Point me to a few viruses for BeOS, OS/2 / eComStation, SolarOS, or Menuet.
Android is the most hackable mobile phone OS out there? Sure, but if you're going to argue that that discredits the security of the kernel like you seem to be saying, go ahead and point out how much of that is due to kernel bugs. (As far as I can tell, the main kernel bug was Samsung's ugo+rwx access to system memory--which would only be an issue for those who haven't updated).
The real issue is twofold: First, many eyes won't do a thing to help if half the phones never get a single update after source code goes public. If you can find and fix every single bug within a month of the source drop, it will not change security for a device that's stuck with a rom from just after the code drop.
Second, notice that term "code drop". To quote Rob Landley, "Android isn't open source, it's regularly updated abandonware." "Many eyes" cannot work when there's no source before release; when you look at Honeycomb, applying it to Android becomes even more absurd.
You're forgetting Apple and their fanboys. They fully expect tablets to displace PCs entirely. It doesn't matter if the task is well suited to the tablet form factor or not.
is phrased as flamebait, the other half is worthy of a +1 in my book:
Now with an HDMI and USB port, there's no good reason one kind of general purpose computer can't act like another. The main limitation is policy and whether or not the guy you buy your device wants to try and lock it down.
If there's anything you can't do after simply connecting the right peripherals (and installing any drivers!)...the only reasonable justification is that the software hasn't been written or installed yet. "You haven't licensed that functionality yet" is not an acceptable reason. (Note that I said functionality, not software.)
Selling a license to use a device as you see fit is not the same as making it not locked down.
Just to make something clear: the opposition to transgenic organisms is not all by those who claim that it "violates the sanctity of gods creations"-the "or somesuch" can range to people who are miles away from fundamentalists. I see quite a bit from liberals who complain about doing things that "Mother Nature" didn't do. Not that there is really a difference when it comes down to the line of argument, it's just that "zealot idiots" can come from anywhere. And I also see a number of people among conservatives who are in favor of transgenics.
But really, it seems to me that the opposition would weaken when someone stops to explain things like: -the universal genetic code -the origin of the genes in transgenic organisms (other species) -why they are doing what they do -the scope of a single gene relative to the whole genome And also when they see someone they know doing it, rather than "those evil big companies". So while it could be risky to participate, it is probably the step which is most likely to result in a real dialog and understanding.
~A conservative Christian who has worked for Pioneer and has prepared genetically modified E. coli during university coursework for his ag degree.
Better yet, outlaw curtains and make all new housing construction out of clear acrylics. Then outlaw clothes and mandate full body scanners to prevent anybody from carrying a concealed weapon.
Hey! Banning clothes is restricting my religious freedom!
(Yes, the line should be drawn a lot sooner than that.)
There's also checking what level of confidentiality is involved (claim 18), what appears to be a buzzword killer (claim 16), alerting the company (claim 15), referring the employee to a company policy web page (claim 20) and some other such things.
It looks like this might be what Google needed for a certain email that got dragged out in the Oracle case. It would also be useful for any developer of business email clients or office suites...
And in my case, I live near the largest and cheapest (especially when it comes to internet) city in Northern California. Don't assume that everyone on the internet is from Northern Europe, or that internet connections are always as cheap as they are where you are.
True, but surprise at a BSD using GCC (note that the OP did not mention GPL3) is not warranted by the existing conditions (note that deprecating does not mean dropping, it means stating that they intend to drop it, at FBSD 11 or later).
Judging by the fact that Thorsten Glaser (MirBSD/MirOS developer/maintainer) is a frequent poster on the pcc-devel list, I'd guess PCC. Also, some of the other MirOS projects are in a similar vein.
I suspect that NetBSD will use whatever compiler supports more of their architectures, with a preference for Clang when it's roughly equivalent...
How about the fact that only one of the two occurred "on the grounds of Bartow High School"?
I'm going to bet that if the kid who accidentally shot his brother with a BB gun had instead brought it to school (not shot it at school, just had it in his possession on campus), he would have been in major trouble.
Not that prosecuting this is sensible, just that the real issue may be "think of the children" gone wrong again rather than race.
> Or... they're patenting it before Amazon or Apple do so in order to avoid being sued themselves. My suspicion is that you're correct--judging by Apple's past behavior, Apple would just add this to their patent on the corner of page curl animation and sue Samsung for "infringing" the patent they worked around.
> What is happening here? Heretics dare to use GNU code on a BSD system? Sacrilege!
BSD has used GNU tools for ages, but most of them have lagged behind. FreeBSD is sticking to GPL2, which is GCC 4.2.1--but you can get a newer one from ports. OpenBSD has 4.2.1 and 2.95.3. NetBSD is the other BSD to use a GPL3 GCC, with GCC 4.5 as far as I can tell. MirBSD is on GCC 3.4.
And the most fun difference: DragonFly BSD uses git as vcs, rather than cvs/svn.
> Sucks to be you. > I pay $16 for a 20Mbps DSL connection.
> I bet there are other options in your area, you are just too stupid or lazy to look around. Try your local phone company, or even sattelite provider if you don't care too much about latency.
Have you ever seen the prices for satellite? In my area it starts at $70. 10 Mbps down/1.5 Mbps up DSL for $25 here,IIRC (my brother pays for it, and it's "fast enough to use"). But the same plan would cost $35 now and that's the cheapest DSL gets here.
So here's your "clueless provincial urbanite" award.
> I bought one last year. They're still for sale and being manufactured. If it ain't broke, don't fix it.
Yup, and I've got one that's probably about 5 years old. Specifically, the WRT54G v8, which needs a micro build (2 MB flash, 8 mb ram). I'm half wondering if it would be possible to build a musl/busybox system with a newer kernel...
I read and searched TFA (net-security.org and blog.sucuri.net), and the words "sum", "hash", and "checksum" do not occur on either page.
The closest it comes is saying that the timestamp is the same as the original, and that rpm -V won't work IF you use cPanel--because that's outside the package management system.
They suggest grepping for open_tty, though it would be possible to circumvent that with upx... in which case file would report a corrupted ELF file.
Probably because it's a pretty standard target for genetic experiments.
And by the way...it may grow slow, but it reproduces pretty fast (you could get 2+ generations in a year, and it produces up to 10,000 seeds per plant...which means potentially 1 transformed plant -> 100 million in a year, though a lot of those would die in reality)
>Sorry, my facts are correct... after all we have history classes in germany as well. >But you are in so far right that indeed the south started shooting on a fortress occupied by northern troops. >However the fortress was in southern territory. No idea and too lazy now to look up how and why the north could occupy it.
It was a previously-existing Federal fort, which under the Constitution is Federal territory. And if you didn't know that, the quality of your "history" class is suspect.
Secession was motivated entirely by the fact that Lincoln was a noted abolitionist, from a party where the previous candidate had campaigned with this slogan: "Free men, free soil, and Fremont!" Anyone who says it was "states' rights" and not slavery has no clue: they are contradicting themselves, since states rights really meant "the right of states to determine the status of slavery within their own borders, to have that decision enforced against runaway slaves by every other state, and to secede (as long as they do so for the sake of protecting slavery; if it's for the sake of making slavery illegal, that's treason!)" And yes, every one of those points is illustrated by an aspect of history: the objections to the Missouri Compromise, the Fugitive Slave Act, and the Southern response to New England's threats of secession. And even if it were about secession, it's also about the reason for secession...which was very much slavery.
But the North, while not wanting slavery, did not fight for its abolition until 1863; that is correct.
Which is why Pioneer (one of the other companies that does "GMO" crops, as well as conventional) is working on corn engineered for more efficient use of nitrogen.
(Of course, this isn't something that will be acknowledged as a yield improvement; it doesn't raise the maximum "intrinsic yield", just yield when you can't put as much in. Which most GMO critics like to ignore, even if it means that the average farmer could raise more bushels with less water pollution...)
Also, more efficient use of resources allows more land to be fallowed and rotated. If you have 10% more yield you can put 9% of the land in alternate crops, such as legumes.
Good comment until you started talking about life cycles. "Life cycles" are an issue in some (read numerous) circumstances. More or less, it's a question of how much of #2 you need to do yourself. And you know that supporting LibreOffice yourself is probably not an option, but avoiding an office suite is also a bad idea. Scenarios:
1) You need to provide a device that's supported for the next 3 years. Upstream has high churn. If there's a regularly-updated stable branch/other support policy, you can just use that. If there isn't you need to backport applicable bugfixes yourself, or write your own.
2) Software is written by a freelancer in between jobs. He says nothing about support. If he finds a job, you probably will be maintaining it yourself.
3) Company offers a solution, then a year later the needed code in GTK/Perl/PHP/... gets replaced/broken/... If they support their solution for the next five years, you have nothing to worry about. If you're using Fedora and have to support it yourself, good luck.
Very good point. Also I'll look at -the last few months of commit logs--how many contributors, patch series, recurring contributors... If you don't have a repo and browser, that's a bad start. If it's tarballs only, I'd better know that it's something interesting. And if you do something non-predictable like archives on mediafire, good luck. -the community mailing list archive or forum When that's empty, it's a bad sign unless you can tell that the project is used elsewhere. Spam there is a VERY bad sign. -popcon/similar statistics from distros. These tell how many users install it. Use in the base of multiple distros is a particularly good sign. -look at the source code, look at the developers' reputations, review policies, etc. I mean, Theo may be belligerent, but if he (or Rich Felker) is involved, it means they are probably concerned about code quality. Which means that it's more likely to be maintainable than $RANDOM_PROJECT. If Linus has some say in the project (as opposed to periodically sending a "You're doing it WRONG!" email), one can expect a measure of functionality. If every random patch gets committed (WORST CASE EVER: tcc "mob branch"), run the other way. A fairly prompt code review for moderately small patches (I'm thinking new functions or 10-20 line changes) is a very good sign. -attitude towards standards. I'm not after standards-worship, but if pointing out that xyz is nonconformant gets any response besides fixing or a _sound_and_intelligeable_ explanation of why the standard is broken, go elsewhere. That way is the path to lockin and frustration. By the same token, "implements xyz according to RFCs 12345 and 6789" is a good sign. When there is a standard that's suitable, it should be used.
Slashdot Mirror
Point me to a few viruses for BeOS, OS/2 / eComStation, SolarOS, or Menuet.
Android is the most hackable mobile phone OS out there? Sure, but if you're going to argue that that discredits the security of the kernel like you seem to be saying, go ahead and point out how much of that is due to kernel bugs. (As far as I can tell, the main kernel bug was Samsung's ugo+rwx access to system memory--which would only be an issue for those who haven't updated).
The real issue is twofold:
First, many eyes won't do a thing to help if half the phones never get a single update after source code goes public. If you can find and fix every single bug within a month of the source drop, it will not change security for a device that's stuck with a rom from just after the code drop.
Second, notice that term "code drop". To quote Rob Landley, "Android isn't open source, it's regularly updated abandonware." "Many eyes" cannot work when there's no source before release; when you look at Honeycomb, applying it to Android becomes even more absurd.
is phrased as flamebait, the other half is worthy of a +1 in my book:
If there's anything you can't do after simply connecting the right peripherals (and installing any drivers!)...the only reasonable justification is that the software hasn't been written or installed yet. "You haven't licensed that functionality yet" is not an acceptable reason. (Note that I said functionality, not software.)
Selling a license to use a device as you see fit is not the same as making it not locked down.
Just to make something clear: the opposition to transgenic organisms is not all by those who claim that it "violates the sanctity of gods creations"-the "or somesuch" can range to people who are miles away from fundamentalists.
I see quite a bit from liberals who complain about doing things that "Mother Nature" didn't do. Not that there is really a difference when it comes down to the line of argument, it's just that "zealot idiots" can come from anywhere.
And I also see a number of people among conservatives who are in favor of transgenics.
But really, it seems to me that the opposition would weaken when someone stops to explain things like:
-the universal genetic code
-the origin of the genes in transgenic organisms (other species)
-why they are doing what they do
-the scope of a single gene relative to the whole genome
And also when they see someone they know doing it, rather than "those evil big companies".
So while it could be risky to participate, it is probably the step which is most likely to result in a real dialog and understanding.
~A conservative Christian who has worked for Pioneer and has prepared genetically modified E. coli during university coursework for his ag degree.
Hey! Banning clothes is restricting my religious freedom!
(Yes, the line should be drawn a lot sooner than that.)
+1 insightful.
There's also checking what level of confidentiality is involved (claim 18), what appears to be a buzzword killer (claim 16), alerting the company (claim 15), referring the employee to a company policy web page (claim 20) and some other such things.
It looks like this might be what Google needed for a certain email that got dragged out in the Oracle case.
It would also be useful for any developer of business email clients or office suites...
And in my case, I live near the largest and cheapest (especially when it comes to internet) city in Northern California. Don't assume that everyone on the internet is from Northern Europe, or that internet connections are always as cheap as they are where you are.
True, but surprise at a BSD using GCC (note that the OP did not mention GPL3) is not warranted by the existing conditions (note that deprecating does not mean dropping, it means stating that they intend to drop it, at FBSD 11 or later).
Judging by the fact that Thorsten Glaser (MirBSD/MirOS developer/maintainer) is a frequent poster on the pcc-devel list, I'd guess PCC. Also, some of the other MirOS projects are in a similar vein.
I suspect that NetBSD will use whatever compiler supports more of their architectures, with a preference for Clang when it's roughly equivalent...
How about the fact that only one of the two occurred "on the grounds of Bartow High School"?
I'm going to bet that if the kid who accidentally shot his brother with a BB gun had instead brought it to school (not shot it at school, just had it in his possession on campus), he would have been in major trouble.
Not that prosecuting this is sensible, just that the real issue may be "think of the children" gone wrong again rather than race.
> Or... they're patenting it before Amazon or Apple do so in order to avoid being sued themselves.
My suspicion is that you're correct--judging by Apple's past behavior, Apple would just add this to their patent on the corner of page curl animation and sue Samsung for "infringing" the patent they worked around.
> What is happening here? Heretics dare to use GNU code on a BSD system? Sacrilege!
BSD has used GNU tools for ages, but most of them have lagged behind.
FreeBSD is sticking to GPL2, which is GCC 4.2.1--but you can get a newer one from ports.
OpenBSD has 4.2.1 and 2.95.3.
NetBSD is the other BSD to use a GPL3 GCC, with GCC 4.5 as far as I can tell.
MirBSD is on GCC 3.4.
And the most fun difference: DragonFly BSD uses git as vcs, rather than cvs/svn.
> Sucks to be you.
> I pay $16 for a 20Mbps DSL connection.
> I bet there are other options in your area, you are just too stupid or lazy to look around. Try your local phone company, or even sattelite provider if you don't care too much about latency.
Have you ever seen the prices for satellite? In my area it starts at $70.
10 Mbps down/1.5 Mbps up DSL for $25 here,IIRC (my brother pays for it, and it's "fast enough to use"). But the same plan would cost $35 now and that's the cheapest DSL gets here.
So here's your "clueless provincial urbanite" award.
> I bought one last year. They're still for sale and being manufactured. If it ain't broke, don't fix it.
Yup, and I've got one that's probably about 5 years old.
Specifically, the WRT54G v8, which needs a micro build (2 MB flash, 8 mb ram).
I'm half wondering if it would be possible to build a musl/busybox system with a newer kernel...
I read and searched TFA (net-security.org and blog.sucuri.net), and the words "sum", "hash", and "checksum" do not occur on either page.
The closest it comes is saying that the timestamp is the same as the original, and that rpm -V won't work IF you use cPanel--because that's outside the package management system.
They suggest grepping for open_tty, though it would be possible to circumvent that with upx...
in which case file would report a corrupted ELF file.
>Microsoft had preemptive multitasking in 1995 Apple only got that in 2001.
WOOSH!
Roundup. ;)
Probably because it's a pretty standard target for genetic experiments.
And by the way...it may grow slow, but it reproduces pretty fast (you could get 2+ generations in a year, and it produces up to 10,000 seeds per plant...which means potentially 1 transformed plant -> 100 million in a year, though a lot of those would die in reality)
>Look where 1984 as required reading got us. If they make the kids read Lovecraft, I fear what the next generation of politicians will be like.
Good joke...or should I say that would be bad?
I think you stopped reading as soon as you hit the "because"
;)
"Stalemate" is the right word, he's making a pun on the similarity of the misspelling to the name of a certain rather well-known bearded man.
>Sorry, my facts are correct ... after all we have history classes in germany as well.
>But you are in so far right that indeed the south started shooting on a fortress occupied by northern troops.
>However the fortress was in southern territory. No idea and too lazy now to look up how and why the north could occupy it.
It was a previously-existing Federal fort, which under the Constitution is Federal territory.
And if you didn't know that, the quality of your "history" class is suspect.
Secession was motivated entirely by the fact that Lincoln was a noted abolitionist, from a party where the previous candidate had campaigned with this slogan:
"Free men, free soil, and Fremont!"
Anyone who says it was "states' rights" and not slavery has no clue: they are contradicting themselves, since states rights really meant "the right of states to determine the status of slavery within their own borders, to have that decision enforced against runaway slaves by every other state, and to secede (as long as they do so for the sake of protecting slavery; if it's for the sake of making slavery illegal, that's treason!)"
And yes, every one of those points is illustrated by an aspect of history: the objections to the Missouri Compromise, the Fugitive Slave Act, and the Southern response to New England's threats of secession.
And even if it were about secession, it's also about the reason for secession...which was very much slavery.
But the North, while not wanting slavery, did not fight for its abolition until 1863; that is correct.
Which is why Pioneer (one of the other companies that does "GMO" crops, as well as conventional) is working on corn engineered for more efficient use of nitrogen.
(Of course, this isn't something that will be acknowledged as a yield improvement; it doesn't raise the maximum "intrinsic yield", just yield when you can't put as much in. Which most GMO critics like to ignore, even if it means that the average farmer could raise more bushels with less water pollution...)
Also, more efficient use of resources allows more land to be fallowed and rotated. If you have 10% more yield you can put 9% of the land in alternate crops, such as legumes.
Good comment until you started talking about life cycles.
"Life cycles" are an issue in some (read numerous) circumstances.
More or less, it's a question of how much of #2 you need to do yourself. And you know that supporting LibreOffice yourself is probably not an option, but avoiding an office suite is also a bad idea.
Scenarios:
1) You need to provide a device that's supported for the next 3 years. Upstream has high churn.
If there's a regularly-updated stable branch/other support policy, you can just use that.
If there isn't you need to backport applicable bugfixes yourself, or write your own.
2) Software is written by a freelancer in between jobs. He says nothing about support. If he finds a job, you probably will be maintaining it yourself.
3) Company offers a solution, then a year later the needed code in GTK/Perl/PHP/... gets replaced/broken/...
If they support their solution for the next five years, you have nothing to worry about. If you're using Fedora and have to support it yourself, good luck.
Very good point.
Also I'll look at
-the last few months of commit logs--how many contributors, patch series, recurring contributors...
If you don't have a repo and browser, that's a bad start. If it's tarballs only, I'd better know that it's something interesting.
And if you do something non-predictable like archives on mediafire, good luck.
-the community mailing list archive or forum
When that's empty, it's a bad sign unless you can tell that the project is used elsewhere. Spam there is a VERY bad sign.
-popcon/similar statistics from distros. These tell how many users install it. Use in the base of multiple distros is a particularly good sign.
-look at the source code, look at the developers' reputations, review policies, etc.
I mean, Theo may be belligerent, but if he (or Rich Felker) is involved, it means they are probably concerned about code quality. Which means that it's more likely to be maintainable than $RANDOM_PROJECT. If Linus has some say in the project (as opposed to periodically sending a "You're doing it WRONG!" email), one can expect a measure of functionality.
If every random patch gets committed (WORST CASE EVER: tcc "mob branch"), run the other way.
A fairly prompt code review for moderately small patches (I'm thinking new functions or 10-20 line changes) is a very good sign.
-attitude towards standards. I'm not after standards-worship, but if pointing out that xyz is nonconformant gets any response besides fixing or a _sound_and_intelligeable_ explanation of why the standard is broken, go elsewhere. That way is the path to lockin and frustration. By the same token, "implements xyz according to RFCs 12345 and 6789" is a good sign. When there is a standard that's suitable, it should be used.
> Are any of these Enterprise distros?
> I don't know of any of those that distribute any of the kernel modules are speaking of.
http://ftp.scientificlinux.org/linux/scientific/6rolling/x86_64/addons/zfs/
Now you do.
Here:
http://media.smithsonianmag.com/audio/alexander-graham-bell.mp3
Not rifling, but the revolver...and several other repeaters.
As you would see if you had looked up the inventors named.