I was all down on Slashdot editors earlier today in a post to "Scientist Must Pay to Read His Own Paper."
So I guess I have at least something in common with this kdawson person. OTOH, various replies lead me to believe that applying a mental kdawson filter would be a Good Thing. A random sig spotted today seems appropriate. To paraphrase: You are probably an asshole, and don't know it.
Thanks Scott. I didn't consider the editor, and seldom do. Coincidentally, I took a shot at that them earlier today, which I shouldn't repeat here, out of context. Fair is fair.
I don't get to read Slashdot as often as I'd like. Time is usually short, so being able to apply a mental kdawson filter is a useful tool.
I've possibly overstated the case, due to reading psych studies done several years ago. Those were mostly based around burger chains. But you've made one argument that might actually support me, which would be the association of green with fresh. On one level, that seems so obvious as to be stupid--but that's the entire point.
Green for Starbucks? Well, I could grasp at unprovable nonsense, such as green being closely associated with Washington and the PacNW. Maybe there's something in corporate history involved. That's a reach, too, as I've no clue, and don't mean to research it. Besides which "...I don't think "soothing" is the image they try to achieve." is such a master-stroke that I have to bow.
WTF? I don't see a reference to kdawson. Not the submitter, and not a poster or respondent. Is this something where you have to hang out here daily in order to get it? Part of me wants to know how to judge the quality of any poster's efforts, but most of me wonders if assuming specific knowledge of a poster's biases (and I have my own, as most of us probably do) isn't setting the bar a bit high.
Well, at least needing to settle a bet makes some frapping *sense*. The second link, I can't make heads nor tails of. Which is an indictment. It seems to me that an omnipotent, omniscient, supernatural figure who wants to be recognized as the One True Bad-Ass, might realize that his references eventually lose meaning in a changing world. Maybe even revisit the problem, and just generally *clear things up*.
"...isle that is called Patmos" doesn't really impart a lot of *meaning*, considering that I'm supposed to correctly choose the One True Bad-Ass, on pain of eternal suffering. OTOH, if you were interested in building some *infrastructure*, and covering personnel and other expenses as, say, a *corporation* might do, this would be a sweet approach.
Execution (no pun intended) might be tricky. You'd need to deploy some serious Politics of Fear. Fear of death! Yeah, that's the ticket! That will *always* work.
The sad bit is that on balance, connecting as directly as possible to the brainstem usually *does* work. Organized religions do it, governments do it, and corporations do it. That last idea can be particularly entertaining to research. Google around for psychological color responses, and you'll soon see why fast-food restaurants, Denny's, etc., always use red, yellow, orange, and white (warm colors) in logos, signage, etc. No blues and greens allowed--that's for IT firms that want to project a soothing, 'we can fix all your problems' vibe.[1]
At some level, we are all subject to a bit of social engineering. I'm a security guy. You might be surprised at how many times I'm specifically not allowed to perform social engineering attacks. The reason, quite simply, is that they work. Corporations generally want to establish the fact that they are secure, not that I can call employees and trust the brainstem to provide me with the keys to the kingdom in an afternoon's work.
[1] Red Hat is an exception, but that goes back to corporate history, and can't be easily changed.
"It can be difficult to remember what you did once you become sober, but easier if you again become intoxicated."
Recommend you have several beers. Just to gain a bit of experience, and perhaps avoid some Thetan brain-rays, or whatever is the current source of what seems to be a serious cerebral malfunction.
Wow. I always knew that there would be some circumstance where beer would be medicinal. It took entirely too long, but was worth the wait.
OK. I visited your links. Now I have to wonder if he's a fifth-columnist, in which case I'd have to say, "Nicely done!"
But see posts later in the discussion, regarding a Slate post that CoS isn't any weirder than others, just newer. http://www.slate.com/id/2171416/
At some level, religion of any stripe disturbs me, as I see it all as both irrational and irrelevant. That said, at least some religions seem able to at least maintain a bit of dignity in their celebrations, and not *completely* insult the intelligence of their followers. I thought lost tribes of Israel present in central America (contrary to genetic evidence, but then we're not speaking of people who would believe in genetics), and wearing underwear that seems to serve the function of a wearable Post It note was a bit odd.
Now I'm trying to quantify the limits of weird, thinking of how reincarnation would rate, etc. At some point, my head will explode. Have you seen Tim Burton's _Mars Attacks_? Yeah, like that.
Think of it as evolution in action. There is no correlation between 'media personality' and 'intelligent', despite what tabloid readers seem to believe. One more reason for me to wonder why we have various concerts, etc., which are purported to support one thing or another.
Perhaps we're getting beyond this, as a couple of events this year have fallen pretty flat. Personally, I don't feel the need to hear about the latest lunacy of slut du jour, musicians deploying the Single Name PR Strategy (Sting, Madonna, Bono), or even OSS folk (media stars in a small world) who deploy the initials or login strategy (RS, ESR).
Even in the rare case of people of this ilk actually having something meaningful to say, they tend to not be people that I'd want to be trapped in an elevator with. I'm a Linux and BSD guy, so it's somewhat hard to admit that last. But truthfully, I'd sooner be trapped in an elevator with Bill Gates than Richard Stallman or Eric Raymond. I already get the advantages of the approaches preached by the last two, but Gates could at least tell me things I definitely don't know about the threat of malaria as an endemic disease. http://www7.nationalgeographic.com/ngm/0707/featur e1/
Excellent link. There I was, thinking Scientology was even weirder than most other belief systems based upon the supernatural or pseudoscience (astrology, etc.). It put things into a more reasonable perspective for me. A quick read, and well worth the time.
I found parent -1, Flamebait. Personally, I find Slashdot editor performance uneven at best, at several levels. Those range from accepting submissions with stupid questions at the end, to a failure to avoid sensationalism. Nor have I ever heard of a Slashdot 'editor' performing a basic editorial function--working with an author. So I question the usage of the very *term* Slashdot editor.
In this respect, I'd call Slashdot Fox News for Nerds. Stuff That Can Deliver Eyeballs to Advertisers.
That's just my opinion, and you know what they say about opinions. But whoever modded this Flamebait certainly didn't read the moderator guidelines.
For what it's worth, I do find things on Slashdot that I likely wouldn't know of if Slashdot didn't exist. I value it as an information source. I just don't think the 'editors' are equal to their job titles.
From TFA: "Thrown into that mix is the private sector, a factor that was never imagined in 1957." It certainly was imagined. Heinlein _The Man Who Sold the Moon_ in 1951, etc. The exploration of space has always been advocated by visionaries, and beset by nay-sayers.
You're describing the colonization of space in terms of return on investment. What you've said has been said by many others, for decades. History certainly doesn't justify this, as national prestige was what drove the original space race. The huge economic returns brought through miniaturization, materials, weather forecasting, etc., were largely serendipitous. Yet they've paid for every dime ever spent on space, many times over.
Nor do I think that a prediction based on ROI will be any more accurate in future than it's been in the past.
Available technologies (which could radically alter the I in ROI) do not remain fixed. What about the 'R'? I doubt that the desire for national prestige will disappear. It's also quite possible that we, as a species, might gain the ultimate R--survival. A couple of scenarios for that might include having a self-sustaining colony away from earth when some bio-weapon is used, whether by a nation, or a non-state actor. Or having enough experience doing industrial-scale things in space to deflect an asteroid or comet if necessary.
There are other arguments, but these will do to go on with.
Well, that's the issue, in a nutshell. The odds are very good that different pieces of the sync software *would* accept input that it shouldn't. Exploits based upon a failure to sanitize input are common as dirt. Assuming that there are competent developers at work may or may not be the best of all possible plans.
Creating secure software is *hard*, in and of itself, business reasons often preclude taking the extra time to get it right, etc. And often, people just never hear about potential downsides. That sounds insane, but it really does happen. WEP in wireless networking is a good example. When IEEE trotted the thing out as a standard, I argued against it with a couple of clients who asked me about. Not just the software, but the very *term*.
They didn't deploy it, and I called that a win as the cracking tools came out. But my neighbor never even considered it might be an issue, bought a broad-band modem with wireless networking, and was badly bitten. He's not an IT guy, much less a security guy. But nor is he stupid.
He's just some guy. Actually, a really nice guy. Friendly, good conversationalist, knowledgeable on many subjects. Does a good barbeque, and is generally a pleasant neighbor to tip a beer with. He owns a couple of nurseries which are well-regarded amongst the local lawn-and-garden set, and does well at it. So he's not without business sense.
He even knew he a had a security guy living next door to him. But it never occurred to him to ask me about wireless. He just assumed that large companies like telcos and cable companies wouldn't be pushing this stuff at him if there were a problem. Yet he didn't realize that a risk/reward consideration was even in play--despite complaining to me about the cost of having to pay for anti-virus, etc., on the several machines he needs to run his business. So he does have some notion of the importance of security.
IMHO, we have a continuous litany of exploited systems because people in the security industry (that would me) have *failed*. There must be an effective method of teaching this stuff, at a level that people just *get*. But I'll be damned if I know what it is. You've met an exploit developer, and you've some knowledge of the development process. That means that you're miles ahead of the average user. But you want a convenience or coolness thing, and you've made assumptions about the way it would work, and the way it would be built, that history completely fails to justify. The entire idea should have rung all your alarm bells.
This is mondo depressing. Security practitioners and educators (again, that would be me) clearly have a long way to go before before we can even see success as a hazy thing on a distant shore.
Last year, some iPods shipped with a virus. A fairly small number (around 1%), but a concern. But given the number of times Facebook has been hacked, you might want to reconsider. Facebook would likely be far more damaging. Unless you think all problems with Facebook have now been fixed, and no new ones will be introduced as the software evolves?
If the value/risk proposition is acceptable to you, go for it. The sad part will be that relatively few people would realize that there *was* a risk/value proposition that required evaluation. They'd just do it for the convenience, a perceived coolness factor, or whatever. Then we'd get to read all about it on Slashdot, in an article filled with outraged posts. Same old thing, different day.
Re:Fair??? Language, please...
on
SCO Loses
·
· Score: 1
"BILLIONS of dollars in lawyers fees, thousands of hours of (taxpayer-funded) court costs, and millions of manpower hours"
You're off by *at least* by at least one order of magnitude in any of that. In manhours, I can't even imagine. At least three, just to get it into millions of manhours. Do you have a resume on file with Fox News, yet? Though being off by a factor a thousand may be a bit much, even for them.
"my understanding is that they're not copyrightable"
OK, a derived work, even with the nonstandard bits. Good! I thought I'd heard something intimating that part of IBM being sued by SCO was about header files. But I'd gotten numb and quit following a lot of that, so maybe I misheard, and that was part of why the suit was bogus. Or maybe I heard a bunch of bull.
You can't run it on a production server. You still need a debug kernel. Given how hard it can be (sometimes nearly impossible on complex systems) to duplicate an issue, the ability to run on a production system can be *the* killer feature.
OTOH, it's getting better fast, and the feature list is already pretty sweet.
But, I also have to comment that SystemTap was developed as a response to DTrace. This would have been useful to kernel devs a long time ago. But SystemTap didn't happen until the Solaris folk paved the way, and people began to clamor. Yet more proof that competition is good, and Solaris bears keeping an eye on.
"GNU is an operating system, just like Solaris and BSD."
That will be true the day they announce that The Hurd, or some other kernel, is ready to go. Until then, it's a userland. The purpose of an operating system is to allocate memory, schedule processes, handle networking, manage filesystems, and other I/O, etc.
GNU software can be very good, and I don't mean to take anything away from them. But calling GNU an operating system is just wildly wrong. See http://en.wikipedia.org/wiki/Operating_system or a huge number of other references.
The kernel rewrites can be a problem. I'm overdue to look through the IBM Linux Performance Tuning Redbook (July, 2007 version) again. Grovel through 168 pages of PDF. Compare to my current production kernels, compare to notes from the previous Redbook version, sort through my last round of production performance metrics. Test any changes, and fold into the configuration management system. Gack. The manhours do stack up, but luckily, I can do it in my copious free time. http://www.redbooks.ibm.com/redpapers/pdfs/redp428 5.pdf
I'm hoping that the LSB 3.0 will make life easier in future, regarding the ABI stability issue. They're claiming six years, going forward. But Sun have always done an excellent job there. Perhaps the best job amongst the commercial Unix vendors. They beat the stuffing out of HP-UX, but I've little experience with AIX, so I really can't claim they're best. But they're definitely someone to beat. http://www.linux-foundation.org/en/Application_Com patibility
I don't know that I'd recommend switching any workloads to OpenSolaris if and when they release on GPL, as there are costs for hardware, support contracts, and staff to consider. So that would have to be determined on a client-by-client basis. But take a very serious look at it? Oh, yeah.
I don't see how he's doing anything at the 'expense' of Debian or GNU folk. Without a specific reference, I don't see how he's 'exploiting' anything. The FSF, with the GPL3 changes made in response to the SuSE imbroglio, have proven that they can and will react to exploitation. But I've not seen anything done in response to any of Murdoch's or Sun's actions.
Is the man considered to be somehow ensnared in Debian his entire life, just because he created the project? In '93, as a student? Lives evolve. Responsibilities change. For all I know, he's trying to put children through college, and a Debian relationship just wasn't making that possible. I'm a *long* way from willing to brand the man a some sort of traitor to the cause. Did he, or did he not, create something valuable to many people, which is being built upon to this day, 14 years later? At what point do you decide, "Yeah, he's been an overall force for good," and cut the man some slack?
I know that Sun have contributed to a lot of open standards, etc., over the years. I'm not claiming that they are 100% knights in shining armor. But they are slowly moving in good directions. Given Murdoch's early environment, I'd expect him to be rather more a force for long-term good than some sort of Dark Lord of the Sith.
Our differences here (if you're one poster) are about religion. I value free (as in freedom) software as well. But I don't think I'm willing to be nearly as rigid as you. Quoting a manifesto is a far cry from evidence, and I simply don't see how he has harmed Debian, or GNU.
I can understand that viewpoint. I'm benefitting from it even as we speak.
But if it's just some header lines, this will have no affect on you. That's *why* I doubt anyone will care very much. In some respects, particularly where there's some push for a move to GPL, as is the case with OpenSolaris, header commonality might even be regarded as desireable.
Some people, for instance, do small-scale systems administration. One source of information on some of the limits of their systems, such as how many characters can be used in a filename, are defined in a header file called, appropriately enough, limits.h. So, say,/usr/include/linux/limits.h contains a line reading: #define NAME_MAX 255/* # chars in a file name */
It's probably better that users of different systems didn't have to keep track of different naming conventions, especially as comments may not be provided. In a system I'm looking at right now, for instance, there's another line: #define RTSIG_MAX 32
That's the maximum number of realtime signals supported. But there's no comment--you just have to know it. Standardization has it's points. Ideally, you learn this once, for one Unixy OS, and it's portable knowledge. It's a far cry from a wholesale rippoff, including algorithms and their implementations, such as SCO implied in the SCO-IBM insanity.
I seriously doubt very much that more than a tiny percentage Linux kernel developers would have a problem with a few header lines. I'm far from certain that copyright law should be pushed that far, lest developers have to use values with names like THIS_NAME_SO_WE_DO_NOT_INFRINGE. That's not in *anyone's* best interest.
Slashdot Mirror
"I hope your mother gets diarrhea tonight."
But think of the children!
I was all down on Slashdot editors earlier today in a post to "Scientist Must Pay to Read His Own Paper."
So I guess I have at least something in common with this kdawson person. OTOH, various replies lead me to believe that applying a mental kdawson filter would be a Good Thing. A random sig spotted today seems appropriate. To paraphrase: You are probably an asshole, and don't know it.
I should probably filter my own posts.
Thanks Scott. I didn't consider the editor, and seldom do. Coincidentally, I took a shot at that them earlier today, which I shouldn't repeat here, out of context. Fair is fair.
I don't get to read Slashdot as often as I'd like. Time is usually short, so being able to apply a mental kdawson filter is a useful tool.
I've possibly overstated the case, due to reading psych studies done several years ago. Those were mostly based around burger chains. But you've made one argument that might actually support me, which would be the association of green with fresh. On one level, that seems so obvious as to be stupid--but that's the entire point.
Green for Starbucks? Well, I could grasp at unprovable nonsense, such as green being closely associated with Washington and the PacNW. Maybe there's something in corporate history involved. That's a reach, too, as I've no clue, and don't mean to research it. Besides which "...I don't think "soothing" is the image they try to achieve." is such a master-stroke that I have to bow.
WTF? I don't see a reference to kdawson. Not the submitter, and not a poster or respondent. Is this something where you have to hang out here daily in order to get it? Part of me wants to know how to judge the quality of any poster's efforts, but most of me wonders if assuming specific knowledge of a poster's biases (and I have my own, as most of us probably do) isn't setting the bar a bit high.
Well, at least needing to settle a bet makes some frapping *sense*. The second link, I can't make heads nor tails of. Which is an indictment. It seems to me that an omnipotent, omniscient, supernatural figure who wants to be recognized as the One True Bad-Ass, might realize that his references eventually lose meaning in a changing world. Maybe even revisit the problem, and just generally *clear things up*.
"...isle that is called Patmos" doesn't really impart a lot of *meaning*, considering that I'm supposed to correctly choose the One True Bad-Ass, on pain of eternal suffering. OTOH, if you were interested in building some *infrastructure*, and covering personnel and other expenses as, say, a *corporation* might do, this would be a sweet approach.
Execution (no pun intended) might be tricky. You'd need to deploy some serious Politics of Fear. Fear of death! Yeah, that's the ticket! That will *always* work.
The sad bit is that on balance, connecting as directly as possible to the brainstem usually *does* work. Organized religions do it, governments do it, and corporations do it. That last idea can be particularly entertaining to research. Google around for psychological color responses, and you'll soon see why fast-food restaurants, Denny's, etc., always use red, yellow, orange, and white (warm colors) in logos, signage, etc. No blues and greens allowed--that's for IT firms that want to project a soothing, 'we can fix all your problems' vibe.[1]
At some level, we are all subject to a bit of social engineering. I'm a security guy. You might be surprised at how many times I'm specifically not allowed to perform social engineering attacks. The reason, quite simply, is that they work. Corporations generally want to establish the fact that they are secure, not that I can call employees and trust the brainstem to provide me with the keys to the kingdom in an afternoon's work.
[1] Red Hat is an exception, but that goes back to corporate history, and can't be easily changed.
"It can be difficult to remember what you did once you become sober, but easier if you again become intoxicated."
Recommend you have several beers. Just to gain a bit of experience, and perhaps avoid some Thetan brain-rays, or whatever is the current source of what seems to be a serious cerebral malfunction.
Wow. I always knew that there would be some circumstance where beer would be medicinal. It took entirely too long, but was worth the wait.
OK. I visited your links. Now I have to wonder if he's a fifth-columnist, in which case I'd have to say, "Nicely done!"
But see posts later in the discussion, regarding a Slate post that CoS isn't any weirder than others, just newer.
http://www.slate.com/id/2171416/
At some level, religion of any stripe disturbs me, as I see it all as both irrational and irrelevant. That said, at least some religions seem able to at least maintain a bit of dignity in their celebrations, and not *completely* insult the intelligence of their followers. I thought lost tribes of Israel present in central America (contrary to genetic evidence, but then we're not speaking of people who would believe in genetics), and wearing underwear that seems to serve the function of a wearable Post It note was a bit odd.
Now I'm trying to quantify the limits of weird, thinking of how reincarnation would rate, etc. At some point, my head will explode. Have you seen Tim Burton's _Mars Attacks_? Yeah, like that.
Think of it as evolution in action. There is no correlation between 'media personality' and 'intelligent', despite what tabloid readers seem to believe. One more reason for me to wonder why we have various concerts, etc., which are purported to support one thing or another.
r e1/
Perhaps we're getting beyond this, as a couple of events this year have fallen pretty flat. Personally, I don't feel the need to hear about the latest lunacy of slut du jour, musicians deploying the Single Name PR Strategy (Sting, Madonna, Bono), or even OSS folk (media stars in a small world) who deploy the initials or login strategy (RS, ESR).
Even in the rare case of people of this ilk actually having something meaningful to say, they tend to not be people that I'd want to be trapped in an elevator with. I'm a Linux and BSD guy, so it's somewhat hard to admit that last. But truthfully, I'd sooner be trapped in an elevator with Bill Gates than Richard Stallman or Eric Raymond. I already get the advantages of the approaches preached by the last two, but Gates could at least tell me things I definitely don't know about the threat of malaria as an endemic disease. http://www7.nationalgeographic.com/ngm/0707/featu
Please specify closet. I've been hearing rumors of an intimate cross-species relationship with Xenu.
Excellent link. There I was, thinking Scientology was even weirder than most other belief systems based upon the supernatural or pseudoscience (astrology, etc.). It put things into a more reasonable perspective for me. A quick read, and well worth the time.
Scientology is so bizarre that I can't tell if you're being facetious or not.
I found parent -1, Flamebait. Personally, I find Slashdot editor performance uneven at best, at several levels. Those range from accepting submissions with stupid questions at the end, to a failure to avoid sensationalism. Nor have I ever heard of a Slashdot 'editor' performing a basic editorial function--working with an author. So I question the usage of the very *term* Slashdot editor.
In this respect, I'd call Slashdot Fox News for Nerds. Stuff That Can Deliver Eyeballs to Advertisers.
That's just my opinion, and you know what they say about opinions. But whoever modded this Flamebait certainly didn't read the moderator guidelines.
For what it's worth, I do find things on Slashdot that I likely wouldn't know of if Slashdot didn't exist. I value it as an information source. I just don't think the 'editors' are equal to their job titles.
I've just burned my mod points, and parent post should be closer to 5 than 2. Would someone please take care of this?
From TFA: "Thrown into that mix is the private sector, a factor that was never imagined in 1957." It certainly was imagined. Heinlein _The Man Who Sold the Moon_ in 1951, etc. The exploration of space has always been advocated by visionaries, and beset by nay-sayers.
You're describing the colonization of space in terms of return on investment. What you've said has been said by many others, for decades. History certainly doesn't justify this, as national prestige was what drove the original space race. The huge economic returns brought through miniaturization, materials, weather forecasting, etc., were largely serendipitous. Yet they've paid for every dime ever spent on space, many times over.
Nor do I think that a prediction based on ROI will be any more accurate in future than it's been in the past.
Available technologies (which could radically alter the I in ROI) do not remain fixed. What about the 'R'? I doubt that the desire for national prestige will disappear. It's also quite possible that we, as a species, might gain the ultimate R--survival. A couple of scenarios for that might include having a self-sustaining colony away from earth when some bio-weapon is used, whether by a nation, or a non-state actor. Or having enough experience doing industrial-scale things in space to deflect an asteroid or comet if necessary.
There are other arguments, but these will do to go on with.
Well, that's the issue, in a nutshell. The odds are very good that different pieces of the sync software *would* accept input that it shouldn't. Exploits based upon a failure to sanitize input are common as dirt. Assuming that there are competent developers at work may or may not be the best of all possible plans.
Creating secure software is *hard*, in and of itself, business reasons often preclude taking the extra time to get it right, etc. And often, people just never hear about potential downsides. That sounds insane, but it really does happen. WEP in wireless networking is a good example. When IEEE trotted the thing out as a standard, I argued against it with a couple of clients who asked me about. Not just the software, but the very *term*.
They didn't deploy it, and I called that a win as the cracking tools came out. But my neighbor never even considered it might be an issue, bought a broad-band modem with wireless networking, and was badly bitten. He's not an IT guy, much less a security guy. But nor is he stupid.
He's just some guy. Actually, a really nice guy. Friendly, good conversationalist, knowledgeable on many subjects. Does a good barbeque, and is generally a pleasant neighbor to tip a beer with. He owns a couple of nurseries which are well-regarded amongst the local lawn-and-garden set, and does well at it. So he's not without business sense.
He even knew he a had a security guy living next door to him. But it never occurred to him to ask me about wireless. He just assumed that large companies like telcos and cable companies wouldn't be pushing this stuff at him if there were a problem. Yet he didn't realize that a risk/reward consideration was even in play--despite complaining to me about the cost of having to pay for anti-virus, etc., on the several machines he needs to run his business. So he does have some notion of the importance of security.
IMHO, we have a continuous litany of exploited systems because people in the security industry (that would me) have *failed*. There must be an effective method of teaching this stuff, at a level that people just *get*. But I'll be damned if I know what it is. You've met an exploit developer, and you've some knowledge of the development process. That means that you're miles ahead of the average user. But you want a convenience or coolness thing, and you've made assumptions about the way it would work, and the way it would be built, that history completely fails to justify. The entire idea should have rung all your alarm bells.
This is mondo depressing. Security practitioners and educators (again, that would be me) clearly have a long way to go before before we can even see success as a hazy thing on a distant shore.
With that convenience you also assume risk.
Last year, some iPods shipped with a virus. A fairly small number (around 1%), but a concern. But given the number of times Facebook has been hacked, you might want to reconsider. Facebook would likely be far more damaging. Unless you think all problems with Facebook have now been fixed, and no new ones will be introduced as the software evolves?
http://mashable.com/2007/07/31/facebook-downtime/
would tend to argue against that. The link appearing on a site named mashable.com has a certain irony.
If the value/risk proposition is acceptable to you, go for it. The sad part will be that relatively few people would realize that there *was* a risk/value proposition that required evaluation. They'd just do it for the convenience, a perceived coolness factor, or whatever. Then we'd get to read all about it on Slashdot, in an article filled with outraged posts. Same old thing, different day.
"BILLIONS of dollars in lawyers fees, thousands of hours of (taxpayer-funded) court costs, and millions of manpower hours"
You're off by *at least* by at least one order of magnitude in any of that. In manhours, I can't even imagine. At least three, just to get it into millions of manhours. Do you have a resume on file with Fox News, yet? Though being off by a factor a thousand may be a bit much, even for them.
Thanks for the word. I'll jog him for what, if any problems he may have had, and maybe take it for a spin.
"my understanding is that they're not copyrightable"
OK, a derived work, even with the nonstandard bits. Good! I thought I'd heard something intimating that part of IBM being sued by SCO was about header files. But I'd gotten numb and quit following a lot of that, so maybe I misheard, and that was part of why the suit was bogus. Or maybe I heard a bunch of bull.
You can't run it on a production server. You still need a debug kernel. Given how hard it can be (sometimes nearly impossible on complex systems) to duplicate an issue, the ability to run on a production system can be *the* killer feature.
OTOH, it's getting better fast, and the feature list is already pretty sweet.
But, I also have to comment that SystemTap was developed as a response to DTrace. This would have been useful to kernel devs a long time ago. But SystemTap didn't happen until the Solaris folk paved the way, and people began to clamor. Yet more proof that competition is good, and Solaris bears keeping an eye on.
"GNU is an operating system, just like Solaris and BSD."
That will be true the day they announce that The Hurd, or some other kernel, is ready to go. Until then, it's a userland. The purpose of an operating system is to allocate memory, schedule processes, handle networking, manage filesystems, and other I/O, etc.
GNU software can be very good, and I don't mean to take anything away from them. But calling GNU an operating system is just wildly wrong. See http://en.wikipedia.org/wiki/Operating_system or a huge number of other references.
The kernel rewrites can be a problem. I'm overdue to look through the IBM Linux Performance Tuning Redbook (July, 2007 version) again. Grovel through 168 pages of PDF. Compare to my current production kernels, compare to notes from the previous Redbook version, sort through my last round of production performance metrics. Test any changes, and fold into the configuration management system. Gack. The manhours do stack up, but luckily, I can do it in my copious free time.8 5.pdf
m patibility
http://www.redbooks.ibm.com/redpapers/pdfs/redp42
I'm hoping that the LSB 3.0 will make life easier in future, regarding the ABI stability issue. They're claiming six years, going forward. But Sun have always done an excellent job there. Perhaps the best job amongst the commercial Unix vendors. They beat the stuffing out of HP-UX, but I've little experience with AIX, so I really can't claim they're best. But they're definitely someone to beat.
http://www.linux-foundation.org/en/Application_Co
I don't know that I'd recommend switching any workloads to OpenSolaris if and when they release on GPL, as there are costs for hardware, support contracts, and staff to consider. So that would have to be determined on a client-by-client basis. But take a very serious look at it? Oh, yeah.
I don't see how he's doing anything at the 'expense' of Debian or GNU folk. Without a specific reference, I don't see how he's 'exploiting' anything. The FSF, with the GPL3 changes made in response to the SuSE imbroglio, have proven that they can and will react to exploitation. But I've not seen anything done in response to any of Murdoch's or Sun's actions.
Is the man considered to be somehow ensnared in Debian his entire life, just because he created the project? In '93, as a student? Lives evolve. Responsibilities change. For all I know, he's trying to put children through college, and a Debian relationship just wasn't making that possible. I'm a *long* way from willing to brand the man a some sort of traitor to the cause. Did he, or did he not, create something valuable to many people, which is being built upon to this day, 14 years later? At what point do you decide, "Yeah, he's been an overall force for good," and cut the man some slack?
I know that Sun have contributed to a lot of open standards, etc., over the years. I'm not claiming that they are 100% knights in shining armor. But they are slowly moving in good directions. Given Murdoch's early environment, I'd expect him to be rather more a force for long-term good than some sort of Dark Lord of the Sith.
Our differences here (if you're one poster) are about religion. I value free (as in freedom) software as well. But I don't think I'm willing to be nearly as rigid as you. Quoting a manifesto is a far cry from evidence, and I simply don't see how he has harmed Debian, or GNU.
I can understand that viewpoint. I'm benefitting from it even as we speak.
/usr/include/linux/limits.h contains a line reading: /* # chars in a file name */
But if it's just some header lines, this will have no affect on you. That's *why* I doubt anyone will care very much. In some respects, particularly where there's some push for a move to GPL, as is the case with OpenSolaris, header commonality might even be regarded as desireable.
Some people, for instance, do small-scale systems administration. One source of information on some of the limits of their systems, such as how many characters can be used in a filename, are defined in a header file called, appropriately enough, limits.h. So, say,
#define NAME_MAX 255
It's probably better that users of different systems didn't have to keep track of different naming conventions, especially
as comments may not be provided. In a system I'm looking at right now, for instance, there's another line:
#define RTSIG_MAX 32
That's the maximum number of realtime signals supported. But there's no comment--you just have to know it. Standardization has it's points. Ideally, you learn this once, for one Unixy OS, and it's portable knowledge. It's a far cry from a wholesale rippoff, including algorithms and their implementations, such as SCO implied in the SCO-IBM insanity.
I seriously doubt very much that more than a tiny percentage Linux kernel developers would have a problem with a few header lines. I'm far from certain that copyright law should be pushed that far, lest developers have to use values with names like THIS_NAME_SO_WE_DO_NOT_INFRINGE. That's not in *anyone's* best interest.