It is in some states that are two party consent states (it's not the video, it's the audio). Even in one party states people get charged with breaking wiretapping laws sometimes. An explicit law that permitted this would be a very good thing, but it won't happen especially with Republicans in charge of one house of Congress.
"If I've been following the case correctly, the states demanding action are states where Amazon has a business presence and a customer."
RTFA.
FTA: “We’re no different from other big chains of retailers,” Bezos said. “They don’t collect sales taxes in states where they don’t have [employees], either.”...
First of all, most of where we do business — Europe, Japan, some of the states here in the United States – we collect sales tax.
Depends on what the software is doing. If it's controlling my car or my mom's pacemaker I want the developers having a passion for making sure the code is 100% correct.
"The thing is, JavaScript is inherently source visible."
Ever try to comprehend highly minified JavaScript? It's difficult at best. Keep in mind it's not just white space and comments, it's also things like symbols getting renamed to very short (not meaningful) names. Is it possible to figure out? Yes. Is it easier. Not at all.
""Derivative Works" shall mean any work, whether in Source or Object form, that is based on (or derived from) the Work and for which the editorial revisions, annotations, elaborations, or other modifications represent, as a whole, an original work of authorship. For the purposes of this License, Derivative Works shall not include works that remain separable from, or merely link (or bind by name) to the interfaces of, the Work and Derivative Works thereof."
The thing about attributions and Apache License are at least part BS. The Apache license (which I just re-read) only requires attributions when a DERIVATIVE work is distributed. In most cases, I'm betting that companies are not distributing derivative works, but the original work. It's a hole in the license, but that't not the user's fault.
Despite the F designation, the F-117 is an attack aircraft, not a fighter. The F-35 is a multi-role aircraft, ala the F-16. The F-22 is a true air superiority fighter, ala the original F-15. That' why we've scaled back F-22 production and ramped up F-35 production. Hasn't been much need for a U.S. air superiority fighter in the last 20 years. On other hand, we've needed lots of attack aircraft in Iraq and Afghanistan.
All of this stuff should count as an interface, and therefore not covered by copyright under US law.
Yes and no. With Java, the package, class and method names would be a part of the interface and thus not covered. However, using the symbol XYZ to be some specific value that is used as a parameter to a method would be covered. It's the difference between an interface and an expression of an interface. Of course, this would be a small amount of what Oracle is claiming, but it's still valid.
The code copying is the thing that is most potentially damaging. And that all depends on what code was copied (if it was) and what the license for that code was.
It probably has to do with the queue depth on the disk drives.
With a write task in the background it could easily fill the queue on the disk drive. Drives often use an elevator algorithm to fulfill I/Os. So, if the file being written by the background task is written sequentially on disk all of those I/Os will be completed before a read from another location on the disk gets performed. It potentially is much worse than that, suppose the write task fills the queue (e.g. 32 sequential I/Os). One of those gets completed and a read from the foreground process gets inserted into the queue. It very well could be for a location with a lower LBA on the disk than the writes so it's not going to get processed right away because of the elevator algorithm. The drive completes another write from the background task --- and guess what? Another write comes down for the same file. Because it's got a higher LBA it's going than the read it's going to be processed before the read because of the elevator algorithm. Another write comes down, same thing. Another write, same thing.
If the data from the file being written is on the same disk then there's a good chance that a read for that file will break this chain and the read from the foreground process will be performed. However, if the file data is coming from another disk or from a network then that foreground process could be starved.
Variations of the elevator algorithm can prevent total starvation, but it is designed to be maximize total throughput.
This is not a thread scheduling issue, it's a disk scheduling issue. If CPU utilization is only 1-2% and things aren't snappy then the issue is because the foreground process's I/Os aren't given higher (high enough?) priority. Easy enough to believe too, a whole lot of writes get cached and then queued up. With an elevator algorithm they'll likely all get performed before any reads required by the foreground process.
For purposes of U.S. export controls, a hash function is considered encryption software. There's not a system in the world that provides a login capability that doesn't at least do hashing.
This is what happens when you have non-technical people making rules for the use of technology.
Yet another overreaction by the paranoid. Personally, I'm sick and tired of these paranoid "security" people trying to take away every little piece of information that some terrorist could conceivably use. Living in a free society entails a certain amount of risk. Deal with it.
"Everyone runs Linux and the McAfee investment turns out to be a boondoggle."
ROFLMAO!!!
Dude, seriously. You need to start scanning the CVEs that come out for all the ones that apply to Linux.
Anybody who's worked at both ends knows this. Big companies don't do much organic growth because the small markets don't generate enough income and it's too hard to know what the big markets will be 3-5 years down the road. It's cheaper and safer just to let the market place itself out and then buy a promising company rather than invest in developing something and then probably end up buying a promising company because they have better technology or started from a somewhat different premise.
A perfectly reasonable assumption. Take a look at the CVE list some time and understand that most of those vulnerabilities were not being exploited when they were discovered.
Slashdot Mirror
"It's not illegal to film them"
It is in some states that are two party consent states (it's not the video, it's the audio). Even in one party states people get charged with breaking wiretapping laws sometimes. An explicit law that permitted this would be a very good thing, but it won't happen especially with Republicans in charge of one house of Congress.
"If I've been following the case correctly, the states demanding action are states where Amazon has a business presence and a customer."
RTFA.
FTA: “We’re no different from other big chains of retailers,” Bezos said. “They don’t collect sales taxes in states where they don’t have [employees], either.” ...
First of all, most of where we do business — Europe, Japan, some of the states here in the United States – we collect sales tax.
Depends on what the software is doing. If it's controlling my car or my mom's pacemaker I want the developers having a passion for making sure the code is 100% correct.
You forgot the /s at the end.
Within a couple of months, ALL the companies will have this provision in their contracts. Then there is no choice if you want the service.
Some "free" market.
Which is probably the goal of the five conservatives on the court.
Perhaps Congress will change the statute.
"The thing is, JavaScript is inherently source visible."
Ever try to comprehend highly minified JavaScript? It's difficult at best. Keep in mind it's not just white space and comments, it's also things like symbols getting renamed to very short (not meaningful) names. Is it possible to figure out? Yes. Is it easier. Not at all.
That's ONE.
Bottom line is it's hard to make a lot of money in open source. Great for consumers, bad for investors.
ROFL
Dude, read the license:
""Derivative Works" shall mean any work, whether in Source or Object form, that is based on (or derived from) the Work and for which the editorial revisions, annotations, elaborations, or other modifications represent, as a whole, an original work of authorship. For the purposes of this License, Derivative Works shall not include works that remain separable from, or merely link (or bind by name) to the interfaces of, the Work and Derivative Works thereof."
And you might comprehend better next time. Giving someone a copy of the license is not the same thing as giving attributions.
The thing about attributions and Apache License are at least part BS. The Apache license (which I just re-read) only requires attributions when a DERIVATIVE work is distributed. In most cases, I'm betting that companies are not distributing derivative works, but the original work. It's a hole in the license, but that't not the user's fault.
Despite the F designation, the F-117 is an attack aircraft, not a fighter. The F-35 is a multi-role aircraft, ala the F-16. The F-22 is a true air superiority fighter, ala the original F-15. That' why we've scaled back F-22 production and ramped up F-35 production. Hasn't been much need for a U.S. air superiority fighter in the last 20 years. On other hand, we've needed lots of attack aircraft in Iraq and Afghanistan.
The first two questions in particular are certainly not applicable in many environments.
All of this stuff should count as an interface, and therefore not covered by copyright under US law.
Yes and no. With Java, the package, class and method names would be a part of the interface and thus not covered. However, using the symbol XYZ to be some specific value that is used as a parameter to a method would be covered. It's the difference between an interface and an expression of an interface. Of course, this would be a small amount of what Oracle is claiming, but it's still valid.
The code copying is the thing that is most potentially damaging. And that all depends on what code was copied (if it was) and what the license for that code was.
It probably has to do with the queue depth on the disk drives.
With a write task in the background it could easily fill the queue on the disk drive. Drives often use an elevator algorithm to fulfill I/Os. So, if the file being written by the background task is written sequentially on disk all of those I/Os will be completed before a read from another location on the disk gets performed. It potentially is much worse than that, suppose the write task fills the queue (e.g. 32 sequential I/Os). One of those gets completed and a read from the foreground process gets inserted into the queue. It very well could be for a location with a lower LBA on the disk than the writes so it's not going to get processed right away because of the elevator algorithm. The drive completes another write from the background task --- and guess what? Another write comes down for the same file. Because it's got a higher LBA it's going than the read it's going to be processed before the read because of the elevator algorithm. Another write comes down, same thing. Another write, same thing.
If the data from the file being written is on the same disk then there's a good chance that a read for that file will break this chain and the read from the foreground process will be performed. However, if the file data is coming from another disk or from a network then that foreground process could be starved.
Variations of the elevator algorithm can prevent total starvation, but it is designed to be maximize total throughput.
This is not a thread scheduling issue, it's a disk scheduling issue. If CPU utilization is only 1-2% and things aren't snappy then the issue is because the foreground process's I/Os aren't given higher (high enough?) priority. Easy enough to believe too, a whole lot of writes get cached and then queued up. With an elevator algorithm they'll likely all get performed before any reads required by the foreground process.
For purposes of U.S. export controls, a hash function is considered encryption software. There's not a system in the world that provides a login capability that doesn't at least do hashing.
This is what happens when you have non-technical people making rules for the use of technology.
Yet another overreaction by the paranoid. Personally, I'm sick and tired of these paranoid "security" people trying to take away every little piece of information that some terrorist could conceivably use. Living in a free society entails a certain amount of risk. Deal with it.
"Everyone runs Linux and the McAfee investment turns out to be a boondoggle." ROFLMAO!!! Dude, seriously. You need to start scanning the CVEs that come out for all the ones that apply to Linux.
Anybody who's worked at both ends knows this. Big companies don't do much organic growth because the small markets don't generate enough income and it's too hard to know what the big markets will be 3-5 years down the road. It's cheaper and safer just to let the market place itself out and then buy a promising company rather than invest in developing something and then probably end up buying a promising company because they have better technology or started from a somewhat different premise.
Wow, a whole lot of speculation followed by "Seems RMS was right"...
"... since we all agreed that tab indenting for code was properly two spaces."
I've never worked anywhere where that was used.
The FBI must think that Wikipedia and Wikileaks are connected somehow.
A perfectly reasonable assumption. Take a look at the CVE list some time and understand that most of those vulnerabilities were not being exploited when they were discovered.
Yeah, help the malware writers by telling them where to look for issues.