Elections 2020

 Elections. Pointless fucking turning on of democracy every 4 years. Saving the soul of America, by making us choose between two rapists. One corrupt, foul-mouthed, decadent gangster - baiting all those left behind by elite America and their technocratic choices of outsourcing jobs, all those condemned to a wasteland of opium addiction, incarceration and camo-wearing, automatic-weapon-wielding "freedom" loving "Christians". The other, an appointee of Wall St., anointed to keep any simmering whispers of the most modest redistribution of wealth at bay. Barely managing to emit coherent sentences. And the elite suburbs I live in put out yard-signs that say "Vote" as if to remind me (or themselves?) how we must exercise our choice (the subtext - vote out that uncouth person please - we don't like his style), our freedom. The same kinds of freedom we have no problems denying Syrians. Or Bolivians when it comes to Lithium. Not for them the yard-signs to vote but just an elaborate maze of semi-truths, untruths and impenetrable foreign policy mumbo-jumbo that is too distant, too irrelevant. 

This election is beyond depressing. I am trying to look away and focus on writing code for the moment. Wake me up when it is over and they have decided that they agree on the count and people have gone back home to their lives of quiet desperation.

MSI Wind PC based NAS

This project was long overdue. I wanted the NAS to be unobtrusive and to have low power consumption and noise and I was not really dying for RAID-5. While I do get hugely added performance (check NAS benchmarks) I was not willing to pay the price for it - my uses were mainly for backup and for streaming audio. So it was going to be RAID-1 for me. I also did not see myself needing more than 1 TB if I left video out of the picture. If audio grows at an order of magnitude faster than text / mail etc., video is two orders of magnitude larger.

There were two alternatives - to find a Linux friendly NAS device or to build one. Among the Linux friendly devices (either running Linux or *BSD). Products such as the Synology DS 409 or QNAP TS-419P were examples of products that were feature rich (hot swappability, RAID-5, 6 support, Online RAID Migration etc.) and would have been reasonable products to base the NAS on - but these were in the $ 400 - 500 range. Given my requirements, they were overkill. I wasn't too impressed with the 2-bay NAS products considering that they were more than $ 200 and were not offering much by way of features either.

So, that's when I started considering building my own. Initially, I walked down the path of building a Micro-ATX case, Atom based system with a Mobo / Processor combo like this one - http://www.newegg.com/Product/Product.aspx?Item=N82E16813131396. Soon, came upon the MSI Wind PC and the price and size were very attractive indeed.

One aborted attempt with a motherboard with a BIOS problem that was annoying but then, putting this together was really easy. Got a 16 GB (wondering why I didn't just buy an 8 GB card!) Compact Flash card for the OS, 2GB (yes, it would have been nice to have had two DIMM slots and 4 GB) memory and two 1 TB low power Seagate disks. Used a 3.5" to 5.25" conversion rails to use the other bay for the second drive. Installed Ubuntu 9.04 from a USB disk to the Compact Flash drive and rebooted. Wanted to reduce the willingness of the OS to swap and reduced vm.swappiness to 5 in /etc/sysctl.conf which is set to 60 by default.

Installed mdadm and set up a RAID-1 array with the two disks, LVM on /dev/md0 and the logical volumes on a volume group called logos and presto - I had my entire NAS setup up and running. Rsync'ed my existing non-RAID LVM on the desktop to the NAS and measured 12.8 MB/s - not bad when you compare with the benchmarks.

What can be better?
1. More memory would be nice - could get a performance boost from caching
2. RAID monitoring - need to set up mdadm --monitor to at least get e-mail alers
3. Slightly smaller form factor - considering how far MSI went to design this one, they could have gone a little further and reduced some more white space. As it is, it is better than my original plan of building it in a Micro-ATX tower.
4. Hot swap - which I plan to accomplish later using an external eSATA enclosure. Could even do RAID-5 that way.

As always, Sundar provided me great advice. His NAS is a 100 MB /s + RAID-5 setup with a hardware RAID controller from Areca!

Revival

Coming back to this blog with renewed resolve to journal my life (at least the more interesting aspects of it) more regularly.

His Master's Voice - Stanislaw Lem

After many months back to the blog and that too because of two pages of scribbled lines on a flight to Delhi many months ago that I found while re-arranging the room

The recurrent theme for Lem - Solaris, Fiasco and as I discovered, in "His Master's Voice" of the problems of communication and any meaningful contact with civilizations and cultures different from our own. In HMV, there is a cosmic "letter" that has been received on earth - transmitted by means of a neutrino-beam that can penetrate through almost any kind of matter. The beam (originating from "Canis Major") requires it to be powered by something as powerful as the sun indicating a civilization capable of astral engineering. An impressive array of scientists and humanists assemble to "solve" the puzzle in a desert in an abandoned nuclear facility.

Some random notes - thinking points:

a. Lem goes into satirical commentary at length on how science has sold its conscience to military research and imperialism
b. The protagonist himself (along with another physicist) discovers a unique effect in the neutrino beam - "Frog's Eggs" (consisting of weak nuclear interactions) - which incidentally is a suspension of large molecules (perhaps like primordial life suggesting extra-terrestrial origins of life). The neutrino beam or the letter as the property of increasing the probability of life's molecules to survive. The TX effect is the discover of "fission" at a distance - instantaneous. However, as the pursue this project in a clandestine way, they come to realize that the TX (Tele-eXplosion) effect decreases in precision of where the explosion happens as the intensity of the explosion increases - a manifestation of Heisenberg's uncertainty.
c. One theme is of how culture affects understanding and shapes our world view. There is an interesting piece about how natural languages are cultural and how, for instance, genetic language is acultural. A good fried of the protagonist Rappaport, a Jew, who has escaped Nazi Germany (gruesome description of a scene where there is a slaughter and R escapes).
d. Various explantions are posited by different experts. An interesting view is one involving the oscillating universe - expanding "red phase" engendering life and contracting "blue phase" annihiliating it over billions of years. This theory proposes that the letter is something, perhaps from an earlier civlization, that encoded this life-engendering message into the neutrinos which result from the blue phase's contraction and are emitted by stars in the red phase. A regeneration of life in every new universe, the legacy of a hugely advanced civilization. There are many ideas and explanations in the book - one could just take one of them and write entire fantasy about it.
e. There is a discussion on how Judeo-Christian cultures are driven by a sense of guilt from the original sin. The eastern societies are driven by a sense of shame and therefore cultivate a lot of ritual so that there is a more "perfect" way of being. The point stressed was that empericism and the pursuit of modern science and physics was possible only in the Judeo-Christian traditions. The eastern traditions shunned empiricism. Hmmm...
f. Finally, the protagonist, comes to some understanding of his own - he thinks that somehow, the sender made the "letter" such that only a civilization that had reached a certain stage could fully understand it. He feels that they had also taken pains to ensure that a civilization not at that stage cannot misuse the message as shown by the TX effect and the uncertainty around it. There is in that a benevolence that he perceives in the sender - an acceptance of something good. There is probably some Christian acceptance here?

Indian Agriculture

There was a time when Baba was an authority on Indian agriculture. There was a time when I could have spoken to him about it and mostly listened to his views on "food security" in the country. With only a weekly call bridging us today, I hardly can do that. I read this article by Vandana Shiva http://www.zmag.org/sustainers/content/2006-02/10shiva.cfm which got me reminiscing.

As we continue to trade away our freedoms to pander to an increasingly vocal media elite and as we begin to genuflect at Bush and America's altar, what niggles is thought of what is to follow. The middle class in India has finally awakened to "life and freedom" and now have a very keen sense of their deserving. Not so when our parents were growing up - no not when our "tryst with destiny" was truly made. Only now does the "dil maange more".

Middle class India have well appropriated the media to increasingly focus on the vapid and lifeless paraphernelia of their own secure lives. There is little coverage on widespread protests of Bush's visit to India. No farmer's suicide makes front page news but the bill of fare lavished on Bush by the master chef of the Taj group at banquets held by our obsequious heads of state does. And as middle class India salivates and drools and congratulates itself on having finally arrived on the world scene, 600 million people look askance at a government that came into power defying all predictions of the self same elite media. These 600 million defied the predictions because they did not want CEOs as Chief Ministers and they did not want neo-liberalism or Hindu fascism or what have you. The new bosses had a more "brilliant disguise" and the defences of the 600 million have been broken.

One day, these sallow, defenceless faces will yield no longer. No more will they put in those ballots into boxes to allow self-serving, war-mongering stuffed shirts to control their destiny. Proud ploughs and furrows in hand, they will take back what is theirs and if they spill some blood of lily-livered city folk with their newspapers and fashions and web-sites and sports idols - well, that would truly be "collateral damage" - of little or no significance.

Cognitivism and theories of the mind

Cognitivist thinking has come into a lot of prominence of late and not least because of some popular science literature, particularly from Steven Pinker and Daniel Dennett. Noam Chomsky made a strong departure from the conditioning based learning paradigm when he proposed that humans are born with the "Language Instinct". As neural sciences progress, more evidence is mounting that favours theories which suggest that the mind is not really a "blank slate" or tabula rasa that John Locke thought it to be and that influenced many generations of philosophers, psychologists and mind theorists.

Over the last few years, I read a few books by Pinker - largely because he made theories of how the mind works accessible to me. It is very clear that like Dawkins, he has embraced the route of popular science. The trouble with popular science is that matters of such gravity are presented without scientific rigour. In The Blank Slate, Pinker takes up the cause of cognitivism and critiques years of philosophical and political motivations for the "denial of human nature". That the brain is completely algorithmic and like a digital computer may be the absurd extrapolation of the cognitivist view point. However, many, including Roger Penrose, the mathematician-physicist and John Searle, the philosopher have written extensively about why the algorithmic / computational theory of the mind may not be feasible. Some very interesting debate is alive in this area.

Be that as it may, and the debate is far from settled, no matter what Pinker believes, it is still true that morality and ethics cannot be the underpinnings of physical science just as empirical physics or biology cannot be the basis for morality and ethics. I think that his point is well illustrated by Dawkins. He points to the philosophy of David Hume and of his separation of the realm of morality from that of scientific enquiry.

I think that the understanding of the working of the brain (there is a very interesting podcast that I chanced upon of a radio-show called "All in the Mind" from Australian Broadcasting Corp. - a beautiful exploration of such issues) is going to be the next biggest challenge for science - and this includes not just neural scientists and psychologists but may take the thinking of mathematicians and physicists to unravel. It is certainly an area that has a lot of debates, strong opinions, a smattering of data and very few profound theories. It is of tremendous interest for computer scientists too because the von Nuemann model of computation that has been the bedrock for all theory of computation may well yield place of privilege to newer models of computation.

The intent of this slightly pretentious post is to really annotate some bookmarks in this area:

John Searle's APA Presidential Address: "Is the Brain a Digital Computer" in which he continues from his Chinese Room Argument (1980) to argue against the mind as a computer program. He posits four difficulties with the thesis of the brain being a digital computer -
- Syntax is not intrinsic to physics
- The Homunculus Fallacy is Endemic to Cognitivism
- Syntax has to Causal Powers
- The Brain Does Not Do Information Processing
Read on...

A very interesting essay by the Philosophy professor D. C. Mathur on the similar positions by three quite unrelated historical figures: The historical Buddha (Gotama), Hume, and James on the self: Comparisons and evaluations
This may not be terribly germane to a theory of the mind as much as it is to a theory of ethics that dissociates itself from the reductionist Western science

Consciousness is the biggest mysery in the already complex mystery of the functioning of the mind. An interesting school of theorists belive in a quantum theory of the mind. This is of course very speculative at this point. There are many links at the following Wikipedia link.

"All in the Mind", an absolutely wonderful radio program with podcast and transcripts available.
Finally, Steven Pinker's home on the web.

It is a gripping story of how the plot unravels. A bit like "Sophie's World" and the little subplot in it.

The "Drawing Room Twitter" of Vikram Seth

Vikram Seth is prodigious with words. He hasn't looked back since declaring his genius with "Golden Gate", a book that I read with utter fascination many years ago. Since then, I read "All you who sleep tonight" and "An Equal Music" - both books mesmerized me with their charm. Recently, an interview appeared of him in the Outlook magazine which has been the subject of much discussion on the internet. That is what set me off thinking about Seth's writing. That, and my recent re-reading of Rushdie's "Imaginary Homelands".

You see, the trouble with genius is that everyone starts to expect more from you. So with Seth, we try to look beyond his silver phrases and the veneer of urbane poetry and prose. And what you find, unfortunately is "drawing room twitter" and an unwillingness to engage with any issue - political, psychological or otherwise. Just a lot of sophisticated badinage!

Paul Graham - stick to writing about technology

I love Paul Graham's writing on Technology but he should stick to that. I love his writing about technology because I feel that he has great insight into the pyschology of programming. For instance, in Hackers and Painters, he argues that hacking (read great programming) is more an art than a science - something that is learnt through practice and by studying the masters.
"The fact that hackers learn to hack by doing it is another sign of how different hacking is from the sciences. Scientists don't learn science by doing it, but by doing labs and problem sets. Scientists start out doing work that's perfect, in the sense that they're just trying to reproduce work someone else has already done for them. Eventually, they get to the point where they can do original work. Whereas hackers, from the start, are doing original work; it's just very bad. So hackers start original, and get good, and scientists start good, and get original."
Writing about why Lisp is so great, Paul compares Lisp's popularity with respect to other languages (read Java) to the popularity of Jane Austen compared to the more commonly read John Grisham! Lisp is indeed difficult to learn as a first language but provides the brevity of expression and richness of metaphor that is unique. Ask any emacs afficionado.

So, why am I disappointed with Paul Graham, in his latest essay writes about Inequality and Risk? If I were to take Graham's conclusions at the end of this essay to their bitter end, I would be led to believe that technology and "progresss" are altars on which every other human endeavour must in the end bow. He argues that societies must reward risk for it is only through risk-taking "startups" that we renew ourselves. The potential reward for the founder of a startup is the possibility of acquiring a huge amount of wealth and that is the motivation for risk taking. If we make the assimilation of wealth not as desirable by very high tax rates, for instance, we would remove that motivation. Therefore, the desire for economic equality in and of itself takes away the entrepreneurial spirit which is so precious. And then, as if to assuage his conscience, Paul Graham goes on to say this:
"We don't need to prevent people from being rich if we can prevent wealth from translating into power."
So, we should be attacking the linkages between wealth and power and not wealth itself.


Graham parrots the very Western capitalistic mantra of competition and risk taking being values to be cherished for a healthy society. His vehicle for realization of the American dream are startups over which he obsesses. And the American dream of wealth is ok as long as it does not lead to corrruption and exploitation of power. Because wealth is only about security according to Graham. Alas, that is just too simplistic reasoning. Wealth allows you to participate in systems that are innately unequal and which foster inequality. Wealth and power are inextricably linked and merely logging transactions will not alter an economic system that rewards capital over labour. I do not believe that technological innovation does not happen in the absence of huge economic rewards. On the contrary, most technological innovators have rarely turned wealthy because of their inventions. Even in the field of computers, examples abound. The elegance of Unix invented in Bell Labs did not lead to much wealth for Messrs. Ritchie, Kernighan etc. Linus, made little from Linux. As for your own favourite - Lisp's inventor John McCarthy despite his right-wing views himself made little money from the idea of Lisp which was conceived in academia and has following mostly in academia. Wealth has usually been obtained, not by technical innovation alone, but by the ability to exploit markets using technology. So, I am not so worried about technical innovation dying because there is no wealth incentive. As long as there continue to remain institutions where the pursuit of knowledge and the respect of peers ranks above accumulation of wealth, there will be innovation. And with the equalizing effect of the internet, such institutions have only proliferated.

I have another objection to Graham's essay - of placing technological growth as a desirable in and of itself. As a race, we have constantly struggled in our sociological and psychological frameworks to keep pace with technology. This has increased in the last hundred years. We also struggle to keep pace with our moral and ethical compasses to deal with issues such as privacy that are under constant threat from technological progress. Without sounding too much like a Luddite (and that would be too hypocritcal considering that I earn my bread through technology), I think that it is time that we put some other things above technological progress. That we had more debates in every sphere about the ethical implications of RFID or of genetic engineering without relegating and labeling all such arguments as nihilistic or anti-development. Bill Joy's (cofounder of Sun) article in Wired (August 2000) - "Why the future doesn't need us" is worth reading as an alternative view point of another technologist.

The other two...

The other two documentaries - one called Born Into Brothels directed by photojournalist Zana Brinsky is about the children of sex workers in Sonagachi, Calcutta's famous red light district. Zana has been working in Sonagachi for over 3 years and this movie is a culmination of her efforts to teach photography to a group of about ten children. She also valiantly campaigns to remove these children from their situation and enroll them in schools, a process which involves not only advocating their cause with school authorities (who would much rather not have prostitute's children spoil their reputation amongst their largely unctuous middle-class students and families) but also spending hours discussing and convincing the parents or guardians of these children that education in an environment removed from their current one is indeed a desirable option for them. In the end, there are a few things that stand out from this film in my mind. Firstly, of course, that the middle-class, who are the primary clientele for the sex workers, choose to completely shut them out and deny their existence when it comes to admission to schools or getting identity cards for the children. The second thing was that despite Zana's best efforts and intentions and after having secured seats for many of the children against overwhelming odds, two or three years down the line, all these children are out of these residential schools and back where they were. Some left of their own volition and others were pulled out by their guardians. It is not only about placing these children there but that there is a continuous process of reinforcement that is required and that requires a movement beyond the compassion of one individual, perhaps. Finally, the pictures that these children take. There is spontaneity. There is unbridled enthusiasm when Zana takes them out for shooting trips to the zoo or to Digha. Aveejeet, a sensitive young boy, speaks perceptively about photographs. He is also the rebelious one. Finally, he is chosen to go to an international exhibition of child photographers. There is some hope there - some way to cut through the barriers of conventional education and learning to achieve something substantial and to find some recognition through a modern and yet more easily acccessible art form.

Lastly, Michael Moore's Roger and Me. A soulful account of the loss of jobs in Flint, Michigan when GM decided to move manufacturing to "harness" Mexico's cheap 70 cents an hour. Michael Moore is like a terrier. He plugs away at something with a maniacal obsession when he sets his mind to it. In this case, he wants to meet the GM CEO, Roger Smith and confront him with the decisions that he has made to move jobs out of Flint. He follows him across the country and his thwarted at every step - by "security" at the GM headquarters, by curt staff at "Members Only" exclusive clubs that Roger Smith is a member of. He even poses as a GM stock holder and enters a general body meeting of stock holders but alas, even as he stands up to ask a question they end the meeting. Moore has an unequivocal stance as always - that people matter over profits, that GM and Roger Smith owe an explanation to the city of Flint and its residents for leaving them high and dry in this manner.

Three Documentaries, First Mondovino

Saw three documentary films in the last few days. Mondovino is directed by Jonathan Nossiter, an American film maker who lives in France who is also a trained sommelier and a wine critic. At one level, this is a wistful look at the changing world of wine production. In a series of interviews with some small artisinal wine makers and some large manufacturers including the Mondavi brothers in Napa, California, Nossiter tells the tale of how most wines are now mass-produced to satisfy the median taste of consumer-researched tastes, stamped with high ratings from acclaimed wine critics. Two people represent this corporatized, mass-marketed, mass-produced world in the film. First, Robert Parker, one of the most influential wine critics around. Second, Michel Rolland, a wine consultant with his ludicrous mantra of "micro-oxygenation" and aging in oak barrels.
There is no commentary from Nossiter - he lets the interviews present his point of view. Yet, the subsersive movement of the camera, the choice of questions, the juxtaposition of shots all tell the sad story of how artisinal wine makers are increasingly being swept aside by forces much larger and more fearsome more lucidly and with better effect than anything that a direct commentary could. For this is the story, not merely of wine but of all things small and beautiful that are being swept aside in this furious rat race of corporate profits, globalization and mass-produced life style. You don't have to have to be a wine lover but it sure helps.
Nossiter's stance against these forces of globalization is echoed most evocatively in the voices of a few wine makers from France (Bordeaux) and Italy most notably Hubert de Monteille, his son Etienne and his daughter Alix. These wine makers speak passionately about "terrior" (land) and about letting the wine speak for itself rather than the vanilla oakiness to satisfy Robert Parker's pallette. New York-based importer Neal Rosenthal is another who speaks against globalization. And there is the amateur-like swinging of the camera that speaks for Nossiter - Robert Parker's flatulent dogs, the poignant voiceless Mexican worker at the large estate of the Napa wine growers the Staglins, the beam of pride in Hubert when his daughter announces that she will leave the winery she is working at because she finds herself opposed to the policy of marketing the same wine under different labels.