Well, unless you plan to invent a time machine and die in the past, the odds of you living when this scheme gets implemented are pretty good, because it have already been implemented here in Danmark, where all current danish cards does have a chip. And the solution to backward compability is quite simple. All cards and card-readers include both the old and new solution.
But the banks have issued new cards to all users, and required all atms to be able to read the chip. So the backward compability is currently only used with foreign cards.
If you did it yes, but if you instead told the Police that someone had bought stolen goods, then the police would come and get it back.
But imagine the stolen goods was a toy aircraft which could fly home to the real owner. In that case activating the "Fly home" button would be ok, but breaking into the house, and taking the aircraft back would not.
But the problem is that the user had no right to the book in the first place, because it was stolen property(Amazon stole it, from the people who had really, the right to sell it).
Lets take an example from the physical world. If I steal 1000 printed books of "1984" and sell them to you, then you have no right to keep them, even thou you paid me.
So Amazon are fucked, no matter what they do. If they did nothing, the result would be a court order which would order them to get all the distributed copiers back. And if Amazon waited for that court order, the headline would instead be
"Amazon sell stolen books, and refuse to pay the publisher", not a good position to be in from Amazon either.
Except for the fact, that this would require all their 50 million users to upgrade their Skype software. Because Ebay can't make an compability version of the prodotol due to patents.
(And many of those skype installs are on mobile phones, where an upgrade may not be that easy for most users).
What people said was that "no one is going to make money developing FLOSS software", and it's almost as true now as it was then. Very few(I can't name any from the top of my head) companies have developed open-source software, and then earned enough money on that(Including support) to pay for the development of that software.
Remember that >98% of the software included with "Redhat Linux" have been developed and paid for by companies other then Redhat. (Some of it, by stupid investors in the.com crash)
Well, most of the things are already supported by compilers, or will be soon. (Se forexample http://gcc.gnu.org/gcc-4.5/cxx0x_status.html for the list of that gcc 4.5 (The newest released gcc version) supports. So it is likely that c++0x will be supported almost fully, within a year of release. (Microsoft and intel have also implemented much of the standard).
The only thing I miss from the gcc is an implementation of "Lambda expressions" and they are working on those. (They have a branch which afair kinda works:}
And it's called synthesiagame now. Something about Activision not liking them using the name "Piano hero"
And the good thing is, it runs perfect in linux using wine:-} (Just remember to uninstall pulseaudio if you use a recent "fedora core" based distro, and need to use timidity for midi.
Is it possible to use Google wave for all users now? wave.google.com just have a "notify me when it's done" button. Is there any way to download it now, or is it only for for a few select developers yet?
So instead of just writing a gui(And a driver to do mesh network), they wrote almost an entire Operation system. Sound like a giant case of "Not invented here" syndrom.
"But what we did...was we had Sugar do the power management, we had Sugar do the wireless management--it became sort of an omelet. The Bios talked directly with Sugar, so Sugar became a bit of a mess."
I always thought that sugar was the graphics interface(Window manager, desktop and widgets, something like kde or gnome, but lighter).
So is he confused about what sugar is, or is sugar really a kind of mini os, operating on a linux kernel? Why should sugar for example include a wireless driver as part of sugar, instead of including the wireless driver as a standard linux driver, and then letting sugar communicate with it, as it would with any other driver.
The compiler is allowed to assume that no other code(Including no other thread, running the same code) change the value of a variable behind its back(As long as the variable it not volatile(Volatile got it's own can of worms)), so the optimization is safe.
so in the code
int *data=myFunc(); val=*data; printf("%d\n",val); val=*data;// The compiler is allowed to optimize this call out. printf("%d\n",val);
And the compiler is allowed to turn this into nothing:
int *val=myFunc();// Returns a valid pointer to an int. *val=10; if(*val==11) {
*val=42;// The compiler may remove anything in here. }
If multi-threaded code were allowed to do what you describe, then it would be impossible to do most of the optimizations that good c compilers do.
If gcc had a specific case to find use of pointer, and then later test of same pointer for 0 That would be evil and useless(And should give a warning).
But what most likely happens is that gcc have one optimization pass, that analyse pointer values. And this detect(And remember) that tun does point to a valid object in the entire block. (Or that the code is invalid, but the compiler always assume that the code is valid).
Then later, the compiler run an other optimization pass, to remove code with no effect. It then look at the if(tun==null) test, and see that a previous pass found out that tun would always point to a valid object. Thus the test is redundent and removed.
But it's (tun->sk) not &(tun->sk). That is: The code is looking at the value of the member sk in the struct pointed to by tun. Looking at this value is undefined if tun is null. It does not take the address of tun or tun->sk.
This sound more like a gcc/embeded os bug. There is no requirement in c/c++ that the null pointer is (int)0. That is: It don't have to be all 0 bits. It just need to be distinct from any valid pointer, so if you run on a platform where you use memory address 0(Valid, but still wierd), you need to config gcc(If possible, I don't know if gcc supports this) to use an other bit pattern as null pointer(say 0xeffff), and then you need to configure your embeded os, to never return a memory address/buffer that contain the address you have used as null pointer.
I think the compiler is correct. If tun is null, then tun->sk is undefined and the compiler can do what even optimization it want.
So when the compiler see tun->sk it can assume that tun is not null, and do the optimization, because IF tun is null, then the program is invoked undefined behavier, which the compiler don't have to preserve/handle. (How do you keep the semantic of an undefined program??)
But who should fork MySQL? >80% of the development is made my the MySql inc company(now sun, soon oracle) so there is not many developers left to do a fork.
Yes but what should the Mozilla developers do when they run on a system such as Linux which can't legally support H.264 in USA and other parts of the world?
Should they just do a "Not support, please install Windows?"
Re-reading my text: Nope I don't mention mysql. Most likely because I like PostgreSQL much more, and only really use Mysql on legacy systems. (The kind I take over from someone else, because the fired the original developer).
But can you clerify on the
PostgreSQL has no problem returning to me a list of rows contained in a single column of a single row in a result. The tree you are looking for is easy to return, I have stored procedures that work this way myself. It may be PostgreSQL specific,
I looked in the manual, but I could not find anything doing what you described above. Got any sql samples or tutorials that explain how you do it?
It may be a bit faster on searching, but it take ~5 times as long to generate the index, and use twice as much memmory when searching so it may just be a different trade off between index time and search time.
And it's a bad search test, because the total search time is less them 2 seconds, thus not including the cost of the gc for java.
hint to people doing benchmark: When benchmarking a component which use gc or similary memmory handling methods, remember to have the test dataset be large enough that you cause enough gc cycles to make the performance of any single cycle noise.
And to be fair to the gc language, set minimum memmory=maximum memmory, so it will use as much memmory as you allow and don't waste time allocating more memmory.
Gc is more effective, the more memory you allow it to use, because the runtime cost of gc mostly depend on the number of live objects, not the number of allocated objects.
Yes but if you use an Relationel to OO mapper such as Torque, or hibernate you loose much of the flexibility of sql, without getting the benefit of an oo database. So you get the worst of both worlds. If you know your database are going to store objects you might as well just use an oo database instead of an relational database with an adaptor.
The main reason for using a relationel databases with an OO-adaptor(I use torque) is that the relationel databases have used the last 20 years for tuning, so they are stable and quite fast.
My Biggest problem with sql is that i have defined my database as a graph, but I newer get data back as a graph. Example for slashdot.org:
There are users, who can post stories and there are comments to each story. Imagine that I want all users that have postet a story, where a comment contains the word Amiga. And I also want the stories, and the comments that contain the word. With sql that's easy, I just do a join between user and story, and between story and comment and take the comments that contain the word amiga.
No problem, EXCEPT that what I get back is a bag of data with no specific order(And with much redundancy). That's not what I want.
What I want is a list of users, and for each user I want the stories that he have submitted with at least one comment that contain the given word. And for each story I want the all the comments that contains the word. The sql does contain all the information i need, but extracting it is quite some work.
My database is defined as a graph where foreign keys are the links, so why does sql(And relational algebra) insist on not using this graph when returning data?
One can hope, but the current battle.net is unable to handle this situation at all.
The problem is that the battle.net server don't know my internal ip address, and it don't know the internal ip address of other users behind the same nat. This can be solved but it's not that simple. (Or maybe it is, you just need to let the client upload the internal address to the server, so it can distribute it the other clients). I wonder why they have not implemented this yet.
Slashdot Mirror
Well, unless you plan to invent a time machine and die in the past, the odds of you living when this scheme gets implemented are pretty good, because it have already been implemented here in Danmark, where all current danish cards does have a chip. And the solution to backward compability is quite simple. All cards and card-readers include both the old and new solution.
But the banks have issued new cards to all users, and required all atms to be able to read the chip. So the backward compability is currently only used with foreign cards.
If you did it yes, but if you instead told the Police that someone had bought stolen goods, then the police would come and get it back.
But imagine the stolen goods was a toy aircraft which could fly home to the real owner. In that case activating the "Fly home" button would be ok, but breaking into the house, and taking the aircraft back would not.
But the problem is that the user had no right to the book in the first place, because it was stolen property(Amazon stole it, from the people who had really, the right to sell it).
Lets take an example from the physical world. If I steal 1000 printed books of "1984" and sell them to you, then you have no right to keep them, even thou you paid me.
So Amazon are fucked, no matter what they do. If they did nothing, the result would be a court order which would order them to get all the distributed copiers back. And if Amazon waited for that court order, the headline would instead be
"Amazon sell stolen books, and refuse to pay the publisher", not a good position to be in from Amazon either.
Except for the fact, that this would require all their 50 million users to upgrade their Skype software. Because Ebay can't make an compability version of the prodotol due to patents.
(And many of those skype installs are on mobile phones, where an upgrade may not be that easy for most users).
No, because the studio did not look for "health benefits", it looked for "Nutritional value".
What people said was that "no one is going to make money developing FLOSS software", and it's almost as true now as it was then. Very few(I can't name any from the top of my head) companies have developed open-source software, and then earned enough money on that(Including support) to pay for the development of that software.
Remember that >98% of the software included with "Redhat Linux" have been developed and paid for by companies other then Redhat. (Some of it, by stupid investors in the .com crash)
Well, most of the things are already supported by compilers, or will be soon. (Se forexample http://gcc.gnu.org/gcc-4.5/cxx0x_status.html for the list of that gcc 4.5 (The newest released gcc version) supports. So it is likely that c++0x will be supported almost fully, within a year of release.
(Microsoft and intel have also implemented much of the standard).
The only thing I miss from the gcc is an implementation of "Lambda expressions" and they are working on those. (They have a branch which afair kinda works :}
And it's called synthesiagame now. Something about Activision not liking them using the name "Piano hero"
And the good thing is, it runs perfect in linux using wine:-} (Just remember to uninstall pulseaudio if you use a recent "fedora core" based distro, and need to use timidity for midi.
Is it possible to use Google wave for all users now? wave.google.com just have a "notify me when it's done" button. Is there any way to download it now, or is it only for for a few select developers yet?
So instead of just writing a gui(And a driver to do mesh network), they wrote almost an entire Operation system. Sound like a giant case of "Not invented here" syndrom.
Reading this quote made me think:
"But what we did...was we had Sugar do the power management, we had Sugar do the wireless management--it became sort of an omelet. The Bios talked directly with Sugar, so Sugar became a bit of a mess."
I always thought that sugar was the graphics interface(Window manager, desktop and widgets, something like kde or gnome, but lighter).
So is he confused about what sugar is, or is sugar really a kind of mini os, operating on a linux kernel? Why should sugar for example include a wireless driver as part of sugar, instead of including the wireless driver as a standard linux driver, and then letting sugar communicate with it, as it would with any other driver.
The compiler is allowed to assume that no other code(Including no other thread, running the same code) change the value of a variable behind its back(As long as the variable it not volatile(Volatile got it's own can of worms)), so the optimization is safe.
so in the code
int *data=myFunc(); // The compiler is allowed to optimize this call out.
val=*data;
printf("%d\n",val);
val=*data;
printf("%d\n",val);
And the compiler is allowed to turn this into nothing:
int *val=myFunc(); // Returns a valid pointer to an int. // The compiler may remove anything in here.
*val=10;
if(*val==11) {
*val=42;
}
If multi-threaded code were allowed to do what you describe, then it would be impossible to do most of the optimizations that good c compilers do.
If gcc had a specific case to find
use of pointer, and then later test of same pointer for 0
That would be evil and useless(And should give a warning).
But what most likely happens is that gcc have one optimization pass, that analyse pointer values. And this detect(And remember) that tun does point to a valid object in the entire block. (Or that the code is invalid, but the compiler always assume that the code is valid).
Then later, the compiler run an other optimization pass, to remove code with no effect. It then look at the if(tun==null) test, and see that a previous pass found out that tun would always point to a valid object. Thus the test is redundent and removed.
But it's (tun->sk) not &(tun->sk). That is: The code is looking at the value of the member sk in the struct pointed to by tun. Looking at this value is undefined if tun is null.
It does not take the address of tun or tun->sk.
This sound more like a gcc/embeded os bug. There is no requirement in c/c++ that the null pointer is (int)0. That is: It don't have to be all 0 bits.
It just need to be distinct from any valid pointer, so if you run on a platform where you use memory address 0(Valid, but still wierd), you need
to config gcc(If possible, I don't know if gcc supports this) to use an other bit pattern as null pointer(say 0xeffff), and then you need to configure your embeded os, to never
return a memory address/buffer that contain the address you have used as null pointer.
I think the compiler is correct. If tun is null, then tun->sk is undefined and the compiler can do what even optimization it want.
So when the compiler see tun->sk it can assume that tun is not null, and do the optimization, because IF tun is null, then the program is invoked undefined behavier, which the compiler don't have to preserve/handle. (How do you keep the semantic of an undefined program??)
What's wrong with the way it scale text? (You do know that the zoom menu have a "zoom text only" which sound exactly like what you want.
But who should fork MySQL? >80% of the development is made my the MySql inc company(now sun, soon oracle) so there is not many developers left to do a fork.
But how the hell do they/you identify good drives? If they produce 100000 harddisks of the same model, how do you identify the good one?
Yes but what should the Mozilla developers do when they run on a system such as Linux which can't legally support H.264 in USA and other parts of the world?
Should they just do a "Not support, please install Windows?"
Re-reading my text: Nope I don't mention mysql. Most likely because I like PostgreSQL much more, and only really use Mysql on legacy systems. (The kind I take over from someone else, because the fired the original developer).
But can you clerify on the
PostgreSQL has no problem returning to me a list of rows contained in a single column of a single row in a result. The tree you are looking for is easy to return, I have stored procedures that work this way myself. It may be PostgreSQL specific,
I looked in the manual, but I could not find anything doing what you described above. Got any sql samples or tutorials that explain how you do it?
It may be a bit faster on searching, but it take ~5 times as long to generate the index, and use twice as much memmory when searching so it may just be a different trade off between index time and search time.
And it's a bad search test, because the total search time is less them 2 seconds, thus not including the cost of the gc for java.
hint to people doing benchmark: When benchmarking a component which use gc or similary memmory handling methods, remember to have the test dataset be large enough that you cause enough gc cycles to make the performance of any single cycle noise.
And to be fair to the gc language, set minimum memmory=maximum memmory, so it will use as much memmory as you allow and don't waste time allocating more memmory.
Gc is more effective, the more memory you allow it to use, because the runtime cost of gc mostly depend on the number of live objects, not the number of allocated objects.
Yes but if you use an Relationel to OO mapper such as Torque, or hibernate you loose much of the flexibility of sql, without getting the benefit of an oo database. So you get the worst of both worlds. If you know your database are going to store objects you might as well just use an oo database instead of an relational database with an adaptor.
The main reason for using a relationel databases with an OO-adaptor(I use torque) is that the relationel databases have used the last 20 years for tuning, so they are stable and quite fast.
My Biggest problem with sql is that i have defined my database as a graph, but I newer get data back as a graph. Example for slashdot.org:
There are users, who can post stories and there are comments to each story. Imagine that I want all users that have postet a story, where a comment contains the word Amiga. And I also want the stories, and the comments that contain the word. With sql that's easy, I just do a join between user and story, and between story and comment and take the comments that contain the word amiga.
No problem, EXCEPT that what I get back is a bag of data with no specific order(And with much redundancy). That's not what I want.
What I want is a list of users, and for each user I want the stories that he have submitted with at least one comment that contain the given word. And for each story I want the all the comments that contains the word. The sql does contain all the information i need, but extracting it is quite some work.
My database is defined as a graph where foreign keys are the links, so why does sql(And relational algebra) insist on not using this graph when returning data?
One can hope, but the current battle.net is unable to handle this situation at all.
The problem is that the battle.net server don't know my internal ip address, and it don't know the internal ip address of other users behind the same nat.
This can be solved but it's not that simple. (Or maybe it is, you just need to let the client upload the internal address to the server, so it can distribute it the other clients). I wonder why they have not implemented this yet.