Huh? The authoritarian fuckers made no effort to hide their ID plans, so why the hell did you vote for them? Tossers like you gave the bastards their mandate.
Tossers like him/her gave the bastards the 23% (or whatever it was) support of the electorate that the bastards needed to dictate to the country for another five years.
Getting slightly over a third of the vote, or under a quarter of the electorate hardly counts as a mandate.
These bastards have taken even our lack of democracy to unplumbed depths but Mrs T. didn't get much more support so it's hardly just a Labour thing.
All that has to happen is that that ssl site, rather than asking you to pick a password, asks you for a certificate (which can be self signed).
Or if the site wants to verify your email address, asks you for a certificate request and it signs that and then emails it back to you.
For the more clueless, the server can even offer to generate the certificate and key and let you download them. Assuming we are using SSL for this connection then there isn't really a problem. Of course, potentially, the server could give your key to someone else as well and then they could impersonate you to that server but the server could currently give away your password.
Note there is no need for "Certificate Authorities". They only apply if two people want to use a third party to verify who they are.
Now the "key" is only stored on the users machine. There is no need for revocation lists because the certificate isn't being used to verify your identity to a third party - I upload a new cert (change my password) and the server just forgets the old one so even if someone did get the key to that cert it is only as useful as an old password.
... "Anyone offering that price must receive a license to that patent."
I'm not sure this doesn't allow people to price everyone else out though.
Lets say you own a patent that gets you $10M profit a year selling a product that nobody else can duplicate. Furthermore assume you have priced such that you are getting the maximum possible return - selling at a higher or lower price will result in less profit overall. Also assume that the tax is 10% of your patent valuation.
If you value your patent at $50M you still make $5M profit after tax each year.
Now if each licensee has to pay $50M for the rights to the patent it is unlikely that anyone would be able to put forward a business case that makes sense. Even if they can capture 100% of the market it will still take them 5 years to break even, and by taking the market away from the inventor he/she will have to revalue the patent at zero anyway because otherwise he/she would have to keep paying that tax on what is now no income so other people would come into the market to compete without forking out the $50M.
The fundamental problem is how to fairly value the IP.
But how about the owner of the IP valuing it. And what this valuation means is that, if, in the next 12 months from valuation, they are paid the amount of their valuation (from any source), the IP becomes public domain. And if nobody choses to buy it they pay x% tax.
Maybe there needs to be a "no tax, effectively infinite value" period initially to allow companies to exploit and assess the value of their IP - say 5 years (maybe 3 for things like software that don't need large factories to exploit)
Or maybe these first years are "deferred" tax. This would also give companies an incentive to exploit their invention fast and then donate it to the public domain after 5 years (they value it at zero after five years are up - the last five years of taxes being calculated on this initial valuation.
"The reality of what tends to happen is that very few people ever actually read the source code of OS products, much less modify it. And if the bug happens to get past the original developers, there is very little chance that a stranger to the code will find it."
But this isn't always true:
My original post - and at this point I had never compiled, nor run Monotone. http://lists.gnu.org/archive/html/monot one-devel/2 004-07/msg00003.html
And the original programmers reply: http://lists.gnu.org/archive/html/monotone -devel/2 004-07/msg00012.html "oh gosh! that's a bad one."
Unfortunately, I still haven't had time to really get to know Monotone, either as a user or a programmer despite the fact that I had started writing something with similar design goals before I found Monotone. (My program hadn't got further than import and checkout by the time I found Monotone so I wasn't throwing away much)
I suspect you spend most of your time looking at, and writing your own code. I spend most of my time looking at, and fixing, other peoples code.
And references give people power. They can change what was an input parameter into an input/output parameter by changing void myfunc(int); into void myfunc(int&); and not touching the caller.
The fact is that probably 80% of the people who write code should never have been allowed infront of an editor.
do_something((const)i); does not compile. And even if you change it to do_something((const int)i); it still won't compile if someone has written void do_something(int&); I've seen this where do_something(int) would have been sufficient but the original programmer obviously thought "This is C++. We use references in C++". Later, another programmer makes a change to the function do_something and ends up changing the value of the parameter. Now I'm left with a situation where I know a value was correct at line 100 and was incorrect at line 200 in a different function and I have to work out where the value changed. And I didn't write ANY of this code.
Had the first programmer written const int& then yes, the problem wouldn't have occurred. But, maybe, if they had had to write do_something(&x) if the prototype was do_something(int&) and do_something(x) if the prototype was do_something(const int&) we might get more const correctness.
And a similar issue occurs when using polymorphism. I know things were right at one line, and I know they had gone wrong at another line but it is impossible to determine where things went inbetween. Polymorphism can be a good thing but it can, and more often than not is, be abused which leads to more maintenance headaches, not less.
class A { public: virtual void do_something() { std::cout << "A" << std::endl; } }; class B:public A { public: virtual void do_something() { std::cout << "B" << std::endl; } };
void myfunc(A& a) {
a.do_something(); }
Does this code print A or B when myfunc is called? Which do_something() is executed? Unless you are using function pointers, this ambiguity does not exist in C (or FORTRAN)
For the first case I wrote: There is no way of telling just by looking at this code.
in C if I have int i; do_something(i); I know that i doesn't change. I don't need to look at the prototype (ok do_something could be a macro)
What I would have liked is for the caller to a non-const reference to still have to write that '&'. The callee gets the benefit of the reference and anyone looking at the caller code immediately knows that there might be something happening that isn't explicit in the code they are looking at.
In theory a smart syntax highlighter could do this but it would probably have to almost be a compiler to handle this, especially if you want it to be correct, even in the face of macros. And I don't want to have to wait for a compilation every time I load a file up into my editor.
An editor could not handle something like: #ifdef REALLY_SCREW_THINGS_UP void shit(int& i) { i=i+1; } #else void shit(int i) { } #endif
int main() {
int i=0;
shit(i); }
But a compiler could have refused to compile unless that call in main was changed to shit(&i); if REALLY_SCREW_THINGS_UP was defined.
C++ also has some pitfalls not disimilar to the problems FORTRAN can bring (maybe later versions of Fortran are better)
function.cpp int i;... do_something(i);...
There is no way of telling just by looking at this code whether the value of i changes in the function call. In C you know it doesn't.
It can be a great help when trying to trace back from a coredump where a crazy value came from to know that a function call couldn't possibly have affected the value without having to look at headers or, even worse, the source of the function.
Another problem C++ brings to the table:
void myfunc(myclass& c) {... c.do_something();... }
Here it is generally impossible to determine what code will be executed in the call to do_something without knowing the code for the caller to myfunc.
These things don't have to cause problems but in the wrong hands they are perfect for writing brittle code
It's no wonder there is so much crap code about when people claim things for a language they clearly don't know. #include <stdio.h> void foo(char (*x)[3][3]); int main(void){
char x[3][3];
printf("%u\n",(unsigned)sizeof(x));/* Prints 9 */
foo(&x);
return 0; } void foo(char (*x)[3][3]){
printf("%u\n",(unsigned)sizeof(*x));/* Prints 9 */ }
As long as your types match you can pass multi-dimentional arrays around in C quite happily.
Don't like the extra dereference? grab a c++ compiler instead (but carry on writing "C") #include <stdio.h> void foo(char (&x)[3][3]); int main(void){
char x[3][3];
printf("%u\n",(unsigned)sizeof(x));/* Prints 9 */
foo(x);
return 0; } void foo(char (&x)[3][3]){
printf("%u\n",(unsigned)sizeof(x));/* Prints 9 */ }
Want to be able to pass arbitrary sized arrays? Now you are limited to C++. #include <stdio.h> template<int Y, int X> void foo(char (&x)[Y][X]); int main(void){
char x[3][3];
printf("%u\n",(unsigned)sizeof(x));/* Prints 9 */
foo(x);
return 0; } template<int Y, int X> void foo(char (&x)[Y][X]) {
printf("%u\n",(unsigned)sizeof(x));/* Prints 9 */ }
$ grep www.microsoft.com/etc/hosts 127.0.0.1 www.microsoft.com $lynx -dump http://www.microsoft.com/EULA.htm The software is public domain. There are no restrictions whatever on what you can do with it. $
And if you implemented this in a proxy? The person viewing the EULA could have no idea.
And https:// won't work for microsoft.com. The certificate for www.microsoft.com has been signed by an unrecognised CA according to firefox. It's trivial to create a certificate that looks the same.
Linux just doesn't seem to do colour management at all. (At least I can't find any)
The Gimp can't even tell me if the colours in an image are out of gamut. But until the Gimp will accept input and output profiles I can't see how it could do that anyway.
To follow up on that I've just found: http://www.photo-news.com/pnl/1997-pnl/710 27-PNL.h tm
Which has: The first phase of the new Kodak Picture Network Europe, to be launched in the UK, Germany and Sweden in December, will be accessible by computer users via standard Web browsers, and will enable them to order prints of their favorite pictures. The next phase, expected in early 1998, will broaden the service to select retailers, enabling access and use by consumers who may not have a suitable PC for access to the Internet.
So Kodak was planning to launch their service end 1997.
I'm almost certain that there is implemented prior art to this. (Kodak had "Kodak picture network Europe" doing all of this implemented and usable in October 1998 (when I joined them). I can remember us supporting belgium franc/flemish, belgium franc/belgium french, french franc/french, english pound/english as a minimum. (I'm pretty sure there were more - german, dutch, swedish? but I couldn't swear to it)
Using google turned up http://europa.eu.int/ISPO/docs/services/trends/doc s/78.pdf dated feb/march 1997 has kodak buying a stake in picturevision and integrating it with KPN. PV was image store/email etc and KPN was the printing and distribution.
The only real issue we had was - If you take orders from a customer in one country, on a webserver in a second, banking in the currency of the customer into a bank in England, for printing an image in a third country for delivery to someone in a fourth country (all in the EU) then who should you pay the VAT to and what rate should you charge?
It's trivial to prove that there is an arbitrarily long sequence of numbers with no primes in it.
(n+1)!+2... (n+1)!+n+1 is a run of n numbers none of which are prime.
Of course, this doesn't mean that you have to go all the way to (n+1)! before you can find a run of n numbers without a prime, merely that such a run must exist.
MnO4 will catalyse 2H2O2 -> 2H2O + O2. Doesn't necessarily require a protein. Most terrestrial organisms use iron for transporting Oxygen, why shouldn't an extraterrestrial organism use something else?
Proxima Centauri (or Alpha Centauri C) is only 4.22 light-years away. (393 927 289 812km)
365*24*3600*4.22*300000
39924576000000.00
Hmm, you seem to be out by two orders of magniture there.
4*3.14159*39924576000000*39924576000000
20030423076323425935360000000.00000
The earths radius is 6.3781 x 10^3km. multiply by pi to get the area
projected area is pi r^2 not pi r
6378*6378*3.14159
127796375.18556
20030423076323425935360000000/127796375
156737020720058968302
Or about 1.6*10^20
Doesn't hugely affect the conclusion though.
When you have water under pressure, you can superheat it far above boiling (up to the structural limits of the pressurizer).
This is not so. You can only heat water and keep it a liquid up to the critical point. After that it is a vapour even if you have it under enough pressure to keep it as dense as liquid water. Critical point of water is around 600K IIRC.
>You've got it right. No one *has* to buy the damned operating system.
I have just (last weekend) bought a new laptop. I expressly asked for it not to have windows but that wasn't possible.
I knew I didn't want windows ME installed and wanted to claim the refund so I booted a linux installer directly. Fdisked and installed. I never ran even one byte of M$hafs code. I also didn't open the shrinkwrapped ME manual (and I assumed CD)
It turns out that the ME software was on the harddisk only, no CD supplied so not only was I forced to buy it but I can't even use it now even if I wanted to.
There are a few web companies that will supply laptops without windows but, for a laptop at least, I want to be able to pick it up and feel its weight and size and make sure that I will be comfortable with it. Keys move around on keyboards etc.
So yes, M$haft used their monopoly position to force me to pay for vapourware.
The certificate of authenticity "sticker" is on the bottom of the laptop (probably undetachable) so I probably can't even sell the licence for ME to someone else who wants it, I believe the right to sell on a licence like this is a statutory right in EU law and M$haft can't deny it in an EULA even if I had agreed to one. I don't use any M$haft software on any of my personal machines so m$haft can't even claim I ought to have known. Infact the last piece of M$haft software I did use was w98 (first edition) although I have acquired licences for w98SE and now wME over the years.
www.dnuk.com charges 80GBP for winME (including install) and I would expect that most of that goes to M$crewTheUsersForEveryPenny
M$ have _STOLEN_ c80GBP of my money this week and I want it back.
In total I suspect that M$ have extracted about 120GBP in licences for software I have never used in the last 12 months.
[root@feynman tim]#/sbin/fdisk -l/dev/hda
Disk/dev/hda: 255 heads, 63 sectors, 2432 cylinders
Units = cylinders of 16065 * 512 bytes
Device Boot Start End Blocks Id System
/dev/hda1 1 3 24066 83 Linux
/dev/hda2 4 101 787185 82 Linux swap
/dev/hda3 102 2432 18723757+ 5 Extended
/dev/hda5 102 167 530113+ 83 Linux
/dev/hda6 168 233 530113+ 83 Linux
/dev/hda7 234 495 2104483+ 83 Linux
/dev/hda8 496 757 2104483+ 83 Linux
/dev/hda9 758 2432 13454406 83 Linux
p.s. Does anyone know whether the FSF has a "donate windows licences to us so we can get the refund" policy?
>When somebody tries to interact with a service on it, say, FTP, not only does it keep a full record of the session, it also portscans them, fingers them, WHOISs them, tries to get banners from their FTP, mail, and web servers, and all that good stuff. Why? Because there is NO reason that anybody would ever hit it. So we want to know about the people who are.
So when I accidently type the wrong IP into a URL (And yes I do use IPs from time to time, expecially when I am setting up boxes at work that don't yet have a dns record) and mis-type it you are going to scan the IP that that connection came from.
Woopie doo. If I connect from home you will manage to scan my ISP's proxy. From work you will scan the parent companies proxy. At my previous job the proxy wasn't even in the same country as I was in.
And these are cases where I have made no attempt to disguise myself. Does your honeypot require a full connect before you instigate your attacks or will a single SYN packet be enough? Perhaps if your machine starts scanning random IPs across the globe you might do something about your security.
If you are going to set up honeypots then you should first learn all the different attacks that the black hats might use and also how, just occasionally, an innocent party might accidently stumble at the gate.
Slashdot Mirror
Shameless plug :-)
http://http-console.sourceforge.net/
Tossers like him/her gave the bastards the 23% (or whatever it was) support of the electorate that the bastards needed to dictate to the country for another five years.
Getting slightly over a third of the vote, or under a quarter of the electorate hardly counts as a mandate.
These bastards have taken even our lack of democracy to unplumbed depths but Mrs T. didn't get much more support so it's hardly just a Labour thing.
All that has to happen is that that ssl site, rather than asking you to pick a password, asks you for a certificate (which can be self signed).
Or if the site wants to verify your email address, asks you for a certificate request and it signs that and then emails it back to you.
For the more clueless, the server can even offer to generate the certificate and key and let you download them. Assuming we are using SSL for this connection then there isn't really a problem. Of course, potentially, the server could give your key to someone else as well and then they could impersonate you to that server but the server could currently give away your password.
Note there is no need for "Certificate Authorities". They only apply if two people want to use a third party to verify who they are.
Now the "key" is only stored on the users machine. There is no need for revocation lists because the certificate isn't being used to verify your identity to a third party - I upload a new cert (change my password) and the server just forgets the old one so even if someone did get the key to that cert it is only as useful as an old password.
... "Anyone offering that price must receive a license to that patent."
I'm not sure this doesn't allow people to price everyone else out though.
Lets say you own a patent that gets you $10M profit a year selling a product that nobody else can duplicate. Furthermore assume you have priced such that you are getting the maximum possible return - selling at a higher or lower price will result in less profit overall.
Also assume that the tax is 10% of your patent valuation.
If you value your patent at $50M you still make $5M profit after tax each year.
Now if each licensee has to pay $50M for the rights to the patent it is unlikely that anyone would be able to put forward a business case that makes sense. Even if they can capture 100% of the market it will still take them 5 years to break even, and by taking the market away from the inventor he/she will have to revalue the patent at zero anyway because otherwise he/she would have to keep paying that tax on what is now no income so other people would come into the market to compete without forking out the $50M.
Tim.
I liked the idea as well.
The fundamental problem is how to fairly value the IP.
But how about the owner of the IP valuing it. And what this valuation means is that, if, in the next 12 months from valuation, they are paid the amount of their valuation (from any source), the IP becomes public domain. And if nobody choses to buy it they pay x% tax.
Maybe there needs to be a "no tax, effectively infinite value" period initially to allow companies to exploit and assess the value of their IP - say 5 years (maybe 3 for things like software that don't need large factories to exploit)
Or maybe these first years are "deferred" tax. This would also give companies an incentive to exploit their invention fast and then donate it to the public domain after 5 years (they value it at zero after five years are up - the last five years of taxes being calculated on this initial valuation.
Tim.
"The reality of what tends to happen is that very few people ever actually read the source code of OS products, much less modify it. And if the bug happens to get past the original developers, there is very little chance that a stranger to the code will find it."
t one-devel/2 004-07/msg00003.html
e -devel/2 004-07/msg00012.html
But this isn't always true:
My original post - and at this point I had never compiled, nor run Monotone.
http://lists.gnu.org/archive/html/mono
And the original programmers reply:
http://lists.gnu.org/archive/html/monoton
"oh gosh! that's a bad one."
Unfortunately, I still haven't had time to really get to know Monotone, either as a user or a programmer despite the fact that I had started writing something with similar design goals before I found Monotone. (My program hadn't got further than import and checkout by the time I found Monotone so I wasn't throwing away much)
Tim.
The UK government isn't going to bother with parliament anymore - they're going to bypass it: http://www.theregister.co.uk/2005/04/12/uk_passpor t_fingerprints/
Not only that but the 7805 is going to be dissipating more power than all the cpus put together if it's being driven from 12V.
I suspect you spend most of your time looking at, and writing your own code. I spend most of my time looking at, and fixing, other peoples code.
And references give people power. They can change what was an input parameter into an input/output parameter by changing void myfunc(int); into void myfunc(int&); and not touching the caller.
The fact is that probably 80% of the people who write code should never have been allowed infront of an editor.
do_something((const)i); does not compile. And even if you change it to do_something((const int)i); it still won't compile if someone has written void do_something(int&); I've seen this where do_something(int) would have been sufficient but the original programmer obviously thought "This is C++. We use references in C++". Later, another programmer makes a change to the function do_something and ends up changing the value of the parameter. Now I'm left with a situation where I know a value was correct at line 100 and was incorrect at line 200 in a different function and I have to work out where the value changed. And I didn't write ANY of this code.
Had the first programmer written const int& then yes, the problem wouldn't have occurred. But, maybe, if they had had to write do_something(&x) if the prototype was do_something(int&) and do_something(x) if the prototype was do_something(const int&) we might get more const correctness.
And a similar issue occurs when using polymorphism. I know things were right at one line, and I know they had gone wrong at another line but it is impossible to determine where things went inbetween. Polymorphism can be a good thing but it can, and more often than not is, be abused which leads to more maintenance headaches, not less.
For the second case:
#include <iostream>
class A { public: virtual void do_something() { std::cout << "A" << std::endl; } };
class B:public A { public: virtual void do_something() { std::cout << "B" << std::endl; } };
void myfunc(A& a) {
a.do_something();
}
Does this code print A or B when myfunc is called? Which do_something() is executed? Unless you are using function pointers, this ambiguity does not exist in C (or FORTRAN)
For the first case I wrote:
There is no way of telling just by looking at this code.
in C if I have int i; do_something(i); I know that i doesn't change. I don't need to look at the prototype (ok do_something could be a macro)
What I would have liked is for the caller to a non-const reference to still have to write that '&'. The callee gets the benefit of the reference and anyone looking at the caller code immediately knows that there might be something happening that isn't explicit in the code they are looking at.
In theory a smart syntax highlighter could do this but it would probably have to almost be a compiler to handle this, especially if you want it to be correct, even in the face of macros. And I don't want to have to wait for a compilation every time I load a file up into my editor.
An editor could not handle something like:
#ifdef REALLY_SCREW_THINGS_UP
void shit(int& i) { i=i+1; }
#else
void shit(int i) { }
#endif
int main() {
int i=0;
shit(i);
}
But a compiler could have refused to compile unless that call in main was changed to shit(&i); if REALLY_SCREW_THINGS_UP was defined.
C++ also has some pitfalls not disimilar to the problems FORTRAN can bring (maybe later versions of Fortran are better)
... ...
... ...
function.cpp
int i;
do_something(i);
There is no way of telling just by looking at this code whether the value of i changes in the function call. In C you know it doesn't.
It can be a great help when trying to trace back from a coredump where a crazy value came from to know that a function call couldn't possibly have affected the value without having to look at headers or, even worse, the source of the function.
Another problem C++ brings to the table:
void myfunc(myclass& c) {
c.do_something();
}
Here it is generally impossible to determine what code will be executed in the call to do_something without knowing the code for the caller to myfunc.
These things don't have to cause problems but in the wrong hands they are perfect for writing brittle code
It's no wonder there is so much crap code about when people claim things for a language they clearly don't know.
/* Prints 9 */ /* Prints 9 */
/* Prints 9 */ /* Prints 9 */
/* Prints 9 */ /* Prints 9 */
#include <stdio.h>
void foo(char (*x)[3][3]);
int main(void){
char x[3][3];
printf("%u\n",(unsigned)sizeof(x));
foo(&x);
return 0;
}
void foo(char (*x)[3][3]){
printf("%u\n",(unsigned)sizeof(*x));
}
As long as your types match you can pass multi-dimentional arrays around in C quite happily.
Don't like the extra dereference? grab a c++ compiler instead (but carry on writing "C")
#include <stdio.h>
void foo(char (&x)[3][3]);
int main(void){
char x[3][3];
printf("%u\n",(unsigned)sizeof(x));
foo(x);
return 0;
}
void foo(char (&x)[3][3]){
printf("%u\n",(unsigned)sizeof(x));
}
Want to be able to pass arbitrary sized arrays? Now you are limited to C++.
#include <stdio.h>
template<int Y, int X> void foo(char (&x)[Y][X]);
int main(void){
char x[3][3];
printf("%u\n",(unsigned)sizeof(x));
foo(x);
return 0;
}
template<int Y, int X> void foo(char (&x)[Y][X]) {
printf("%u\n",(unsigned)sizeof(x));
}
$ grep www.microsoft.com /etc/hosts
127.0.0.1 www.microsoft.com
$lynx -dump http://www.microsoft.com/EULA.htm
The software is public domain. There are no restrictions whatever on what you can do with it.
$
And if you implemented this in a proxy? The person viewing the EULA could have no idea.
And https:// won't work for microsoft.com. The certificate for www.microsoft.com has been signed by an unrecognised CA according to firefox. It's trivial to create a certificate that looks the same.
Linux just doesn't seem to do colour management at all. (At least I can't find any)
The Gimp can't even tell me if the colours in an image are out of gamut. But until the Gimp will accept input and output profiles I can't see how it could do that anyway.
To follow up on that I've just found:0 27-PNL.h tm
http://www.photo-news.com/pnl/1997-pnl/71
Which has:
The first phase of the new Kodak Picture Network Europe, to be launched in the UK, Germany and Sweden in December, will be accessible by computer users via standard Web browsers, and will enable them to order prints of their favorite pictures. The next phase, expected in early 1998, will broaden the service to select retailers, enabling access and use by consumers who may not have a suitable PC for access to the Internet.
So Kodak was planning to launch their service end 1997.
I'm almost certain that there is implemented prior art to this. (Kodak had "Kodak picture network Europe" doing all of this implemented and usable in October 1998 (when I joined them). I can remember us supporting belgium franc/flemish, belgium franc/belgium french, french franc/french, english pound/english as a minimum. (I'm pretty sure there were more - german, dutch, swedish? but I couldn't swear to it)
c s/78.pdf dated feb/march 1997 has kodak buying a stake in picturevision and integrating it with KPN. PV was image store/email etc and KPN was the printing and distribution.
Using google turned up http://europa.eu.int/ISPO/docs/services/trends/do
The only real issue we had was - If you take orders from a customer in one country, on a webserver in a second, banking in the currency of the customer into a bank in England, for printing an image in a third country for delivery to someone in a fourth country (all in the EU) then who should you pay the VAT to and what rate should you charge?
It's trivial to prove that there is an arbitrarily long sequence of numbers with no primes in it.
... (n+1)!+n+1 is a run of n numbers none of which are prime.
(n+1)!+2
Of course, this doesn't mean that you have to go all the way to (n+1)! before you can find a run of n numbers without a prime, merely that such a run must exist.
Eight addresses. 1 broadcast, 1 network, 1 router, 4 machines. Looks like seven out of eight to me.
That should be MnO2 - Manganese IV oxide. Duh!
MnO4 will catalyse 2H2O2 -> 2H2O + O2. Doesn't necessarily require a protein. Most terrestrial organisms use iron for transporting Oxygen, why shouldn't an extraterrestrial organism use something else?
Proxima Centauri (or Alpha Centauri C) is only 4.22 light-years away. (393 927 289 812km) 365*24*3600*4.22*300000 39924576000000.00 Hmm, you seem to be out by two orders of magniture there. 4*3.14159*39924576000000*39924576000000 20030423076323425935360000000.00000 The earths radius is 6.3781 x 10^3km. multiply by pi to get the area projected area is pi r^2 not pi r 6378*6378*3.14159 127796375.18556 20030423076323425935360000000/127796375 156737020720058968302 Or about 1.6*10^20 Doesn't hugely affect the conclusion though.
When you have water under pressure, you can superheat it far above boiling (up to the structural limits of the pressurizer).
This is not so. You can only heat water and keep it a liquid up to the critical point. After that it is a vapour even if you have it under enough pressure to keep it as dense as liquid water.
Critical point of water is around 600K IIRC.
Or redirect them to www.microsoft.com. :-)
>You've got it right. No one *has* to buy the damned operating system.
/sbin/fdisk -l /dev/hda
/dev/hda: 255 heads, 63 sectors, 2432 cylinders
I have just (last weekend) bought a new laptop. I expressly asked for it not to have windows but that wasn't possible.
I knew I didn't want windows ME installed and wanted to claim the refund so I booted a linux installer directly. Fdisked and installed. I never ran even one byte of M$hafs code. I also didn't open the shrinkwrapped ME manual (and I assumed CD)
It turns out that the ME software was on the harddisk only, no CD supplied so not only was I forced to buy it but I can't even use it now even if I wanted to.
There are a few web companies that will supply laptops without windows but, for a laptop at least, I want to be able to pick it up and feel its weight and size and make sure that I will be comfortable with it. Keys move around on keyboards etc.
So yes, M$haft used their monopoly position to force me to pay for vapourware.
The certificate of authenticity "sticker" is on the bottom of the laptop (probably undetachable) so I probably can't even sell the licence for ME to someone else who wants it, I believe the right to sell on a licence like this is a statutory right in EU law and M$haft can't deny it in an EULA even if I had agreed to one. I don't use any M$haft software on any of my personal machines so m$haft can't even claim I ought to have known. Infact the last piece of M$haft software I did use was w98 (first edition) although I have acquired licences for w98SE and now wME over the years.
www.dnuk.com charges 80GBP for winME (including install) and I would expect that most of that goes to M$crewTheUsersForEveryPenny
M$ have _STOLEN_ c80GBP of my money this week and I want it back.
In total I suspect that M$ have extracted about 120GBP in licences for software I have never used in the last 12 months.
[root@feynman tim]#
Disk
Units = cylinders of 16065 * 512 bytes
Device Boot Start End Blocks Id System
/dev/hda1 1 3 24066 83 Linux
/dev/hda2 4 101 787185 82 Linux swap
/dev/hda3 102 2432 18723757+ 5 Extended
/dev/hda5 102 167 530113+ 83 Linux
/dev/hda6 168 233 530113+ 83 Linux
/dev/hda7 234 495 2104483+ 83 Linux
/dev/hda8 496 757 2104483+ 83 Linux
/dev/hda9 758 2432 13454406 83 Linux
p.s. Does anyone know whether the FSF has a "donate windows licences to us so we can get the refund" policy?
Tim.
>When somebody tries to interact with a service on it, say, FTP, not only does it keep a full record of the session, it also portscans them, fingers them, WHOISs them, tries to get banners from their FTP, mail, and web servers, and all that good stuff. Why? Because there is NO reason that anybody would ever hit it. So we want to know about the people who are.
So when I accidently type the wrong IP into a URL (And yes I do use IPs from time to time, expecially when I am setting up boxes at work that don't yet have a dns record) and mis-type it you are going to scan the IP that that connection came from.
Woopie doo. If I connect from home you will manage to scan my ISP's proxy. From work you will scan the parent companies proxy. At my previous job the proxy wasn't even in the same country as I was in.
And these are cases where I have made no attempt to disguise myself. Does your honeypot require a full connect before you instigate your attacks or will a single SYN packet be enough? Perhaps if your machine starts scanning random IPs across the globe you might do something about your security.
If you are going to set up honeypots then you should first learn all the different attacks that the black hats might use and also how, just occasionally, an innocent party might accidently stumble at the gate.
Tim.