 20, 2006

Certificate of authentic Science

There's a silly debate raging in America right now. Some people believe that Evolution is just a silly theory, and really some divine being created the Earth. This means that you and I are original creations, have nothing in common with monkeys, have never evolved, and so on.

Think about that for a second. A country we think is well educated is having a discussion based on stupid theories, blind faith, and ignorance.

Alright, so I don't want to annoy anyone here, but I want to point out two very important things that are useful to any student of Science.

Small Steps


Firstly, good science proceeds in small steps. New research shines light on a very small puzzle. If you read any research paper, you will notice that stating the problem takes quite some space. Scientists publish very specific results. I am yet to find a research paper that says, "I know everything, everywhere, everytime." So when I see a person who claims to be a scientist, and is very glib, my warning lights go off. A good scientist says, "I don't know" much more often than, "yes, I know." Focussing on a small problem helps learn enough to solve it. Usually, even a big breakthrough impacts a very small section of science. The structure of research is like a grid of a large number of small stones cemented together. Take a very big breakthrough as an example: Einstein's theory of relativity, and the fact that mass is not constant. Before it, mass was held constant, and a lot of research was based on this. Doesn't Einstein's theory throw it all in the water? Well, not really, because at low speeds (even that of a Formula One racer), mass is constant, and the change in mass is almost negligible. So we can ignore it. Mass changes only as we approach the speed of light, which is quite difficult to achieve for most of us. And Einstein's theory changed nothing in Biology: we still have cells that make up humans, eggs create new birdies. And zero impact in Geology: erosion still happens, plate tectonics don't change, and people still don't know how to predict earthquakes. So even a big, huge breakthrough like that has a relatively small impact overall. So when you see a theory that has such a big impact that entire fields of science get called into question, chances are it is false. Consider the changes to Science if there really is _magic_: people can fly (gravity might not exist), pigeons come out of hats (and not eggs), trains disappear (mass is completely destroyed, without energy being created), and women can be cut into half (Biology is wrong, basically). That is a pretty darn broad claim, and chances are that it is false.

Some research might be false, but one little claim cannot disprove absolutely so much existing science. This is one reason why I don't believe in the huge number of Indian Babas, charlatans who claim to do magical stuff. If they could do what they claim, mainly magical stuff, it would mean that 5 or 6 giant areas of Science would be wrong. That kind of impact is too big to not get noticed. Occam's razor says the chances of Babas being crooks is higher than the chance that Biology, Physics, Chemistry and Mathematics are all simultaneously false.

What will prove it wrong?


The second interesting thing about Science is that real Science comes with conditions that will prove the result false. Read that again. If I am a real Scientist, and give you a theory, you get to say, "What will prove you Wrong?", and I should be ready with an answer. If I don't know the answer, I am not a real Scientist. If my answer is "nothing will prove my theory wrong", then my theory is not science. It is blind faith. Take Einstein's theory again[1], where mass can change based on the velocity of the object. What would prove his theory wrong: If you took a 1Kg object, and hurled it to very close to the speed of light, and it was still 1Kg, then clearly, his theory is rubbish. This is something a novice can construct, but there are more elegant circumstances that invalidate his theory. This happens in all fields, even Mathematics[2]. In Computer Science, there is a famous result regarding the Halting problem, saying that you cannot solve the Halting problem. I don't have to believe in the result. It comes with a nice certificate which will prove it false: if you find a program P that can look at any program A and say whether A will halt or not, then you have shown that the Halting problem can be solved (and the result is false). But elegant arguments show that no such P ever exists, so don't go looking. Even such a strong result comes with a condition which will cause it to fail, completely and utterly.

This is one reason why "belief" in Science and "belief" in religion are completely different. I have not seen a single condition which will cause any religious belief to be incorrect. You cannot possibly show that it is wrong, which is surprising. You do not give proof for why it is correct (one is supposed to have faith that it is correct), and there is nothing that can prove it wrong.

See this for a similar argument: "Because of Vik." That is my entire belief. Why is the Earth round: Because of Vik. Why do cows moo: Because of Vik. Why is the atom built of Protons, Neutrons and Electrons: Because of Vik. You're supposed to have faith in those words: Because of Vik. What will prove it wrong: NOTHING. This is absolutely perfect: no burden of proof on anyone: everything is "Because of Vik". And nothing can show it wrong. What use is such a philosophy? It doesn't lead to any new understanding: doesn't lead to Physics, Chemistry, detergents, computers, watches, medicine, cars, and the other fruits of Scientific study. More importantly, since nothing will prove this wrong, this is not science at all. This is blind faith. You either subscribe to it, or you don't. People who subscribe to it cannot say why it is true, and people who don't cannot show why it is false.

So while you could claim I believe in Science, that is a very different thing from belief in religion.

And Science is open to criticism, open to dissenting voices, and open to scrutiny. Try to find holes in Newton's theories, and if you succeed: people will praise you.

Try doing that in a Theocracy.


Footnotes



1. My examples are all in Physics or such, because that is a field I know a little better than others. The ideas here are common to all of Science, and even a Masters student should have no trouble providing examples from their field: Biology, Chemistry, ...

2. Mathematics is a little more elegant, because stuff in Mathematics is shown to be always true. Pythagoras' Theorem is true, and cannot be shown to be false. There are hundreds of proofs for it, and so you could say that Pythagoras' theorem would be false if there was a right angled triangle whose smaller sides (x, y), and larger size (z) did not follow the ratio x*x + y*y = z*z. But in Euclidean geometry, this will never happen. But you're welcome to try, just be good to kittens and teddy bears in the meanwhile.

Friday, May 19, 2006

The Asian Age

Lovely newspaper, I liked it over The Times of India. So I figured I should try to read it online. When I went to their website, I was surprised. Since I don't have Internet Explorer, I'm given the wonderful message, "

"This page can only be seen in IE"

Compatibility:

Your web browser needs to support
HTML4.01, CSS2 and JavaScript to correctly view this page.

Haha! That's funny. Do the site authors know that Internet Explorer doesn't support CSS2 itself? What a joke. I seriously hope this is not the Asian Age. We Asians are much better than that. I think they should rename their site, "The Stone Age" instead. The Asian Age web admins want to come across as smart people, throwing acronyms at their audience to wow them. (Oh they need HTML, CSS, XML and PRQSST! Sweet!) Unfortunately, using acronyms doesn't make you smart. Making good, standards-compliant websites makes you smart. (And attractive to the opposite sex! Growl!)

Thinking that Internet Explorer is a reference browser is an unfortunate attitude. Internet Explorer's standards compliance has never been applauded. Try this simple demo in IE, and then try it in any other standards compliant browser to see just how much difference there is.

Thursday, May 18, 2006

Showing up for war in your pajamas

A friend of mine had a terrible experience recently. Her Windows computer was compromised, and a lot of her financial information was stolen. She got to know of it when the system administrator at the Computer Science department asked her about suspicious activity at her University account. Then she found out that her machine was compromised.

She had a lot of financial information on the computer, and with the threat of identity fraud, she was forced to cancel all credit cards. (And change all her passwords) Even though no financial damage has been done, now she doesn't know if her SSN is out there or not. Even if she used Firefox for financial transactions, her browser cache could have sensitive data.

Fraud prevention websites always say, "Keep your computer secure". This person was a Computer Science Grad student, and quite smart. Despite her best efforts to keep her machine secure, it was compromised.

Contrast this with silly, stupid financial websites that only work in Internet Explorer. In essence, HDFC securities and Geojit are saying: we want you to run a browser that only works on a software whose security is a joke. Oh, and by the way, do large transactions with this insecure browser, ok? Every time I have to interact with their support staff, I have to search for words that will express my level of frustration with this stupid requirement.

Microsoft's Internet Explorer 6 is a five year old browser (released in 2001) with a known track record of abysmal security. Any financial company recommending it is seriously demented. Firefox also has its problems, but they are far smaller than Internet Explorer, even though the numbers might be the same. The issue is not the quantity of bugs but the severity. Combine that with financial websites requiring Internet Explorer. Using Internet Explorer for financial websites is like showing up for war in your pajamas. (I coined that, so give me credit when you say this, ok?)

If your bank, mutual fund, or stock broker requires you to use Internet Explorer, you should complain. Internet Explorer has holes severe enough for security experts to scream and shout about. The financial companies (HDFC, Geojit, and many others) are deliberately forcing their customers to use an inferior and unsafe product. It is our financial security at risk, not theirs.

Within hours of using the Geojit online trading system, my girlfriend and I found a gaping hole in their security. I have been told it has been fixed, but the relative ease with which we found one makes me jittery about using their website. (Both my girlfriend and I are relative nobodies, not security experts.) In any case, their website is nearly unusable, even in Microsoft Internet Explorer.

Security by unusability.

Monday, May 15, 2006

The undervalued Xaviers education

When I was in Xaviers, doing a BSc, I was amazed at how undervalued the education was. Most people thought that people did a BSc solely because they couldn't get into an Engineering college. I actually chose to get into Xaviers. My brother was in an Engineering college, and he said that I should visit his college one day, and then visit Xaviers. And then decide for myself. This is the best possible advice anyone has given me. I went to his college, looked around for a day, saw some students, and some professors (yow!). Then I went to Xaviers, and met some students. My brother was around, and this was the first time that even he had visited Xaviers. It was a nice experience. Didn't meet any professors, but that was ok. The place spoke for itself.

I met some real hackers at Xaviers: pure brilliance in Mathematics (which is what I was studying). Three students that I saw there were sparkling genius, and everyone knew that. Not only did I meet peers, but during Malhar I met some super hackers who were much older than me. And when I went back there after graduating, I met some lovely people much younger than me. The youth and verve of the place is astonishing. At the time I didn't appreciate it much, and wasn't as socially capable as I am now. But to think back, I wouldn't have been this socially capable if it wasn't for Xaviers.

I still remember the first day of class. The first class was by Felix Almeida. He spent the entire first class telling us how we shouldn't be disappointed that we didn't get into Engineering or Medicine, and that there were lots of avenues after BSc. He also told us that he was going to take us from the point of complete indifference towards Math to a point where we'd love the subject so much that we'd have trouble sleeping till we had done some Math in the day. I remember smirking, back in the last bench. After all, who was Felix to break me. But he did. By the third year, I really loved Math, and had developed a crazy interest in the world. It was astonishing. Felix remains a good friend, and he always makes his old students talk to his current students, which I think is a great idea. When I was a student, I appreciated the alumni talking to us about the "real" world.

Now that I look back, I wouldn't change those three years at all. I spent two years with the damn race: 11th and 12th. It was the most miserable time: I met very few interesting people: everyone was interested in the stupid 12th standard exams: the teachers were terrible, and the students were equally bad. Nobody was interested in learning: Just in getting those all important marks. I shudder to think of four more years of that in terms of an Engineering degree. And the Junior College I was in was way better than nearly all Engineering institutes in Bombay in terms of both students and teachers.

So if you're in Xaviers, give yourself a pat in the back. Respect the Math faculty, both Almeida and Gurjar are true cats. And make sure you have fun! If you're making your mind: just visit the options yourself. Don't take my word for it. Two hours in an Engineering institute should have you pulling your teeth out in despair if you are sane.

Thanks to Ruchi for sneaking me into Xaviers at the right time. It was great!

Friday, May 12, 2006

Listen to Frizbee NOW!

Apparently I overlooked this when I posted about Frizbee. Their music is available for download from their website. I really wish Frizbee was on Magnatune, so that I could send them some cash for their other albums.

To listen to good Welsh music, go over to their website, and download the MP3s. Don't worry about the wierd words: their website is in Welsh, and so it looks strange. Just look for the word mp3, and get everything from the album "Hirnos", the big green one with the eskimo on it. And if you like it, please mail them telling them how much you loved it.

Thursday, May 11, 2006

Good news, Great news.

I used to blog about bad sites, but rolling in the mud is a bad way to keep clean (thanks, Aldous). So I'll talk about something positive for a change.

I'll talk about the news that I read these days, as an indicator of the way things are going on the Internet. Clearly blogs are big, but what is bigger are blog aggregators and sites which filter the good news.

The top on my list are Digg, Reddit, and Slashdot, and occassionally I read The Great News Network. And more often than not, I read Bloglines, the feed aggregator for my choice of news and blogs.

Notice the complete absence of print media, because they're just not relevant news anymore. The New York Times website is still a little better than the rest, publishing a CSS feed, so that you can read them in a blog aggregator, and pick and choose just the technology and Health. And toss that worthless Sports section out forever. (It is worthless for me, but many people read just that one section.)

It is instructive to compare the layouts and fonts, and general arrangement of the above sites with a typical Indian newspaper: Times of India. Even after blocking all the ads out, with Firefox, and a specially crafted hosts file, the page is still unreadable, and the news stories are very shallow, sometimes no more than two paragraphs of completely obvious stuff. Even a robot is better than these fellows. A very well made robot, no doubt, but it illustrates the disastrous effects of incompetence. Also, the top story today on the Times of India is how Sonia Gandhi wins the Rai Bareily Lok Sabha seat. How gullible do you have to be to think that this is newsworthy? At best, it is something that should be in a three line statement. Now Sonia Gandhi winning a marathon, or a bicycling championship deserves front page. (Sonia Gandhi is the top Congress polician in India.) The top page photograph is about Brooke Shields being happy after giving birth to her second daughter. Really, why do we care? Sorry Brooke baby, but nobody in India cares about you. We like you, (hell, we like everyone), but you're just not someone we care about.

A few years ago, we had to read the newspapers, because, like my father puts it, There Is No Alternative (TINA). Well, this is 2006, my warm friends. There Is A F#*@ing Great Alternative (TIFGA), and that is called Technology. See, you can choose blogs that list articles of interest, speak from the heart, and aren't swayed by press releases that are completely biased in favor of the PR company.

These are great times to live in. Every morning you get great news, all in clear fonts, nice CSS, and with comments that sometimes are quite lovely. Even Slashdot has moved to CSS now, I admit with a tear in one eye. Of course, Google News might serve your interest well, if you spend five minutes customizing it to your interest.

And when you need some entertainment, you can read a newspaper, and laugh at the page-high ads selling miracle creams. And more entertainment when you read these paper's websites, complete with whooshy-poppy ads.

Come on, Indian kids: make a Reddit clone for India, and get rich. How tough can it be? And when Reddit/Digg goes global, they'll buy your company for one boatload of money.

Tuesday, May 09, 2006

More work, less drama

What do you do when your software vendor uses strong arm tactics? Not much, unless you have your own 100 kilo (200 pound) gorilla for a corporate lawyer.

Software is a very funny product. When you buy it from someone, you can copy it without telling that person, and use it on other machines. If the other person doesn't find out, does it hurt them? Apparently software manufacturers think that it does, and go to great lengths to build in copy protection so that people cannot effortlessly copy their product.

What would happen if Microsoft starts targetting the small companies in India to make them cough up money for software licenses? How many roadside DTP places in Bombay have legal Windows? Or for that matter, how many *cough*colleges*cough* actually have licenses for every copy of Windows they have?

In the US, people are shocked when I tell them that software in India is all free. You can buy a copy of Windows, along with all the usual software on it, for about Rs.50. (That's a little more than a dollar, and with the sinking dollar, it is getting more expensive for tourists to tank up on software when in India!) And from what I have seen, it works a lot better than legal copies of Windows, since it doesn't matter if you lose the CD case with the hundred character CD key. I've heard conspiracy theories about software manufacturers purposely turning a blind eye towards piracy, since it helps their user base grow.

But the thing that people should be thinking about is: what can I do if I get a legal letter from the Software Alliance in India? How much of my life hangs by proprietary software? This is not just for people using pirated software, but also people using legal software. Does your company have the legal weight required to fend off a bullying attack by a software vendor? Infosys and Wipro might be able to fight off a vendor, and a vendor isn't about to bully someone as big as Infosys or Wipro anyway. But smaller companies don't have huge legal teams.

What if a disgruntled employee of yours calls the Business Software Alliance or NASSCOM, to tell them that your company uses pirated software?

What if some students call their BSA or NASSCOM to tip off their college? When I was in college, this was something kids always wanted to do, but didn't.

Tipping off NASSCOM is completely anonymous, by the way, and you can use the services of various anonymous emailers on the web.

And what if your company does pay up regularly, but is strong armed into giving more money? Do you have all the copies of your sales receipts?

With the sudden growth in the use of computers, coupled with the huge amount of piracy, it is just a matter of time before it becomes lucrative for the software vendors to show up demanding money. Some companies, like Ernie Ball, will be forced to think of an exit strategy.

Personally, I like the concept of zero drama, and more work. In any case, a lot of work that Windows machines are put to is so mundane that it can be easily done with Linux + OpenOffice. Even in colleges, the biggest use for computers is to teach BASIC, C, C++ and Java programming, something that actually works better on Linux than on Windows.

Wednesday, May 03, 2006

Music knows no boundaries

Right now I'm listening to a lovely album called "Hirnos" by a band called "Frizbee". Frizbee is a little known Welsh band that my roomie introduced me to. All the songs are in Welsh, and I cannot tell what they mean, or what the words are. My roomie told me what some of the songs were singing about, but Welsh isn't easy for an Indian to pronounce, so I cannot sing along. Even though, the album is brilliant, and I love listening to it. It is the one contemporary album that I listen to, amidst my Led Zeppelin, Stevie Ray Vaughn, and Joseph Haydn sessions.

Absolutely gorgeous music: quite easy on the ears, and very melodious. The kind of music that makes you want to fake the rock star moves. And makes you want to learn Welsh, just so you can sing along...

Thank you Ywain Gwynedd , Owain Jones, and Jason Hughes (band members of Frizbee).