Date: Sun, 14 Oct 2001 00:21:01 -0700 From: Terry Lambert <tlambert2@mindspring.com> To: Rahul Siddharthan <rsidd@physics.iisc.ernet.in> Cc: chat@freebsd.org Subject: Re: Returned mail: see transcript for details Message-ID: <3BC93CDD.484C9806@mindspring.com> References: <20011013010553.A343@lpt.ens.fr>
next in thread | previous in thread | raw e-mail | index | archive | help
Rahul Siddharthan wrote: > I know a fair bit about the protein folding problem, including the > fact that it's not as simple as you think (or as many naive physicists > think). I have one or two tricks up my sleeve, and I know other tricks that others have up their sleeves. If you really are going after the problem, I urge you to pool your information with myself or another Senior Associate of the Foresight Institute. > There will certainly be money in it, but I doubt the money > will go to IBM simply for supplying the computer, any more than money > for solving existing computational problems goes to the computer > manufacturers. People will buy the equipment from the manufacturer who first builds it, in preference to other equipment. It's how people who make business decisions think. > Moreover, the protein folding problem is much more than a > computational problem. In its stripped down form, it is often > stated as "mapping a protein sequence to its fold", and then it is > imagined that doing a molecular dynamics simulation of the protein > chain will be able to predict its fold, if carried out long enough. It's generally accepted that protein folding has an intermediate step, in which there is a "scaffolding" that is erected. One approach that's frequently ignored is the software engineering approach: We already have these little factories called "mitochondria", and we already program them in E. Coli to manufature proteins like human insulain, without really understanding how that happens. Modern software engineering, sadly, seems to include a lot of people who don't understand the code that their compiler is going to output for a given source file. These people treat the compiler as if it were a black box, and as a result, they build code that's not nearly efficient as it could be, if they understood their tools. On the other hand, they can still do useful work, treating it as a black box. > > I think you are mixing passenger railway with freight railway, > > Well, yes, when referring to public services which are not > successfully privatised, I was talking about passenger railway. > > > and concluding because people don't like passenger rail service > > in the U.S. more than they like getting there quickly by plane, > > or having daily control over their own schedules (how many dates > > would you be able to have on the spur of the moment, if you were > > required to take a relatively slow train home before you could > > get in your car and drive to the date location?) > > Train travel is, of course, suitable only to some ranges of distances. > One would not expect to travel from New York to Los Angeles by train. > However, Los Angeles to San Francisco should certainly be possible, > and a train travel between two such major cities separated by such a > difference in France (say, Paris to Marseille) would take 3 or 4 > hours, and there are several trains a day. For LA-SF, you can easily > look at the amtrak site and find you can't do it without changing, > taking bus connections, etc and spending over 8 hours. (I know that > example because I needed to do that recently; but I've been told that > it's the same, or worse, all over the US. It's a bit better on the > east coast in the Washington-New York-Boston zone, I admit; but still > doesn't approach the efficiency of the SNCF in France. (Since the > faster Paris-Marseille track opened a few months ago, air tickets on > that route are going unsold, while the train, which runs several times > a day, is booked out well in advance.) Frankly, I'd rather take a convertable down the Pacific Coast Highway; the journey would be infinitely more enjoyable, not include having to work to get on and off, and not run the risk of some moron throwing Antrax into the environmental controls, since a train is a nice enclosed tube with a lot of people to efficiently attack all at once that way. My risk is a hell of a lot less in my car with only one or two passengers: I'm much less of a target that way anyway. [ ... trains as a means of transportation ... ] The only way you are going to get people to take public transportation in the U.S. is if you make it free, and make private transportation prohibitively expensive. Even then, you will only success in clearing the highways of all but the economic upper classes. Right now, trains and other public transportation in the U.S. is predominantly limited to the economically low end (the bottom end can't afford it, because we like to believe we'[d like people to take it, but in fact, we don't want to make it free, because if we did, it'd be overloaded). > > It's not surprising to me that FedEx and DHL, whose main > > claim to market share in India is that they are able to send > > and deliver internationally, would be more expensive. > > How about France, where also they are more expensive? They are U.S. companies; given their protectionist tariffs on farm produce, I'd expect French companies to end up being cheaper than foreign competitors. > Besides, I was referring to delivery prices within India > (strictly speaking, I was referring to a francisee of FedEx > called Blue Dart, and a franchisee of DHL whose name I forget). India is even more protectionist than France or Japan. They also have a reputation for making it very difficult to start or maintain a business: graft and corruption being rampant, with bribery often being necesary to obtain business licenses, and with waits in excess of 3 months common; in the U.S., it's around two weeks to obtain a license, and in Singapore, you can be up and selling your wares in a single day. It's no wonder that a foreign owned service company would find themselves disadvantaged in such an environment. > > You're wrong. If the postal system were not a state monopoly > > in most countries, it would be easy to compete on purely > > economic terms, using more modern automation than that used > > by them. > > The cost isn't the "automation" in sorting and so on; it's in delivery > at the doorstep of the recipient. Postal departments everywhere have > a reach which extends to the remotest villages, because governments > have supplied such links as parts of basic infrastructure. Couriers > will not find it cost-effective to make door-to-door delivery of > ordinary letters at such rates (40 cents in France, 4 cents in India, > whatever). FedEx has regional centers; initially, before their volume became so large, they sent all the planes to Nashville, Tennesee, sorted them all, and sent them back out. This approach is orders of magnitude more efficient than that used by the U.S.P.S., even if sorting is the only thing that changes. Realize also that the U.S.P.S. is unionized, and thus has a vested interest in resisting automation in order to protect jobs, even if there is a higher cost. > In fact I know the Indian rates are subsidised and not > profitable; the postal department tries to recover the cost from > things like courier services. For the same reason, private > transport operators tend not to serve remote low-populated areas, > and it is often left to government-run transport to serve those > markets; these routes are not profitable for the government either, > but they cross-subsidise them from other routes or other income > sources. Regulations can require service for rural areas in order to permit service in non-rural areas. This is how the U.S. telephone system and road system operate -- as well as the U.S. power grid, etc.. > > This seems to be an issue that's really a problem with the > > WTO, and with India's system, in particular. The U.S. > > patent office actually worked: it rejected the patent. > > For turmeric, and after representation from India. Not for neem > (margosa, I think, in English) and other herbal products, so far. > Those patents still exist. People can use those herbs without the patent. You can't patent naturally occurring materials: you can only patent alternate manufacturing or purification processes for the active ingredients. [ ... abuses of the U.S. patent system ... ] > Sure. The point is it's rather frequent. India is not the only > country affected. And you claimed it's not possible to do this, > which is not true. It's not possible to _successfully_ do this. The Turmeric (Tumeric) patent case proves that. > > It's ridiculous in that countries do not have to overthrow > > such patents, not because of the process involved in doing > > that. > > They most certainly do have to overthrow them, at some point, if they > want an open market for export of these products, which they do rather > actively. Otherwise they'd be violating a patent. Perhaps you're > arguing that they can ignore them and pretend the patent doesn't > exist, and the multinational who got the patent won't sue. I don't > see that happening, and it's probably cheaper to overthrow the thing > to begin with than to sit and wait for a patent-infringement lawsuit > and then tackle that. That's what I'm saying, and that's what I'm claiming India in fact did, using U.S. public opinion as a shield. Yes, this won't work every time, but it worked that time. > > In any case, I was referring to the financial interest of the > > countries, like the U.S. and India, which stand to make a > > much larger amount of money from AIDS treatments than they > > would, were they to actually cure the disease. > > I can't really say about the US, but India and other developing > countries are certainly not making any money from AIDS treatments. You're saying that the drugs are being sold at cost? The news stories from non-U.S. sources (Reuter's, London Times, and D.W.T.V.) which I saw at the time indicated that they were merely undercutting prices, and taking a "minimal profit" -- not a break-even or a loss. Following U.S. public outcry, U.S. companies are now doing the same: taking a minimal profit, and grumbling that the market isn't nearly as profitable as it used to be. > > India did not approach AIDS drug shipments to Africa as the > > purely humanitarian endeavor that they claimed: > > Ah now you're referring to a specific Bombay-based pharmaceutical > company, Cipla, as "India". I could equally well refer to Monsanto > as "The US". That's just silly. By "India" one can mean, most > restrictively, the Indian government, but preferably a broader > index like the GDP. To claim that India would make a gain, one > would have to show that Cipla's gains were offset by the huge > costs of government-sponsored AIDS treatment within India itself. You might as well make the association; the U.S. is not being given credit for humanitarian aid spent in treating AIDS and other ills outside the U.S, they are only held accountable for what U.S. companies do. It's only fair that other nations be held to the same standard you hold the U.S. to. -- Terry To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-chat" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3BC93CDD.484C9806>