Ubuntu improves the desktop presentation of Debian. How is that useful for Sun in the server market? Ubuntu also gives you a Debian server with time-based stable-releases (so, the need for backports is minimized). Actually, I am running Breezy even on my servers nowadays, because there is some server stuff that I want that is not on Woody... and I did not want to wait 'till 6.12 (etch) for something I had in a nice shape in 5.10 (breezy) []s
alien works for me 95% of the cases. works for Oracle, for instance... unless I have strict time constraints, instead of./configure && make && sudo checkinstall make install I take my time to do dh_make, edit debian/*, and fakeroot debian/rules binary.
The viewers (the suckers that pay for the thing on cable telly) will be the ones subdisizing the price for larger CCTV coverage!!! This is really a smart coup.
that's my take (and I was a paralegal in a DA's office for two years, and I researched for prosecuting criminally some cases of copyright infringement):
* a binary is EXACTLY in the same "copyright state" as the sum of the lines of source code that are included on it *
what does that mean? suppose the following program was non-trivial enough to be eligible to copyright protection (it is NOT):
#include <iostream>
using namespace std; int main(int, char**) {
cout << "Hi, Humberto!\n";
return 0; }
when I compile this (w/o --static on the command line), I get a binary that is more or less this way:
** binary executable header ** ** entries for the fixups of "__init_all_libraries", "std::cout", and "std::ostream& std::ostream::operator<<(const char *)", for resolution at load-time ** ** the machine code for main, like this pseudo-assembly:
ALL of those parts are translated automatically, and without any intellectual involvement from a person, from the information in hello.cc (somewhat/influenced/ by the contents of iostream): this binary is not a derivative work, it _is_ hello.cc for all copyright purposes. It is IMHO _not_ a derivative work on a particular iostream implementation, nor on the compiler. Especially because, the way it is written, it could be compiled by any ANSI-compliant C++ compiler and it could be using any of multiple ANSI-compliant STL implementations.
The fact that it was (for instance) compiled with g++ could only affect its license if g++'s license stated "all the binaries generated with g++ must be licensed with such and such terms otherwise g++ license is revoked retroactively" etc etc. And it does NOT. And more -- the GPL states clearly in its text that it does not want to control USAGE of the software, but only copying and creation of derivative works (it even states "derivative works under copyright law" on its most confusing paragraph).
If they are calling it / advertising it as bandwidth, they oughta offer the bandwidth to those that buy it for the bandwidth.
If they say: "buy our 24/7 7 _megabits_ _per_ _second_ for US$ 20", they MUST NOT complain if I want to download four torrents 24/7, saturating my 7Mbps link at all times. If they complain, they are advertising falsely. Period.
Tiered services,/per//se/, are not evil: in a sense, I pay tiered when I pay $35/month for my fat pipe, while a dialup user pays $5/month. The evil is in false advertising, and, from the moment you use a _bandwidth_ measure and an _availability_ measure to advertise, you better deliver.
The whole "kick the user out" or "throttle the user" will not fly with me, either. You promised me -- on TV no less -- you must deliver or PAY ME BACK everything I already paid, plus damages.
the GPL is pretty clear that it requires that GPL applications can't link to non-GPL libraries
No, it does not. Really. All that can be construed from the text of the GPL is that derivative works, if distributed, must be distributed under the terms of the GPL. In the case of the NVidia drivers, only the "shim" is a derivative work of the kernel, and it _is_ distributed under the terms of the GPL. The binary part is _not_ a derivative work of the kernel -- it's the same part that's used by Windows, Solaris, drivers etc.
Main reason is probably that less drunks are the same place at the same time since they go home over a 2h period instead of a 5m period... AND that the drunks are not pissed off because they are FORCED to go home!!:-) (Down here, when the pub's owner wants to send the hardcore drunks out, he starts cleaning the place -- which occasionally involves flooding the floor with soapy water and moping it, and yes, I had my fet wet this way a lot of times...)
Here in the U.S., ethanol can't compete on its own merits. The main thing that is driving its use as a main fuel source (e.g., E85) are the huge government subsidies. I would suspect that besides the obvious market inefficiencies (and what happens when it is so widespread and still requires subsidies?) that this new use may cost us in other ways. If we used ethanol purely as a fuel additive then very high octane gasolines could become commonplace nationwide. This would allow for higher efficiency engines because turbos would be able to use more boost and engines could have higher compression ratios. Unfortunately something like that requires long term thinking and people and governments generally aren't into that.
I forgot to mention that our gasoline is already 20% ethanol (and our ethanol contains methanol and gasoline so you can't [ou shouldn't] drink it)... The US have another disadvantage: corn ethanol is energy-inefficient and expensive. I think the most energy-efficient/cheaper ethanol is our sugarcane ethanol: but we *do* have a problem when the market is hot for sugar (as it's the case right now) (*)
(*) two years ago, the price difference between gas and alcohol was up to 60% (alcohol being 0.4 times the price of gasoline per liter) because the market was not hot for sugar.
Except that, with oil at current prices, it's Waay cheaper to fill your tank with ethanol than to fill it with gas.
But, seriously, though: the same car, on ethanol, makes 10-20% worse mileage than with gas. Down here we have "flex-power" (ethanol/gasoline flexible fuel system) cars, and if a car gets 12km/l(30mpg, 8l/100km) on gas, it usually will get 10+ km/l (25mpg, 10l/100km) on ethanol. Currently, in my town, the pump price for alcohol is about R$ 2,10/l (US$ 3.85/gallon) and the pump price for gas, R$ 2,50/l (US$ 4.59/gallon), which is a 19% difference.
IOW: renewable and non-renewable fuels break even (with a slight advantage for ethanol) on mileage per dollar.
On the performance side, on ethanol cars tend to have a higher final speed than on gas, but they have some 5-10% less torque.
your answer makes sense, even if you are completely wrong:-)
For example, let's say that the filesystem queues up too many requests. If you restart the SCSI driver and resend the requests, the queue will probably be even longer by now, causing it to crash again. Your system is now in an infinite loop of restarting the SCSI driver, while you hope that it's not corrupting the disk every time it gets automatically reset.
Or more likely, the filesystem sent a bunch of requests to the SCSI driver, assuming they would be fulfilled. When the SCSI driver crashes, the filesystem has no idea which requests were completed or not, and has not kept track of which ones were outstanding. Now your filesystem is consistent in-memory but not on-disk. When you restart the SCSI driver it will continue corrupting the filesystem even further. The only way around this is to restart the filesystem on that disk too (I hope it was journaled), meaning that it would probably have to invalidate any open files. If this happened while writing or reading your VM paging file, the system is almost completely hosed.
Scenario #1: when the disks went whacko, I would expect (besides trying again) that the server sends a "cancel" message to every pending-i/o process, and let the apps recover.
Scenario #2: the error is in the "assuming they would be fullfilled". If the SCSI says "I fullfilled your batch", then it's fullfilled, not before.
Scenario #3: disk error on swap partition: the affected processes are killed.
That being said, I agree that having the option of installing a real sun JRE/JDK with apt-get or rpm WITHOUT dropping to the command line or hacking configuration would be a BIG bonus and widen the audience beyond CS students and linux gurus.
like, say, K-MENU [click] System [click] Adept [click] [wait half a second] [type password] [enter] [wait 5 seconds] hmmm... [click on search box] j2 [wait a second] [click on j2re entry on the grid] Install package [click] Apply changes [click] [wait 30 seconds] ???
When I re-watch "Stand by Me" (or read King's Four Seasons) I feel moved. "Braveheart", I never thought very much of it; but almost every Stephen King-based story strikes me profoundly.
Estatistically speaking, and in the long run (meaning tending toward infinity) the "profiles" of people that browse site A and site B would fall in more or less the same -- meaning it's not the overcounting or the undercounting that cancel each other -- is the the over/undercounting for site A is more or less the same over/undercounting for site B.
Therefore, if site A has 10 million IPs logged yesterday and site B has 5 million IPs logged yesterday, it's fair to say that site A has double the traffic of site B.
That is, unless you can prove that the user "profiles" (PPP, PPPoE, people that use lots of different computers versus proxies, NATted people etc) are wildly different between site A and site B -- which can occur e.g. in pr0n sites (which would be blocked by most corporate proxies but not home computers etc), the number of distinct visiting IPs is a good measure of site traffic.
Slashdot Mirror
I Malaysia is blocking my IP, I use an open proxy in, say, Vietnam, and access the "page creation" site from my house. Simple as that.
Ubuntu improves the desktop presentation of Debian. How is that useful for Sun in the server market?
Ubuntu also gives you a Debian server with time-based stable-releases (so, the need for backports is minimized). Actually, I am running Breezy even on my servers nowadays, because there is some server stuff that I want that is not on Woody... and I did not want to wait 'till 6.12 (etch) for something I had in a nice shape in 5.10 (breezy)
[]s
alien works for me 95% of the cases. works for Oracle, for instance... unless I have strict time constraints, instead of ./configure && make && sudo checkinstall make install I take my time to do dh_make, edit debian/*, and fakeroot debian/rules binary.
Alternative 1 (dirty, two steps):
:-)
$ sudo apt-get install rpm
$ sudo rpm --force-all -ivh PACKAGE.rpm
Alternative 2 (cleaner, four steps):
$ sudo apt-get install rpm alien fakeroot
$ fakeroot alien PACKAGE.rpm
$ sudo dpkg -i package.deb
$ sudo apt-get -f install # will install any dependencies
Alternative 3 (suppose multiverse is in sources.list)
$ sudo apt-get install package ## it is probably there
I've never met anyone that *didn't* want to run OS X.
Seriously, I don't want to run OSX, never did. Cross my heart.
The viewers (the suckers that pay for the thing on cable telly) will be the ones subdisizing the price for larger CCTV coverage!!! This is really a smart coup.
that's my take (and I was a paralegal in a DA's office for two years, and I researched for prosecuting criminally some cases of copyright infringement):
/influenced/ by the contents of iostream): this binary is not a derivative work, it _is_ hello.cc for all copyright purposes. It is IMHO _not_ a derivative work on a particular iostream implementation, nor on the compiler. Especially because, the way it is written, it could be compiled by any ANSI-compliant C++ compiler and it could be using any of multiple ANSI-compliant STL implementations.
* a binary is EXACTLY in the same "copyright state" as the sum of the lines of source code that are included on it *
what does that mean? suppose the following program was non-trivial enough to be eligible to copyright protection (it is NOT):
#include <iostream>
using namespace std;
int main(int, char**) {
cout << "Hi, Humberto!\n";
return 0;
}
when I compile this (w/o --static on the command line), I get a binary that is more or less this way:
** binary executable header **
** entries for the fixups of "__init_all_libraries", "std::cout", and "std::ostream& std::ostream::operator<<(const char *)", for resolution at load-time **
** the machine code for main, like this pseudo-assembly:
_1: bytes "Hi, Humberto", #10
_main: load value r1, "std::cout"
load immediate address r2, _1
call "std::ostream& std::ostream::operator<<(const char *)"
load immediate value r1, 0
return
**
ALL of those parts are translated automatically, and without any intellectual involvement from a person, from the information in hello.cc (somewhat
The fact that it was (for instance) compiled with g++ could only affect its license if g++'s license stated "all the binaries generated with g++ must be licensed with such and such terms otherwise g++ license is revoked retroactively" etc etc. And it does NOT. And more -- the GPL states clearly in its text that it does not want to control USAGE of the software, but only copying and creation of derivative works (it even states "derivative works under copyright law" on its most confusing paragraph).
Sue for false advertising.
If they are calling it / advertising it as bandwidth, they oughta offer the bandwidth to those that buy it for the bandwidth.
/per/ /se/, are not evil: in a sense, I pay tiered when I pay $35/month for my fat pipe, while a dialup user pays $5/month. The evil is in false advertising, and, from the moment you use a _bandwidth_ measure and an _availability_ measure to advertise, you better deliver.
If they say: "buy our 24/7 7 _megabits_ _per_ _second_ for US$ 20", they MUST NOT complain if I want to download four torrents 24/7, saturating my 7Mbps link at all times. If they complain, they are advertising falsely. Period.
Tiered services,
The whole "kick the user out" or "throttle the user" will not fly with me, either. You promised me -- on TV no less -- you must deliver or PAY ME BACK everything I already paid, plus damages.
the GPL is pretty clear that it requires that GPL applications can't link to non-GPL libraries
No, it does not. Really. All that can be construed from the text of the GPL is that derivative works, if distributed, must be distributed under the terms of the GPL. In the case of the NVidia drivers, only the "shim" is a derivative work of the kernel, and it _is_ distributed under the terms of the GPL. The binary part is _not_ a derivative work of the kernel -- it's the same part that's used by Windows, Solaris, drivers etc.
oh, I should really use this "preview" thingy sometimes.
Main reason is probably that less drunks are the same place at the same time since they go home over a 2h period instead of a 5m period ... AND that the drunks are not pissed off because they are FORCED to go home!! :-)
(Down here, when the pub's owner wants to send the hardcore drunks out, he starts cleaning the place -- which occasionally involves flooding the floor with soapy water and moping it, and yes, I had my fet wet this way a lot of times...)
good-resolution videoconferencing could burn that much bandwidth
(*) two years ago, the price difference between gas and alcohol was up to 60% (alcohol being 0.4 times the price of gasoline per liter) because the market was not hot for sugar.
Except that, with oil at current prices, it's Waay cheaper to fill your tank with ethanol than to fill it with gas.
But, seriously, though: the same car, on ethanol, makes 10-20% worse mileage than with gas. Down here we have "flex-power" (ethanol/gasoline flexible fuel system) cars, and if a car gets 12km/l(30mpg, 8l/100km) on gas, it usually will get 10+ km/l (25mpg, 10l/100km) on ethanol. Currently, in my town, the pump price for alcohol is about R$ 2,10/l (US$ 3.85/gallon) and the pump price for gas, R$ 2,50/l (US$ 4.59/gallon), which is a 19% difference.
IOW: renewable and non-renewable fuels break even (with a slight advantage for ethanol) on mileage per dollar.
On the performance side, on ethanol cars tend to have a higher final speed than on gas, but they have some 5-10% less torque.
your answer makes sense, even if you are completely wrong :-)
For example, let's say that the filesystem queues up too many requests. If you restart the SCSI driver and resend the requests, the queue will probably be even longer by now, causing it to crash again. Your system is now in an infinite loop of restarting the SCSI driver, while you hope that it's not corrupting the disk every time it gets automatically reset.
Or more likely, the filesystem sent a bunch of requests to the SCSI driver, assuming they would be fulfilled. When the SCSI driver crashes, the filesystem has no idea which requests were completed or not, and has not kept track of which ones were outstanding. Now your filesystem is consistent in-memory but not on-disk. When you restart the SCSI driver it will continue corrupting the filesystem even further. The only way around this is to restart the filesystem on that disk too (I hope it was journaled), meaning that it would probably have to invalidate any open files. If this happened while writing or reading your VM paging file, the system is almost completely hosed.
Scenario #1: when the disks went whacko, I would expect (besides trying again) that the server sends a "cancel" message to every pending-i/o process, and let the apps recover.
Scenario #2: the error is in the "assuming they would be fullfilled". If the SCSI says "I fullfilled your batch", then it's fullfilled, not before.
Scenario #3: disk error on swap partition: the affected processes are killed.
whoa, let's reboot the machine... oops, it encountered the same bug it did before; reboot again; oops; reboot again???
(IOW: if the bug is that persistent, you're hosed anyway.)
like, say, K-MENU [click] System [click] Adept [click] [wait half a second] [type password] [enter] [wait 5 seconds] hmmm... [click on search box] j2 [wait a second] [click on j2re entry on the grid] Install package [click] Apply changes [click] [wait 30 seconds] ???
When I re-watch "Stand by Me" (or read King's Four Seasons) I feel moved. "Braveheart", I never thought very much of it; but almost every Stephen King-based story strikes me profoundly.
It's LaTeX, but graphic. Really nice.
Estatistically speaking, and in the long run (meaning tending toward infinity) the "profiles" of people that browse site A and site B would fall in more or less the same -- meaning it's not the overcounting or the undercounting that cancel each other -- is the the over/undercounting for site A is more or less the same over/undercounting for site B.
Therefore, if site A has 10 million IPs logged yesterday and site B has 5 million IPs logged yesterday, it's fair to say that site A has double the traffic of site B.
That is, unless you can prove that the user "profiles" (PPP, PPPoE, people that use lots of different computers versus proxies, NATted people etc) are wildly different between site A and site B -- which can occur e.g. in pr0n sites (which would be blocked by most corporate proxies but not home computers etc), the number of distinct visiting IPs is a good measure of site traffic.
because that was exactly my point: but it applies to tangible goods, also.
"perception of value" === "value"
s/Mexico/Puerto Rico/
HTH
There is _no_ competitors. Where an Apple product is in -- especially iPods -- its market share is segregated and guaranteed.