Re:Having trouble with 2.4.17, should I get this?
on
2.5.4 Kernel Out
·
· Score: 2
As far as I know, the problem's not been fixed in either tree, so you should probably sit tight with 2.4, and keep with the nopentium flag (or don't use it, and don't be surprised when it's unstable).
The problem is indeed with the combination of AGP and Athlon - disabling 3D support would probably help here, but is unlikely to make it go away completely.
I'd expect it shouldn't be too long (maybe it already happened, and I missed it) before there's a patch available for the problem (just disabling use of 4Mb mappings on Athlon should be the quick fix). When this is available, I imagine it'll go into both 2.4 and 2.5 fairly quickly.
While I'm not sure I agree with the original poster's point, separate off-the-shelf support for the OS and the box probably isn't the solution.
It'll work for any problem that's clearly an OS or clearly a hardware issue (or, by extension, clearly a userland-software issue), but it's not going to help when the problem's unclear. You'll get the A: It's B's fault, B: It's A's fault situation. If everything falls under A's remit, the problem goes away. It might be an OS problem, it might be a hardware problem, it might be a software problem, but in any event, it's A's problem to solve.
I don't think anyone should get a Turing for XML, the inventive step took place in SGML about 15 years earlier. XML is simply a cleanup of SGML that removes incompetence, idiocy and illogic from the SGML design. The basic principle is the same though. SGML was too badly executed to give Goldfarb any awards
Sure - like I said, it wasn't that innovative. I was just pulling examples out of the air to give an idea why it's better to award concepts that have proven themselves winners over time, rather than those that seem like a good idea now.
The problem is that straight after his lunar trip, you could tell that Armstrong was Astronaut of the year. In 1980, though, you might have thought Pascal was a good, recent, world-beating innovation (to pick a possibly bad, but nonetheless useful example). History shows that C won out, and was augmented from things in the OO field to give C++. So, in CS, things take time to prove themselves.
Granted, you could pick XML as a more recent thing that's going to have long-term improving effects. But it's not really a huge innovation, and it may turn out in 5 years that it's not as relevant as everyone thinks it's going to be.
I guess the conclusion would be that buzz-words make for bad awards.
It also seems they didn't do their research right. Great Britain is *not* planning the mandatory introduction of open source software in the public sector.
There have been a couple of initiatives examining the suitability of various open source systems in the public sector (as an example, they just started looking at open source as a potential component of the next police 'IT platform'), but they're by no means about to throw out the massive investment they have in all government departments, merely in order to jump toward open source.
You can already get their gcc-based toolchain for free. This is why they've got a huge developer base, and why it's an important question about whether they're going to continue doing that.
I'm no Free Software Zealot - to tell the truth, when it comes to this question, I don't really mind if it's Free as in Stallman or free as in beer. As far as I'm concerned, the only interesting part is whether it'll still be possible to get it for free, in order to do hobby development with it.
The article's fairly thin on detail. Other than the obvious and much-anticipated port to ARM, does anyone have any details on other new features? What are the new 'multimedia' features? How have they implemented this 'robust security'?
Most importantly, will Palm still be freely allowing development by releasing the SDK for free? (the move to ARM might have given them an opportunity to switch from gcc, thereby making this a question)
According to the law (at least in the US, UK and Germany, which are the three I'm familiar with), it's not legal. If the license says you pay to use it, and you haven't paid, you're not licensed.
You're right that it's not worth anyone's time raiding you about this, but that doesn't make it legal.
As for being a registered user, hey, I'm more than happy if you're a registered user of our stuff, but it's not going to change my opinions - I doubt if I paid you $20 you'd change yours either:-) (Indeed, I'd hope not)
Irrespective of the merits of your approach, one thing it sure isn't is legal. In principle, I've nothing against your behaviour - the practical difficulty is that it makes it impossible to draw the magical line in the sand between you and someone who just won't pay.
Software is a different animal than hardware, but if there was no difference between two things, why would anyone ever compare them? An elephant and a mouse are drastically different animals, but both have 4 legs, both can be grey, both are mammals....
If you want an extended demo, e-mail the authors and ask for one. Often as not, they'll give you one. If they don't, make your decision based on what you have. If your decision's to scrap the software, go for it.
Your example eventually worked out well for the companies in question, but it's just not possible to say that your behaviour's OK without providing the excuse to every pirate that "Oh, I've just been evaluating the software for the last 6.5 years". You end up with an unworkable know-it-when-I-see-it situation.
The companies want to sell the software to you, you should make them work for it, not create extra work for yourself. If they're not co-operative, take this as a sign of how much they're eager to win your custom and move on.
I write shareware - it has a 30 day trial period. If someone mails and asks, this can happily be extended to 60 or 90 days. No features are crippled during the trial. If the people don't ask, how am I to tell the difference between them and a pirate?
He's cost me bandwidth (which has an associated cost), and he's cost the honest users of my software more money, since a limited percentage of the users have to support the entire development costs. It may not be quite as direct, but it's not a victimless crime.
A good real-world comparison might be breaking into a shop which has closed down and stealing whatever stuff's left behind. You've not prevented any sales, and if the shop owner had chosen to give the stuff away (open source), then everyone would pretty much agree that this was fine. But when the shop owner hasn't done this, it's still stealing - piracy works the same way.
I think if you say this to NASA, they'll look at the x billion dollars they've already invested, the inevitable drastic budget cuts if you were to mess something up, not to mention the horrid PR consequences, and still say no.
First off - I don't hold anything against the one-click patent. If they want to waste money on silly patents, that's their problem.
Now that that's out of the way, it's nice to see them prove the nay-sayers wrong. But it doesn't mean a whole lot for the future - in order to get to this Milestone, they've spent upwards of $10 billion. They're paying $120 million a year in interest. If you take their 4th-quarter sales ($1.12 billion) as a guide to the next year's sales (generous - 4th quarter includes Christmas), you come out with 3% of their income going on interest payments.
I really hope they succeed - I order pretty much all my stuff from amazon.de - but this profit is by no means the end of the road. It's nice to see they managed to declare GAAP-standard profits, though, instead of using the pro-forma get-out clause.
This seems both readable and easy to follow to me. I maintain a large and (necessarily) complex firewall using iptables (and DNAT, SNAT, mark-based routing, etc.) I've never found it to be especially difficult to follow the config files, nor awkward to read.
I don't deny things could be just as simple as pf, possibly even easier, but I don't think complexity of configuration is a valid criticism of iptables. On the contrary, I'd have to say I find the example you gave a little counter-intuitive - it's necessary to think for a little too long about whether that's "to any" or "to any port". That's probably just me, though - in any event, this post hopefully makes it clear that the difference between the two is far more a matter of personal taste / how accustomed each person is to the syntax - neither of the syntaxes are (IMHO) intrinsically better.
* The second line's unnecessary if your input/forward chain policy is 'DROP', which would be the case for most sane firewalls I can think of...
RHN is not free. You get a free year (afair) with your copy of Red Hat, but if you want additional machines to use the service, or you want to use the service thereafter, it's $19.95/month
"the only reason that people out there are writing free software is that no one would want to pay for their code."
I'm no expert on Tru64 scalability, but this level of flamebait makes this post highly suspect to my mind. Would someone who both knows something on the subject and can manage to comment without bad-mouthing the competition care to say whether this post is really +3, Insightful?
I believe this used to be the case, but due to the time-criticality of deploying the landing gear, it was later changed to be an automatic system. I seem to recall that this move was greeted at the time with resistance from the crews, but that they now welcome it.
On the other hand, it's perfectly possible that I'm thinking of some other aspect of the landing process.
At which point, you just tune into the same random radio noise, and you have their randomnicity. There are some ways to get untappable random data sources, this is not one of them. See the recent story regarding using photons as OTPs.
Stealing a tool for injection moulding a case is 'free' most places:-)
Once you've stolen your machine, you just need some moulds. You already have examples of exactly how you want the case to be, the moulds can't be too far behind.
...and the point I forgot to mention - if there *is* a discovery here, it's certainly not being reported well by ZDnet. As far as I can tell, the articles consists of about 50% idle speculation and rumour-mongering on the part of rent-a-quote e-commerce types (with the exception of Bruce Schneier), and 50% contrived explanation by someone who doesn't understand anything about PKI.
If I have access to the server, I have access to the code that runs on the server. If I have access to the code, I can trace through that code, find out where it gets its keys from, and do the same thing. This has always been the case.
The key is to keep people from getting access to the server - not to claim that there's something wrong with the infrastructure because it's possible to compromise something outside of it.
Whilst more openness from previously closed companies can never be a particularly Bad Thing, I question Sun's motives - how many people are going to have the time, ability and resources to contribute to this?
Is Sun hoping to benefit more from the traditional advantages of Open Source (strong peer review, new viewpoints providing new enhancements) ? Or is this move intended to attract more press to Sun's recent announcements, and form more of a statistic for future Sun comments on their commitment to Open Source?
Slashdot Mirror
As far as I know, the problem's not been fixed in either tree, so you should probably sit tight with 2.4, and keep with the nopentium flag (or don't use it, and don't be surprised when it's unstable).
The problem is indeed with the combination of AGP and Athlon - disabling 3D support would probably help here, but is unlikely to make it go away completely.
I'd expect it shouldn't be too long (maybe it already happened, and I missed it) before there's a patch available for the problem (just disabling use of 4Mb mappings on Athlon should be the quick fix). When this is available, I imagine it'll go into both 2.4 and 2.5 fairly quickly.
While I'm not sure I agree with the original poster's point, separate off-the-shelf support for the OS and the box probably isn't the solution.
It'll work for any problem that's clearly an OS or clearly a hardware issue (or, by extension, clearly a userland-software issue), but it's not going to help when the problem's unclear. You'll get the A: It's B's fault, B: It's A's fault situation. If everything falls under A's remit, the problem goes away. It might be an OS problem, it might be a hardware problem, it might be a software problem, but in any event, it's A's problem to solve.
I don't think anyone should get a Turing for XML, the inventive step took place in SGML about 15 years earlier. XML is simply a cleanup of SGML that removes incompetence, idiocy and illogic from the SGML design. The basic principle is the same though. SGML was too badly executed to give Goldfarb any awards
Sure - like I said, it wasn't that innovative. I was just pulling examples out of the air to give an idea why it's better to award concepts that have proven themselves winners over time, rather than those that seem like a good idea now.
The problem is that straight after his lunar trip, you could tell that Armstrong was Astronaut of the year. In 1980, though, you might have thought Pascal was a good, recent, world-beating innovation (to pick a possibly bad, but nonetheless useful example). History shows that C won out, and was augmented from things in the OO field to give C++. So, in CS, things take time to prove themselves.
Granted, you could pick XML as a more recent thing that's going to have long-term improving effects. But it's not really a huge innovation, and it may turn out in 5 years that it's not as relevant as everyone thinks it's going to be.
I guess the conclusion would be that buzz-words make for bad awards.
It also seems they didn't do their research right. Great Britain is *not* planning the mandatory introduction of open source software in the public sector.
There have been a couple of initiatives examining the suitability of various open source systems in the public sector (as an example, they just started looking at open source as a potential component of the next police 'IT platform'), but they're by no means about to throw out the massive investment they have in all government departments, merely in order to jump toward open source.
You can already get their gcc-based toolchain for free. This is why they've got a huge developer base, and why it's an important question about whether they're going to continue doing that.
I'm no Free Software Zealot - to tell the truth, when it comes to this question, I don't really mind if it's Free as in Stallman or free as in beer. As far as I'm concerned, the only interesting part is whether it'll still be possible to get it for free, in order to do hobby development with it.
The article's fairly thin on detail. Other than the obvious and much-anticipated port to ARM, does anyone have any details on other new features? What are the new 'multimedia' features? How have they implemented this 'robust security'?
Most importantly, will Palm still be freely allowing development by releasing the SDK for free? (the move to ARM might have given them an opportunity to switch from gcc, thereby making this a question)
According to the law (at least in the US, UK and Germany, which are the three I'm familiar with), it's not legal. If the license says you pay to use it, and you haven't paid, you're not licensed.
:-) (Indeed, I'd hope not)
You're right that it's not worth anyone's time raiding you about this, but that doesn't make it legal.
As for being a registered user, hey, I'm more than happy if you're a registered user of our stuff, but it's not going to change my opinions - I doubt if I paid you $20 you'd change yours either
> I'll figure out how to be legal MY way
Irrespective of the merits of your approach, one thing it sure isn't is legal. In principle, I've nothing against your behaviour - the practical difficulty is that it makes it impossible to draw the magical line in the sand between you and someone who just won't pay.
Software is a different animal than hardware, but if there was no difference between two things, why would anyone ever compare them? An elephant and a mouse are drastically different animals, but both have 4 legs, both can be grey, both are mammals....
If you want an extended demo, e-mail the authors and ask for one. Often as not, they'll give you one. If they don't, make your decision based on what you have. If your decision's to scrap the software, go for it.
Your example eventually worked out well for the companies in question, but it's just not possible to say that your behaviour's OK without providing the excuse to every pirate that "Oh, I've just been evaluating the software for the last 6.5 years". You end up with an unworkable know-it-when-I-see-it situation.
The companies want to sell the software to you, you should make them work for it, not create extra work for yourself. If they're not co-operative, take this as a sign of how much they're eager to win your custom and move on.
I write shareware - it has a 30 day trial period. If someone mails and asks, this can happily be extended to 60 or 90 days. No features are crippled during the trial. If the people don't ask, how am I to tell the difference between them and a pirate?
He's cost me bandwidth (which has an associated cost), and he's cost the honest users of my software more money, since a limited percentage of the users have to support the entire development costs. It may not be quite as direct, but it's not a victimless crime.
A good real-world comparison might be breaking into a shop which has closed down and stealing whatever stuff's left behind. You've not prevented any sales, and if the shop owner had chosen to give the stuff away (open source), then everyone would pretty much agree that this was fine. But when the shop owner hasn't done this, it's still stealing - piracy works the same way.
I think if you say this to NASA, they'll look at the x billion dollars they've already invested, the inevitable drastic budget cuts if you were to mess something up, not to mention the horrid PR consequences, and still say no.
First off - I don't hold anything against the one-click patent. If they want to waste money on silly patents, that's their problem.
Now that that's out of the way, it's nice to see them prove the nay-sayers wrong. But it doesn't mean a whole lot for the future - in order to get to this Milestone, they've spent upwards of $10 billion. They're paying $120 million a year in interest. If you take their 4th-quarter sales ($1.12 billion) as a guide to the next year's sales (generous - 4th quarter includes Christmas), you come out with 3% of their income going on interest payments.
I really hope they succeed - I order pretty much all my stuff from amazon.de - but this profit is by no means the end of the road. It's nice to see they managed to declare GAAP-standard profits, though, instead of using the pro-forma get-out clause.
See the official press release here.
The point would be valid if there weren't a more readable way to configure iptables:
iptables --append firewall --source 10.11.0.0/16 --proto tcp --destination-port ssh --jump ACCEPT
iptables --append firewall --destination-port ssh --jump DROP*
This seems both readable and easy to follow to me. I maintain a large and (necessarily) complex firewall using iptables (and DNAT, SNAT, mark-based routing, etc.) I've never found it to be especially difficult to follow the config files, nor awkward to read.
I don't deny things could be just as simple as pf, possibly even easier, but I don't think complexity of configuration is a valid criticism of iptables. On the contrary, I'd have to say I find the example you gave a little counter-intuitive - it's necessary to think for a little too long about whether that's "to any" or "to any port". That's probably just me, though - in any event, this post hopefully makes it clear that the difference between the two is far more a matter of personal taste / how accustomed each person is to the syntax - neither of the syntaxes are (IMHO) intrinsically better.
* The second line's unnecessary if your input/forward chain policy is 'DROP', which would be the case for most sane firewalls I can think of...
RHN is not free. You get a free year (afair) with your copy of Red Hat, but if you want additional machines to use the service, or you want to use the service thereafter, it's $19.95/month
Only on the assumption that this is your only world-executable suid binary. If you've any others, they're also vulnerable. To check:
cd /
find . -perm -4001 -uid 0
(this isn't quite complete - if you had a file which was suid root, owned by a non-root group and group-executable, that would also be vulnerable)
"the only reason that people out there are writing free software is that no one would want to pay for their code."
I'm no expert on Tru64 scalability, but this level of flamebait makes this post highly suspect to my mind. Would someone who both knows something on the subject and can manage to comment without bad-mouthing the competition care to say whether this post is really +3, Insightful?
I believe this used to be the case, but due to the time-criticality of deploying the landing gear, it was later changed to be an automatic system. I seem to recall that this move was greeted at the time with resistance from the crews, but that they now welcome it.
On the other hand, it's perfectly possible that I'm thinking of some other aspect of the landing process.
That's an XT, not a PC.
At which point, you just tune into the same random radio noise, and you have their randomnicity. There are some ways to get untappable random data sources, this is not one of them. See the recent story regarding using photons as OTPs.
I can see the headlines now - "Slashdot uses closed-source web server". You certainly know how to excite the VA/Andover/Red Hat/AOL conspiracy crowd.
Stealing a tool for injection moulding a case is 'free' most places :-)
Once you've stolen your machine, you just need some moulds. You already have examples of exactly how you want the case to be, the moulds can't be too far behind.
...and the point I forgot to mention - if there *is* a discovery here, it's certainly not being reported well by ZDnet. As far as I can tell, the articles consists of about 50% idle speculation and rumour-mongering on the part of rent-a-quote e-commerce types (with the exception of Bruce Schneier), and 50% contrived explanation by someone who doesn't understand anything about PKI.
If I have access to the server, I have access to the code that runs on the server. If I have access to the code, I can trace through that code, find out where it gets its keys from, and do the same thing. This has always been the case.
The key is to keep people from getting access to the server - not to claim that there's something wrong with the infrastructure because it's possible to compromise something outside of it.
Whilst more openness from previously closed companies can never be a particularly Bad Thing, I question Sun's motives - how many people are going to have the time, ability and resources to contribute to this?
Is Sun hoping to benefit more from the traditional advantages of Open Source (strong peer review, new viewpoints providing new enhancements) ? Or is this move intended to attract more press to Sun's recent announcements, and form more of a statistic for future Sun comments on their commitment to Open Source?