Read what the decision actually says. The source code and object code which implement a system are covered by copyright, but the interfaces and algorithms implemented by that source code are not.
It's a fine distinction, but it essentially says that if you can reverse engineer the requirements of an API by observing it's behaviour, you are free to re-implement that functionality. i.e. You have to use so-called "clean-room" techniques, where the team that did the functional analysis of the APIs to write the specs have absolutely nothing else to do with the team that writes the implementation.
That didn't take much searching. NTM seems to be the most highly recommended utility for doing this.
But if you install it, make sure you configure it. By default it will auto-disconnect under a number of conditions, which could leave some users tearing their hair out and thinking their network is down.
Does anyone know of any metering applications for Ubuntu 10.04.1?
I'm curious as to how much bandwidth I actually consume. SaskTel has no download caps, and I do download a LOT, but I'm curious as to whether I would exceed Comcast's cap or not.
I download every TV show that catches my interest, and watch maybe an hour to two per week out of all those downloads. I download every movie that hits the Pirate Bay's top 100, and watch maybe 1-2 per month at most.
As literally 95% of what I download is "just in case" I want to watch it later, I think I've probably got about the maximum downloads a single user could possibly put on the network.
So if I'm not exceeding the 250Gb cap, I'd have to say the furor over the cap is a tempest in a teapot.
This has been bugging me all day. It really pisses me off when people try to tell me what I mean. I know what I mean. I say what I mean.
The fact that "patriotic Americans" don't like it is their problem. But watch those of them with mod points mod this psot into oblivion, because they think it's a "disagree" to vote things down. Which only proves my point...
When I say the US has a navel gazing, we're superior, our law should trump all others attitude, I MEAN IT.
Your government.
Your banks.
Your MPAA/RIAA.
Your businesses.
Your pharmacorps.
And the list goes on...
Your whole nation's MENTALITY is that you're superior.
You are the very DEFINITION of a fascist country which engenders and encourages blind, national fervour and faith in the waving flag of the nation above all else.
There are many in the country who do not feel that way, and understand what it means to cooperate with the world instead of trying to dominate it.
But apparently there aren't enough of them VOTING.
The Canadian equivalent to ASCAP is so anal retentive about collecting their royalties that they PAY someone to attend every concert to make sure that only the songs on the play list are played, including encores. If anything other than the listed songs is played and the band itself didn't write the track, they get charged a fee.
I forget the name of the organization off hand, but I've known a few people over the years who worked for them monitoring concerts. From their perspective, they got to see the concert for free -- all they had to do was write down the title of each track as it was played.
This sounds an awful lot like the arguments that AT&T brought forth against Berkeley regarding the Unix System V vs. BSD arguments back in the early '80s.
AT&T rightfully lost those arguments, and BSD moved forward.
If Oracle succeeds with this, you can expect whoever holds the AT&T copyright nowadays to come after Linux and other Unix-like systems again, despite them just following what are now documented standards accepted by the industry.
Oracle's arguments should be rejected for the same reasons as AT&T's.
I was under the impression that Google had used the Java GPL source to compile their core jars. I later had it clarified that such is not the case; they used the Apache source.
A decision in favour of Oracle would throw the entire computing industry overboard and cause no end of harm to the industry.
There is nothing like a know-it-all anonymous troll who doesn't realize that neither the Java nor the.Net documentation make any reference to RFC4122, nor do they explain how collisions are avoided (if indeed they are.)
Or are people supposed to just osmotically absorb the RFC numbers without anyone bothering to UPDATE THE FUCKING DOCUMENTATION??!?!?
That's exactly what I was hoping to be able to do -- treat both UUID and GUID generated values as being truly unique, so that client systems could assign an id to each request that they submit to a server, as well as using them to tag server nodes, processes, etc. within a distributed environment.
But I think the only way I could use this approach safely is to code the UUID generators in C#/.Net and use those rather than the actual GUID values in.Net.
As low as the chances of collisions are, there are a lot of algorithms in use that assume a UUID/GUID is truly unique and that would cause data corruption, loss, or data security issues if the value set isn't truly distinct.
Given that both use the same format, the GUID could be used as the storage type in the application code; you'd just need to use the UUID-based C# id allocation functions/algorithms to produce the value. Each of the two can deal with the string-to-*ID parsing just fine, so maybe this is the best compromise for me to look at using.
How the hell did I wander so far off the beaten path of the main article's topic?
Oh yeah -- the idea of universal devices communicating with each other when the implementation standards conflict with the idea of them being truly universal.
But my understanding is the actual VALUES of those masking patterns differ, so there is the possibility of a GUID overlapping with a UUID, even though they both use the same string format for printing.
There's more to a value than it's display format, and the fact that the two id generation algorithms are different is where I see the possible data conflicts arising in production systems. Despite the sameness of the string format of GUIDs and UUIDs, I think you'd be risking some issues if you allow.Net/GUID systems and Java/C/C++/Unix/UUID systems to cohabit the same id space.
That super computer/cluster market is precisely why I would have thought there would be a market for super-bandwidth CPUs. Such systems tend to use the highest of the high end processors already, along with custom memory interfaces and backbones to speed up the communications within the cluster.
Some posters seem to have assumed I was talking about PCs. I specifically said CPU because I wasn't concerned about maintaining compatibility with desktop architectures, but the really big data crunching engines that live in data centers and labs.
I would have thought publishing the proposed patents was being done intentionally to allow submission of prior art notices so bad patents can be blocked before it costs someone millions of dollars to defend against it in court.
Once approved, the new patent holder can go against everyone who copied the patent. And if the only way they can make money off the idea is to license the patent, then they haven't INVENTED something, have they? You need to be able to develop PRODUCTS for a patent to be useful or valid.
Not just bullshit like the fact that one click on a button does something (Amazon) -- exactly like the UI specs for buttons intended in the first place!
Can someone explain to me why general purpose CPU-memory interfaces don't have this kind of bandwidth to keep the newer 6 and 8 core monsters well fed with data and code to crunch?
RedHat neither owns nor pays sponsorship to most of the projects they contribute to.
I'm thinking in terms of groups like Mozilla which receive actual cash funding from companies like Google for their work, not companies which have found a way to leverage the work of others to turn a profit. Or the actual Apache projects, which receive funding from some of the biggest IT firms in the world to develop and maintain "core infrastructure" components for the web world. Another one that comes to mind is Eclipse, with their IBM sponsorship.
Projects like the core GNU tools, on the other hand, were developed without the goal of sponsorship or turning a profit on the tools themselves.
I'm not saying either model or license is "better". Just wondering about the chicken-and-egg question: Did they receive funding because of the license, or did they choose the license because of the funding and influence of the companies involved?
While trying to get some OS installed that includes Gnome, but which allows me to use KDE as the default so I can do configuration and stuff (knowing full well that Gnome doesn't work on my hardware), I ended up redoing the Ubuntu 12.04 installation. This time I didn't let it automatically download updates, so the software manager worked. Sort of. After a minute or two, it stopped registering clicks on the software manager widgets, but the main Unity display manager was still responding to clicks (unlike what happens with Gnome.)
Regardless, I eventually fiddled enough to get KDE installed, and it's been rock-solid.
My main issue now is getting the alien installation of Oracle XE to run. It installs ok, but the configuration fails because it thinks Oracle isn't running. Which is a little bizarre, because the TNS listener is left running when the oracle-xe configure falls over. *shrug*
While the Unity installer and desktop ran fine, the software package manager kept crashing. That kept me from installing Gnome, KDE, and a host of other software that I need.
Maybe 12.04.1 will work.
In the meantime, I'm trying to make time to work with the GTK team to debug the problem. My test partition of Ubuntu 12.04 can't be used for the testing, so I'll have to reinstall Debian 6 at some point in time. But not tonight.:)
Slashdot Mirror
Were this applied to the Oracle-Google Java case, I think this basically says the EU would rule in Google's favour.
Read what the decision actually says. The source code and object code which implement a system are covered by copyright, but the interfaces and algorithms implemented by that source code are not.
It's a fine distinction, but it essentially says that if you can reverse engineer the requirements of an API by observing it's behaviour, you are free to re-implement that functionality. i.e. You have to use so-called "clean-room" techniques, where the team that did the functional analysis of the APIs to write the specs have absolutely nothing else to do with the team that writes the implementation.
That didn't take much searching. NTM seems to be the most highly recommended utility for doing this.
But if you install it, make sure you configure it. By default it will auto-disconnect under a number of conditions, which could leave some users tearing their hair out and thinking their network is down.
netramon.sourceforge.net
Does anyone know of any metering applications for Ubuntu 10.04.1?
I'm curious as to how much bandwidth I actually consume. SaskTel has no download caps, and I do download a LOT, but I'm curious as to whether I would exceed Comcast's cap or not.
I download every TV show that catches my interest, and watch maybe an hour to two per week out of all those downloads. I download every movie that hits the Pirate Bay's top 100, and watch maybe 1-2 per month at most.
As literally 95% of what I download is "just in case" I want to watch it later, I think I've probably got about the maximum downloads a single user could possibly put on the network.
So if I'm not exceeding the 250Gb cap, I'd have to say the furor over the cap is a tempest in a teapot.
This has been bugging me all day. It really pisses me off when people try to tell me what I mean. I know what I mean. I say what I mean.
The fact that "patriotic Americans" don't like it is their problem. But watch those of them with mod points mod this psot into oblivion, because they think it's a "disagree" to vote things down. Which only proves my point...
When I say the US has a navel gazing, we're superior, our law should trump all others attitude, I MEAN IT.
Your government.
Your banks.
Your MPAA/RIAA.
Your businesses.
Your pharmacorps.
And the list goes on...
Your whole nation's MENTALITY is that you're superior.
You are the very DEFINITION of a fascist country which engenders and encourages blind, national fervour and faith in the waving flag of the nation above all else.
There are many in the country who do not feel that way, and understand what it means to cooperate with the world instead of trying to dominate it.
But apparently there aren't enough of them VOTING.
I find it interesting that you would call someone you don't trust and know "friend."
The people I call "friend" (as opposed to what Facebook says) are trustworthy. I wouldn't hang around them if they weren't.
The Canadian equivalent to ASCAP is so anal retentive about collecting their royalties that they PAY someone to attend every concert to make sure that only the songs on the play list are played, including encores. If anything other than the listed songs is played and the band itself didn't write the track, they get charged a fee.
I forget the name of the organization off hand, but I've known a few people over the years who worked for them monitoring concerts. From their perspective, they got to see the concert for free -- all they had to do was write down the title of each track as it was played.
The government is elected by the people.
Stop pretending it's not your fault. You, the people, are the ones who put up with their schite.
It's not quite the same thing as SCO.
SCO was claiming Linux had copied code from SVR.
Oracle's claim would make the POSIX APIs illegal, regardless of who or how the code to implement those APIs was written.
This sounds an awful lot like the arguments that AT&T brought forth against Berkeley regarding the Unix System V vs. BSD arguments back in the early '80s.
AT&T rightfully lost those arguments, and BSD moved forward.
If Oracle succeeds with this, you can expect whoever holds the AT&T copyright nowadays to come after Linux and other Unix-like systems again, despite them just following what are now documented standards accepted by the industry.
Oracle's arguments should be rejected for the same reasons as AT&T's.
I was under the impression that Google had used the Java GPL source to compile their core jars. I later had it clarified that such is not the case; they used the Apache source.
A decision in favour of Oracle would throw the entire computing industry overboard and cause no end of harm to the industry.
Woz was the hardware guru who created the original Apple computers; Jobs was the marketing guy who got them in people's hands.
But as a true "geek", Woz has the decency to respect his competition, and no shame in giving them praise when it's due.
It's a shame so-called "businessmen" couldn't be as generous in their dealings with the competition.
The US has declared wars on drugs, terrorism, copyright violations, crackers, and a whole host of other things.
In doing so they've declared "war" on pretty much every nation in the world, including the very ones that they claim are friends and allies.
So what can we conclude?
The US is Enemy #1 to the world.
There is nothing like a know-it-all anonymous troll who doesn't realize that neither the Java nor the .Net documentation make any reference to RFC4122, nor do they explain how collisions are avoided (if indeed they are.)
Or are people supposed to just osmotically absorb the RFC numbers without anyone bothering to UPDATE THE FUCKING DOCUMENTATION??!?!?
That's exactly what I was hoping to be able to do -- treat both UUID and GUID generated values as being truly unique, so that client systems could assign an id to each request that they submit to a server, as well as using them to tag server nodes, processes, etc. within a distributed environment.
But I think the only way I could use this approach safely is to code the UUID generators in C#/.Net and use those rather than the actual GUID values in .Net.
As low as the chances of collisions are, there are a lot of algorithms in use that assume a UUID/GUID is truly unique and that would cause data corruption, loss, or data security issues if the value set isn't truly distinct.
Given that both use the same format, the GUID could be used as the storage type in the application code; you'd just need to use the UUID-based C# id allocation functions/algorithms to produce the value. Each of the two can deal with the string-to-*ID parsing just fine, so maybe this is the best compromise for me to look at using.
How the hell did I wander so far off the beaten path of the main article's topic?
Oh yeah -- the idea of universal devices communicating with each other when the implementation standards conflict with the idea of them being truly universal.
But my understanding is the actual VALUES of those masking patterns differ, so there is the possibility of a GUID overlapping with a UUID, even though they both use the same string format for printing.
There's more to a value than it's display format, and the fact that the two id generation algorithms are different is where I see the possible data conflicts arising in production systems. Despite the sameness of the string format of GUIDs and UUIDs, I think you'd be risking some issues if you allow .Net/GUID systems and Java/C/C++/Unix/UUID systems to cohabit the same id space.
That super computer/cluster market is precisely why I would have thought there would be a market for super-bandwidth CPUs. Such systems tend to use the highest of the high end processors already, along with custom memory interfaces and backbones to speed up the communications within the cluster.
Some posters seem to have assumed I was talking about PCs. I specifically said CPU because I wasn't concerned about maintaining compatibility with desktop architectures, but the really big data crunching engines that live in data centers and labs.
I would have thought publishing the proposed patents was being done intentionally to allow submission of prior art notices so bad patents can be blocked before it costs someone millions of dollars to defend against it in court.
Once approved, the new patent holder can go against everyone who copied the patent. And if the only way they can make money off the idea is to license the patent, then they haven't INVENTED something, have they? You need to be able to develop PRODUCTS for a patent to be useful or valid.
Not just bullshit like the fact that one click on a button does something (Amazon) -- exactly like the UI specs for buttons intended in the first place!
They're not the first to dream of embedded smart devices. But Java ME owns a huge chunk of that market, from Blu-Ray players on up.
One thing I learned that ticks me off to no end is Microsoft intentionally made the GUID incompatible with the UUID.
What, pray tell, was wrong with the UUID standard other than Microsoft wanting to yet again try to lock customers in with incompatibilities?
Can someone explain to me why general purpose CPU-memory interfaces don't have this kind of bandwidth to keep the newer 6 and 8 core monsters well fed with data and code to crunch?
RedHat neither owns nor pays sponsorship to most of the projects they contribute to.
I'm thinking in terms of groups like Mozilla which receive actual cash funding from companies like Google for their work, not companies which have found a way to leverage the work of others to turn a profit. Or the actual Apache projects, which receive funding from some of the biggest IT firms in the world to develop and maintain "core infrastructure" components for the web world. Another one that comes to mind is Eclipse, with their IBM sponsorship.
Projects like the core GNU tools, on the other hand, were developed without the goal of sponsorship or turning a profit on the tools themselves.
I'm not saying either model or license is "better". Just wondering about the chicken-and-egg question: Did they receive funding because of the license, or did they choose the license because of the funding and influence of the companies involved?
Apache licenses are favoured by projects which have corporate sponsorship and funding.
GPL licenses are favoured by "grass roots" efforts which have no funding.
The question is the funding a cause or an effect of the choice of license?
While trying to get some OS installed that includes Gnome, but which allows me to use KDE as the default so I can do configuration and stuff (knowing full well that Gnome doesn't work on my hardware), I ended up redoing the Ubuntu 12.04 installation. This time I didn't let it automatically download updates, so the software manager worked. Sort of. After a minute or two, it stopped registering clicks on the software manager widgets, but the main Unity display manager was still responding to clicks (unlike what happens with Gnome.)
Regardless, I eventually fiddled enough to get KDE installed, and it's been rock-solid.
My main issue now is getting the alien installation of Oracle XE to run. It installs ok, but the configuration fails because it thinks Oracle isn't running. Which is a little bizarre, because the TNS listener is left running when the oracle-xe configure falls over. *shrug*
Ubuntu 12.04 is a wash for me.
While the Unity installer and desktop ran fine, the software package manager kept crashing. That kept me from installing Gnome, KDE, and a host of other software that I need.
Maybe 12.04.1 will work.
In the meantime, I'm trying to make time to work with the GTK team to debug the problem. My test partition of Ubuntu 12.04 can't be used for the testing, so I'll have to reinstall Debian 6 at some point in time. But not tonight. :)
DEC - Digital Equipment of China.
Ask a Canadian if they feel free.
Ask a Swede if they feel free.
Ask a Norweigan if they feel free.
They'll bitch about the high taxes, but they won't bitch about a lack of health care.
Where is the freedom in knowing the bank can jack your rates and take not only your home, but everything you've paid into it so far?
Where is the freedom in knowing that once $1M has been spent on your child's medical treatments, your only option is bankruptcy?
Where is the freedom in knowing that despite a college degree, your only hope for a job is flipping burgers?
Socialist systems still use money. You still need a job. You still have options as to where to spend that money.
You're thinking communism, where the state controls everything from your career to your housing options.
If you're going to argue politics, you'd best learn what the terms mean before you open your mouth.