10 GBytes/s end-to-end between two processing units is quite unlikely, given high-end processors typically employed in supercomputers (like UltraSparc, Alpha,...). The memory bandwidth of today's CPUs is not so gigantic, after all.
The bisection bandwidth might be 10 GByte/s, though. And _that_ is hardly suprising - you can get Ethernet switches with 30Gbit/s backbone bandwidth for less then $20k these days...
The main difference between GBit Ethernet and tightly coupled network interfaces is the latency, i.e. the time it takes to send a message accross. Fractions of a microsecond for supercomputers, tens of microseconds for Ethernet. </soapbox>
It might be a funny idea to force, say, my local phone company to drop my address information, which I trivially own, from _all_ their databases (phone book, billing databases, etc).
Can I then sue them for copyright infringement if they still send me a bill to my address ??
Also, some of my custer information is probably in databases of multiple companies and agencies (DMV, GTE,...). Will they start sueing each other over my address. Oh joy!
It is true that pharmaceutical corporations need to be profitable in order to fund research. BUT:
not all research in that field is performed by said corporations. Last time I checked, there was also the public sector (universities and such).
There are a lot of companies offering gererica, i.e. identical drugs to the original, without the brand name. For example, ASS(ratiopharm) vs. Aspirin (Bayer). "Brand" corporations like Bayer, Merk, etc still make a quite decent living despite the low-cost competition, so I not all too concerned. Finally, I am wondering if the software industry can serve as a role model for the chemical industry: Although OpenSource is going quite strong these days, few are concerned that OpenSource will throw the world back in the the information dark age because it drives all commercial software vendors out of business.
But my thinking may be flawed, and I don't understand how the pharmaceutical market works. Anyone can give some comment on that? For instance: For a standard new drug, how much is development cost vs. total profit? How does the development cost break up (research, testing, marketing, lobbying,...)
Human genes are not the only one that are dangerous to allow patenting. Financial times has an interesting article on how companies holding patents for gene-manipulated food can get a stranglehold on all of us. Read it up at here.
IMHO, this issue is as least as scary as M$, especially given the history of companies like DuPont (anyone remembers Jack Herer's book ("the emperor wears no clothes"?). Clearly, companies should in general be protected to reap the rewards of their research and work. However, this protection needs to be balanced with the protection of society and (yeah, corny), humankind.
I don't see a lot of serious effort to address this issue politically. IMHO, this is the time to act.
Slashdot Mirror
10 GBytes/s end-to-end between two processing units is quite unlikely, given high-end processors typically employed in supercomputers (like UltraSparc, Alpha,
The bisection bandwidth might be 10 GByte/s, though. And _that_ is hardly suprising - you can get Ethernet switches with 30Gbit/s backbone bandwidth for less then $20k these days
The main difference between GBit Ethernet and tightly coupled network interfaces is the latency, i.e. the time it takes to send a message accross. Fractions of a microsecond for supercomputers, tens of microseconds for Ethernet.
</soapbox>
It might be a funny idea to force, say, my local phone company to drop my address information, which I trivially own, from _all_ their databases (phone book, billing databases, etc).
...). Will they start sueing each other over my address. Oh joy!
Can I then sue them for copyright infringement if they still send me a bill to my address ??
Also, some of my custer information is probably in databases of multiple companies and agencies (DMV, GTE,
It is true that pharmaceutical corporations need to be profitable in order to fund research. BUT:
not all research in that field is performed by said corporations. Last time I checked, there was also the public sector (universities and such).
There are a lot of companies offering gererica, i.e. identical drugs to the original, without the brand name. For example, ASS(ratiopharm) vs. Aspirin (Bayer). "Brand" corporations like Bayer, Merk, etc still make a quite decent living despite the low-cost competition, so I not all too concerned. Finally, I am wondering if the software industry can serve as a role model for the chemical industry: Although OpenSource is going quite strong these days, few are concerned that OpenSource will throw the world back in the the information dark age because it drives all commercial software vendors out of business.
But my thinking may be flawed, and I don't understand how the pharmaceutical market works. Anyone can give some comment on that? For instance: For a standard new drug, how much is development cost vs. total profit? How does the development cost break up (research, testing, marketing, lobbying, ...)
IMHO, this issue is as least as scary as M$, especially given the history of companies like DuPont (anyone remembers Jack Herer's book ("the emperor wears no clothes"?). Clearly, companies should in general be protected to reap the rewards of their research and work. However, this protection needs to be balanced with the protection of society and (yeah, corny), humankind.
I don't see a lot of serious effort to address this issue politically. IMHO, this is the time to act.