TED Community » Arvind Kumar

More About Me

I'm passionate about

Neuroscience, computing, cancer genetics, cricket

An idea worth spreading

Cancer is not a disease but a symptom of altered Information Homeostasis in a cell population which is interfering with the expression of genetic information. So the problem is not just in the genes but also in the machinery that expresses the genes. @ http://theoryofcancer.blogspot.com

Talk to me about

Neuroscience, Evolution, Cricket, Theory of Cancer, Theoretical Biology

Comments

• TEDCred score: +3.40 TEDCred reflects your contribution to the TED community.

• A comment on Talk: Pankaj Ghemawat: Actually, the world isn't flat

  Oct 24 2012: Nice talk. It is indeed important to bring the data to people. If we leave it to estimates in absence of information and data, over and under estimated are guaranteed.
  But I do not quite agree with Ghemawat's take on Globalization. It is not important that the local and global trade, telephone calls and other such variable are matched.
  We know that in modular networks (here network = country/economy) few long range connections -- in this case inter-country connections -- are sufficient to bring in what is called the small-world effect. The point is that the world can be effectively a global economy without appearing to be global. in terms of numbers. We all know how the effect of Lehman Brother fall percolated across different economies.
  Also it is important to know what is being traded? Money is often not the best descriptor. For instance cell phone trade volume (in terms of money) may not compare much with others but it changes the quality of life dramatically.
  
  So quality and not quantity of interactions should be taken as a the indicator of how global or flat (who came up with this term anyways) we are.
• +1

  A comment on Talk: Clay Shirky: How the Internet will (one day) transform government

  Sep 25 2012: Very nice. Clearly internet has the potential to change the way we govern the world. But I am not sure if people are ready or the governments are ready may way, or more importantly the lawyers who dont use github...
  
  The first problem is that internet is not as free as it seems and it will not remain free and accessible. Tim Wu (The Switch) makes a nice case about how every technology starts as open source and eventually becomes inaccessible as governments starts to use it for their ends.
  
  Next if people are not educated they are only going to misuse the media. The laughing cat video was a nice example.
  
  So if we really want to exploit the potential of the internet to govern the world, (1) we need to ensure that people are educated to use the internet as a media properly and (2) very importantly we need to keep the participation cost low and internet free and accessible to those who are educated to use it.
  
  Also, for once Clay goes a bit over borad with the version control systems (git etc.). I am not sure if we want an update or patch on our law every day like we do for Linux systems, on install new LTS version every year. And every one knows problems with the peer review system...(Clay no offense... I very much like your idea)
  
  To me the power of internet is in going away from Democracy and introduce MERITOCRACY where right people decide on the things e.g. scientists decide on science and sportperson on sports policies and so on ... Internet can really create an unbiased but meritocratic system. Projects like Linux have worked out nice because only those who can program and understand the system are making edits not every random person
• +1

  A comment on Talk: Read Montague: What we're learning from 5,000 brains

  Sep 24 2012: Too long introduction. Where is the content in this talk. I wish if he had said something about these 'insights' and 'embryonic discoveries'.
• +1

  A reply on Talk: Sarah-Jayne Blakemore: The mysterious workings of the adolescent brain

  Sep 20 2012: Well, yes you can determine the age from structural MRI, you can find out the age of a person from teeth and other similar methods exist... But you would not say that teeth determine the cognitive abilities or height of a person which also changes during adolescence....
  
  I agree that structural MRI has been very useful in locating certain pathologies in brain. But we have to admit that the link between structure of the brain and its neural activity and function is just not understood. Yes, you can generate awareness for neuroscience, but I am worried about how people who are already not well informed, can take such talks more seriously than they should.

• A comment on Talk: Scott Fraser: Why eyewitnesses get it wrong

  Sep 18 2012: Very nice. But dont forget that eyes adapt and if the kids were playing in such relatively dark, they can see much more than any non-adaptive camera. So the argument is a bit flaky. On top of that if you are familiar with the person....
• +1

  A comment on Talk: Sarah-Jayne Blakemore: The mysterious workings of the adolescent brain

  Sep 18 2012: Structural MRI and its relation to brain function or functional MRI for that matter are nothing but Phrenology equipped with modern technology.
  
  And over simplification is just wrong, nearly everything said in this talk can be argued/debated and most likely would be proven wrong in a scientific crowd. With such situation whats the point of such talks.
  
  We know so little about functional properties of brain and its components that talking about relationship between size of certain brain region and behavior is so bad and absurd that it cant even be termed wrong.
  
  I think people should be more responsible in giving talk on such topics. They only misguide the untrained listeners.

• A comment on Talk: Mina Bissell: Experiments that point to a new understanding of cancer

  Jul 18 2012: Very nice talk. Personally very informative and inspiring. I have been thinking about a theory of cancer myself and there has been one missing link my version. In my theory cancer is not a single cell dysfunction rather it is a tissue dysfunction. So critic said that how a single otherwise healthy cell would know that there is a tissue dysfunction I think now I can safely think of extra-cellular-matrix (the red boundaries that Prof. Bissell showed) as a mode of communication that informs the cells in a tissue about an overall dysfunction located somewhere else.
  Read on for more: http://theoryofcancer.blogspot.com
  I will update the blog post with this link of Dr. Bissell's ideas.
• +1

  A comment on Talk: Pamela Meyer: How to spot a liar

  Oct 14 2011: What about actors, they are trained liars too, I am surprised that they were left of the discussion in this talk.
  In general I think there is so much context to our language, choice of words, body language that such mapping between a certain behavior and lying is very likely to hold as a rule. Only way to spot a lie is to know the truth. This may come as a tautology but this just summarizes the problem of spotting a lie. You will claim something to be a lie only when you know an alternative.

• A comment on Talk: Rajesh Rao: A Rosetta Stone for the Indus script

  Jun 30 2011: When the Science paper on this work came out, I was surprised that it took the linguists studying Indus script such a long time to do the such basic analysis.
  
  Its indeed a good start to study the Indus script, but I remain skeptical of the inference so far. For a start, I doubt the right to left or left to right style of writing can be inferred from the fact that on one side of the stamps the letters are cramped. Think of society that uses the concept of 'stamping' things and they make a mistake on a 'stamp' such that letters are cramped on the right or left side. Its only kids who run into such issues.
  
  Next, I find it a complete stretch that those names sound like ancient Dravidian. Give it to an Iranian or a north Indian or and s/he will find out corresponding names form their own languages. Also, if these are stamps then they should be read from their mirror images.

• A reply on Talk: Sebastian Seung: I am my connectome

  Oct 4 2010: Thank you Sebastian for your reply. I agree with you that knowing the connectome in terms of putative connections among neurons is not sufficient and we need to consider more data about the neurons and synapses to complete the 'connectome' description of the brain. I think connectome data would be very useful in terms of general connectivity rules, once we have that, next challenge will be to solve the dynamics on the connectome.