I think it's bad; I wouldn't frequent such bars.. I think it says something about the owners, and their mentalities. That said.. as long as this is a private venture, and not a government pressured one, it's not so bad....
Music is covered by copyright law. Anything above what you are allowed to do with copyright law would require a license.
Now... the software used or the services used to get digital music, those things have licenses... but not the media itself. It's just.. the DMCA makes it illegal, often, in the US, to circumvent those schemes.... but the music itself is still covered by standard copyright rules.
and having verified on rather expensive test equipment....
100BaseT, one host can put data on the wire at well over 95Mbps... the overhead (inter-frame gap, protocol overhead of ethernet-II or 802.3, and even IP and TCP headers only add up to a small amount of the available time.
So, although TECHNICALLY the 100 refers to the bus bandwidth, not the host to host bandwidht, the practical bandwidht available is so damn close nobody really complains.
Lexmark is also willing to let you trade that freedom (something none of us have any business telling you not to do) for a significant discount.
That's business folks.. and reasonably straightforward business at that. We understand that under this scheme, they make their mone off the print cartridges, not off the printer... we also understand we are free to buy a printer at full price and do what we want with it. To buy it, claim the discount, then ask the courts to say it shoudln't count is dishonest and greedy.
As the use of DNS for looking up web sites is one of may specific applications of DNS, then looking for alternate sites and dealing with typos is OBVIOUSLY an application level problem. Build it into the browser. Verisign, if they want, could develop an extension to DNS that let a browser say "what domains are similar to this" or something.. that would be fine.. as long as they leave things work they way they are supposed to.
You hit it in your last line though.. it's not options that would have saved him... being smart and not risking all his gains in the first place would have saved him.
My point was that options are based on predicted growth... If the market moves as predicted, what you spent in options cancels out any profits..... the only benefit is if you have puts and the market doesn't meet predicted values, or calls and the market beats the predictions.
Unless his expectations were much higher than everyone elses, buying some puts would have been just as good as not putting money in in the first place.
You are SUPPOSED to be able to count on getting "DOMAIN NOT FOUND" errors.... DNS isn't google.. it's a precise, distributed database, that has served us well so far.
I have been hit by this problem already, where typos went unnoticed in scripts because a connection was made, and html returned. I've had mail problems as well, where secondary MX was never tried, because of verisign's new trick.
It's handy for when you mistype.. unfortuntaely, looking up web pages is just one of many uses for the DNS.... and not at all what it was intended for.
- There don't appear to be any options available on SCO, so you can't buy puts or calls.
- "Why you can't short stock".. . care to back that up?
- IT's not amazing at all.. are you implying that you truly understand the stock market, and know how to make money with it for real? It's never as simple as buying puts.. hindsight is 20/20. There is ALWAYS risk. He could have made oney buying puts.. or, had the value increased by the market predicted amount, wasted exactly as much money on puts as he would have made in profit... or worse.
The only people in the stock market who are making guaranteed profit are the market makers.
If I use Intel's compiler, I get different code than with GCC. If I provide binaries built with Intel's compiler... am I violating rules, because you can't produce the same binary? no.
what's teh difference between that, and soem other custom compiler only I have?
Furthermore...
We're talking about a linux kernel here... what OS, exactly, is it running on?
One could argue it just runs on the hardware.. and that any "missign" binary blocks are just that, blocks that are already a standard part of teh target platform (a linksys router)
So please answer my last question then.. at what point does the translation of C into machine code constitute a binary block? How many instructions does a compiler have to produce from a given instruction in C before you consider it "inserting a binary block" and not "compiling"?
You are misinterpreting the intent of that clause.
The intent is for you to not hold out on scripts and whatnot that are part of assembling the final product... but the compiler is not part of that deal.
What you suggest would make it imposisble to , say, port something to Delphi and distribute it, as Delphi is not a normal part of ANY platform.
Similarly, MS VC++ or Visual Basic is not a standard part of the windows platform... yet we have windows code, GPL, and compiled with it.
The GPL was not at all intended to force you to use open, freely available tools.
You do not have to give people everything they need to compile the code; you have to give them all the CODE, including supporting scripts needed to build it. Not the compiler.
Unless that binary block is considered a standard part of the target platform.
I'm not excusing linksys here.. I don't in any way think this applies to them.... this is theoretical.
There is no guarantee that the compiler in question is avialable to you.. and if that compiler happens to translate "BoogaBooga();" into a special routine the compiler knows about...... that's not something you get.
Where do you draw the line between compiling, and inserting a block of code? How many instructions does the compiler have to translate an operation into to qualify as a block of code, as opposed to just compiling?
Tell me, what is the operating system, in this case, that the linux kernel on the linksys router runs on?
This says you do NOT have to include the source for things that are a normal part of the target platform. As the target platform of the derived work is a linksys router....they don't have to provide their compilers.
The point of this clause is that you can distribute a Solaris version of a GPL program without distributing the source to the solaris kernel, or the solaris LIBC implementation.. despite the fact that the executable is statically linked to it... because those are things that are a normal part of the intended platform for the binary.
In this case, the binary is distributed with it's platform.. the router.
In your last example, a custom compiler inserting binary code, it is NOT clearly a violation.
The GPL does NOT require that you are able to compile the code... only that if you make modifications to someone's code, you release those modifications.
The GPL does not requier that you have the technical means to take and use other's code.. only that you are allowed to use it legally, and that the code is made availalbe to you.
If I port something to VC++, it is not a violation just because you don't have a copy of VC++. Using my own custom compiler you CANT get is irrelevant.
So.. the compiler-that-inserts-code example could perhaps be a way to subvert the intent of the GPL..... at least the way I read it.
There is a second subversion also.
The GPL allows you to statically link against things that are a normal part of the target system. You can write a GPL app for Solaris that is linked against Sun's LIBC, which definately isn't open source, for instance.
This, of course, is open to interpretation... the SUN example is obvoius. what if I say that somethign is built for my particular distribution of linux I use in-house, which has some commercial non-free libraries.... am I violating the GPL by distributing osmeting linked against them? Not if the intended target is that platform, and that platform only... by the letter of the GPL, that is allowed. The fact that it LOOKS like another platform that you want to use isn't relevant.
There is no requirement that you are able to compile it... but only with relation to the tools involved. For instance, if it was ported to their custom compiler which used dirctives that ours don't support, that's our problem, not theirs.
If, however, it doesn't compile because they neglected to include the full source code, that's another matter.
These guys didn't just say they can't compile it, they analyzed WHY it wouldn't compile; modules are looking for symbols that do not exist in the kernel.....that means code is taken out.
In your intercal++ example, it would not violate teh GPL in any way.
HyperSCSI on ethernet uses a EtherType field of 0x889a.
With iSCSI, you can route over an IP network to your devices.. you could have a storage subnet, for instance.
We want storage to be as fast as possible, and generally it will be local, so HyperSCSI makes more sense. It's just a new way to build a SAN rather than using Fiberchannel... now we have gigabit ethernet, and so on.
When someone says "TCP/IP" they are referring to the entire protocol suite, which involves TCP, UDP, ICMP, and other stuff on top of IP. It does not at all mean that they are using TCP itself.
When someone says "TCP/IP" they are referring not to a protocol, but a protocol suite. They are NOT referring to "TCP and UDP ONLY".
So, if you say something is not based on tcp/ip,that indeed DOES mean it does not use IP. Furthermore, saying it uses "raw ethernet" indicates that this uses it's own layer 3 protocol, other than IP.
I think if you read again, and this time don't assume every poster above you doesn't already know this, you'll see the point they were trying to make.
If this uses it's own Layer 3 protocol (presumably, and for the sake of argument, called HyperSCSI), then it's NOT IP BASED... and the article summary indicated it was IP based, then contradicted itself.
If you are a linux geek... fine.. you don't need or want to spend the extra money on OSX server.
OSX servers are not overly expensive, and are nice hardware. They come with a nice OS, with a few great admin tools with tight desktop integration.
Again, if you are a unix geek, you don't care, you want to do it all yourself.. but for an office without real unix geeks, tossing in an OSX server is an economical, powerful, easy to integrate solution, and far better than trying to use Win2k or something...
The system as shipped won't let you log in as root, or su to root (though you can use sudo)... it just means the root password is effectively unknown, or locked (I did not check which).
System services still run as root, and SSH is just as vulnerably as on any other unix.
Slashdot Mirror
I think it's bad; I wouldn't frequent such bars.. I think it says something about the owners, and their mentalities.
That said.. as long as this is a private venture, and not a government pressured one, it's not so bad....
Music is covered by copyright law. Anything above what you are allowed to do with copyright law would require a license.
Now... the software used or the services used to get digital music, those things have licenses... but not the media itself.
It's just.. the DMCA makes it illegal, often, in the US, to circumvent those schemes.... but the music itself is still covered by standard copyright rules.
and having verified on rather expensive test equipment....
100BaseT, one host can put data on the wire at well over 95Mbps... the overhead (inter-frame gap, protocol overhead of ethernet-II or 802.3, and even IP and TCP headers only add up to a small amount of the available time.
So, although TECHNICALLY the 100 refers to the bus bandwidth, not the host to host bandwidht, the practical bandwidht available is so damn close nobody really complains.
First, you have to determine what's a fair price. What's the real cost of the printer?
I'd go for a continuous flow inkjet system, perhaps...
http://www.inkbags.com/
And do whatever you like with it.
Lexmark is also willing to let you trade that freedom (something none of us have any business telling you not to do) for a significant discount.
That's business folks.. and reasonably straightforward business at that. We understand that under this scheme, they make their mone off the print cartridges, not off the printer... we also understand we are free to buy a printer at full price and do what we want with it. To buy it, claim the discount, then ask the courts to say it shoudln't count is dishonest and greedy.
As the use of DNS for looking up web sites is one of may specific applications of DNS, then looking for alternate sites and dealing with typos is OBVIOUSLY an application level problem. Build it into the browser. Verisign, if they want, could develop an extension to DNS that let a browser say "what domains are similar to this" or something.. that would be fine.. as long as they leave things work they way they are supposed to.
You hit it in your last line though.. it's not options that would have saved him... being smart and not risking all his gains in the first place would have saved him.
My point was that options are based on predicted growth... If the market moves as predicted, what you spent in options cancels out any profits..... the only benefit is if you have puts and the market doesn't meet predicted values, or calls and the market beats the predictions.
Unless his expectations were much higher than everyone elses, buying some puts would have been just as good as not putting money in in the first place.
You are SUPPOSED to be able to count on getting "DOMAIN NOT FOUND" errors.... DNS isn't google.. it's a precise, distributed database, that has served us well so far.
I have been hit by this problem already, where typos went unnoticed in scripts because a connection was made, and html returned.
I've had mail problems as well, where secondary MX was never tried, because of verisign's new trick.
It's handy for when you mistype.. unfortuntaely, looking up web pages is just one of many uses for the DNS.... and not at all what it was intended for.
- There don't appear to be any options available on SCO, so you can't buy puts or calls.
- "Why you can't short stock".. . care to back that up?
- IT's not amazing at all.. are you implying that you truly understand the stock market, and know how to make money with it for real? It's never as simple as buying puts.. hindsight is 20/20. There is ALWAYS risk.
He could have made oney buying puts.. or, had the value increased by the market predicted amount, wasted exactly as much money on puts as he would have made in profit... or worse.
The only people in the stock market who are making guaranteed profit are the market makers.
If I use Intel's compiler, I get different code than with GCC. If I provide binaries built with Intel's compiler... am I violating rules, because you can't produce the same binary? no.
what's teh difference between that, and soem other custom compiler only I have?
Furthermore...
We're talking about a linux kernel here... what OS, exactly, is it running on?
One could argue it just runs on the hardware.. and that any "missign" binary blocks are just that, blocks that are already a standard part of teh target platform (a linksys router)
We do regulate commercial advertisements... and infomercials have to disclose that they are advertisements.
Where do you draw the line between an infomercial, and a TV show with product placement?
So please answer my last question then.. at what point does the translation of C into machine code constitute a binary block? How many instructions does a compiler have to produce from a given instruction in C before you consider it "inserting a binary block" and not "compiling"?
You are misinterpreting the intent of that clause.
The intent is for you to not hold out on scripts and whatnot that are part of assembling the final product... but the compiler is not part of that deal.
What you suggest would make it imposisble to , say, port something to Delphi and distribute it, as Delphi is not a normal part of ANY platform.
Similarly, MS VC++ or Visual Basic is not a standard part of the windows platform... yet we have windows code, GPL, and compiled with it.
The GPL was not at all intended to force you to use open, freely available tools.
You do not have to give people everything they need to compile the code; you have to give them all the CODE, including supporting scripts needed to build it. Not the compiler.
Unless that binary block is considered a standard part of the target platform.
I'm not excusing linksys here.. I don't in any way think this applies to them.... this is theoretical.
There is no guarantee that the compiler in question is avialable to you.. and if that compiler happens to translate "BoogaBooga();" into a special routine the compiler knows about...... that's not something you get.
Where do you draw the line between compiling, and inserting a block of code? How many instructions does the compiler have to translate an operation into to qualify as a block of code, as opposed to just compiling?
Tell me, what is the operating system, in this case, that the linux kernel on the linksys router runs on?
This says you do NOT have to include the source for things that are a normal part of the target platform. As the target platform of the derived work is a linksys router....they don't have to provide their compilers.
The point of this clause is that you can distribute a Solaris version of a GPL program without distributing the source to the solaris kernel, or the solaris LIBC implementation.. despite the fact that the executable is statically linked to it... because those are things that are a normal part of the intended platform for the binary.
In this case, the binary is distributed with it's platform.. the router.
In your last example, a custom compiler inserting binary code, it is NOT clearly a violation.
The GPL does NOT require that you are able to compile the code... only that if you make modifications to someone's code, you release those modifications.
The GPL does not requier that you have the technical means to take and use other's code.. only that you are allowed to use it legally, and that the code is made availalbe to you.
If I port something to VC++, it is not a violation just because you don't have a copy of VC++. Using my own custom compiler you CANT get is irrelevant.
So.. the compiler-that-inserts-code example could perhaps be a way to subvert the intent of the GPL..... at least the way I read it.
There is a second subversion also.
The GPL allows you to statically link against things that are a normal part of the target system. You can write a GPL app for Solaris that is linked against Sun's LIBC, which definately isn't open source, for instance.
This, of course, is open to interpretation... the SUN example is obvoius. what if I say that somethign is built for my particular distribution of linux I use in-house, which has some commercial non-free libraries.... am I violating the GPL by distributing osmeting linked against them? Not if the intended target is that platform, and that platform only... by the letter of the GPL, that is allowed. The fact that it LOOKS like another platform that you want to use isn't relevant.
There is no requirement that you are able to compile it... but only with relation to the tools involved. For instance, if it was ported to their custom compiler which used dirctives that ours don't support, that's our problem, not theirs.
If, however, it doesn't compile because they neglected to include the full source code, that's another matter.
These guys didn't just say they can't compile it, they analyzed WHY it wouldn't compile; modules are looking for symbols that do not exist in the kernel.....that means code is taken out.
In your intercal++ example, it would not violate teh GPL in any way.
"IP-like" is not "IP"
if it doesn't use IP, it doesn't work "over IP". Either it's IP, or it is not.
By the way, HyperSCSI is a layer 3 protocol, and it has absolutely nothing to do with IP.
The article summary was mixing up iSCSI (which runs over IP only, using TCP) and HyperSCSI (which is it's own protocol)
When someone says "TCP/IP" they are referring to the entire protocol suite, which involves TCP, UDP, ICMP, and other stuff on top of IP.
It does not at all mean that they are using TCP itself.
When someone says "TCP/IP" they are referring not to a protocol, but a protocol suite. They are NOT referring to "TCP and UDP ONLY".
So, if you say something is not based on tcp/ip,that indeed DOES mean it does not use IP. Furthermore, saying it uses "raw ethernet" indicates that this uses it's own layer 3 protocol, other than IP.
I think if you read again, and this time don't assume every poster above you doesn't already know this, you'll see the point they were trying to make.
If this uses it's own Layer 3 protocol (presumably, and for the sake of argument, called HyperSCSI), then it's NOT IP BASED... and the article summary indicated it was IP based, then contradicted itself.
Ease of administration.
If you are a linux geek... fine.. you don't need or want to spend the extra money on OSX server.
OSX servers are not overly expensive, and are nice hardware. They come with a nice OS, with a few great admin tools with tight desktop integration.
Again, if you are a unix geek, you don't care, you want to do it all yourself.. but for an office without real unix geeks, tossing in an OSX server is an economical, powerful, easy to integrate solution, and far better than trying to use Win2k or something...
It's unix.. root always exists.
The system as shipped won't let you log in as root, or su to root (though you can use sudo)... it just means the root password is effectively unknown, or locked (I did not check which).
System services still run as root, and SSH is just as vulnerably as on any other unix.
Thought of it yesterday. Great idea.
The only problem is nobody sells options on sco.