Dark matter! Maybe dark energy! You're confusing yourself by equating dark matter and dark energy. Dark matter is one possible explanation for an observed deficiency in our understanding of gravitational effects of large scale objects. We don't know for sure that dark matter exists, but if it doesn't then there would have to be substantially odd forces at work. Dark matter is just an easy solution.
Dark energy is a mathematical placeholder name. There is an observed force which we can measure, but which we have no tested model to explain. We call this force dark energy.
When you say, "maybe dark energy," you demonstrate that you don't know what that phrase means. That's like saying, "maybe the solution to the problem is x!" X is just a variable name, not an answer to a question.
Even maybe we have to revise theories in astrophysics because we were wrong on something... Which happens all the time as our ability to measure and test the universe around us expands. This is an expected consequence of having more information. Someday, we'll marvel at how little we knew "back then" (e.g. today). For now, we have some very good ideas of how the universe in our local vicinity works, but no one expects to not be surprised by something new.
sigh, why do scientists think they are right now when their forbears were wrong? Why do you think that scientists are some alien species that don't understand basic logic? Of course astrophysics know that they have some things wrong today, but this is how we learn. We build solid ground upon which to base further ideas, and we constantly assail these ideas and their underpinnings in order to determine which parts are reliable enough to continue to bear the weight of many other theories.
Speaking of Astrophysics, if we can look into the sky and only see x millions of years back based off of light years, That's kind of broken statement. Let's try again, shall we? We can measure distance (in ways that range from simple triangulation to measuring red-shift). We know that light travels a certain distance in a certain amount of time. We therefore know how long light from an object would have traveled in order to get to us.
Now that's not quite "seeing x millions of years back," but it's close enough that I understand (I think) where you're going.
how do we know that we are not seeing the opposite side of the big bang curve? What is the "big bang curve?" Do you mean, "how do we know that we're not seeing light that started out at a time before the big bang?"
Well there are several easy reasons for that: 1) The big bang started as a singularity. You can't measure or view anything through a singularity. It's a cosmic wall through which no information can pass 2) If that were true, then the expansion of the universe would change as we looked out into deep space, and those distant objects would be moving toward us. This is not the case.
Of course, your question (at least, as I understand it) assumes that the big bang was "preceded" by a big crunch (the universe collapsing into a singularity). That may or may not be true, and we have no way to prove that it is or isn't, since we can't extract information about what happened before the singularity.
There's also the possibility that the background radiation from highly energetic galaxies prevented life from developing. That is, rather than a catastrophic event repeated at increasing period, the universe was simply hostile to our form of life as a rule. This might be borne out by recent evidence that small periods of Earthly extinction may be the result of Sol's eccentric orbit around the galactic nucleus taking it above the galactic plane, and away from the protective dust that surrounds us.
If the radiation we are exposed to from other galaxies were much more energetic when the universe was smaller by, say, a billion years, then it's quite possible that life which we would recognize simply could not have formed before we did, much less survive long enough for a gamma burst to take them out.
I don't know if the average citizen would even realize if their downstream bandwidth were boosted significantly. Of course they would. They would notice because the services available to them would change radically. It won't be the average consumer pushing out new media models, it will be the next generation of entrepreneurs and applications.
The real battle seems to be with the upstream. Face it, sending photos sucks. If I have to do any sort of large.ear deployment over my work's VPN, it sucks even more. Agreed.
My biggest problem with your entire comment may seem minor. But it's a major issue in mindset at the top of most every company now, and it's spread to the populace by the corporate-owned news media. That problem is the use of the word "consumer". So what you took away from my entire comment was one word, and your interpretation of how I might interpret that word (which happens to be wrong)... great. Perhaps we are well and truly screwed.
They sell to the high performance market, which (outside of trivially distributed applications like serving Web pages or rendering CG) NEEDS the kind of horsepower than Suns can crank out. [...] These are not (other then for a very few apps) parallelized applications. You're confusing distributed computing (something that works very well on piles of cheaper boxes) and parallel execution (something that requires very specialized hardware that's not just designed well for parallelization at the chip level, but throughout the architecture. Sun spends an awful lot of time building that kind of hardware.
Now, I'm no Sun fan boy. I work with Linux exclusively these days because that's the world I live in, and it's what my application wants, but if I were in the business of high-powered computation and I needed a box to run a parallel, non-distributed application, I would run it on a Sun.
There was a very short stint with IA64 as king for huge RAM jobs (128Gig+ in the box) for the last testing stages, but x86_64 overtook the cpus in speed awhile ago, and now you can get 256Gig+ hosts (we've recently ordered a few of those). We've got a small amount of 32bit linux boxes left too.. mostly for support of older tools. I think you live in the world where very fast, single-threaded, memory-hungry computation is king. In that world, Intel running Linux might be the right call. Just don't make the mistake of assuming that all the world is your back yard.
Our response, as tech-savvy consumers, must be several fold:
We must stop buying this media at home. I'm just about ready to cancel my cable subscription over this kind of abuse.
We must work hard to seek out alternative media outlets that want to foster our rights, rather than abuse them for profit.
We must work to reform the campaign finance rules that allow Congress to be bribed into allowing such horrendous abuses of their constituencies.
If we do not do these things, then we have no right to complain about the rampant abuse of the consumer. After all, if we don't care enough to take these steps, who will?
most people just don't care about them. Tom's Hardware is not going to be reviewing them for the enthusiast market, for example, they are waaaaay out of that range. Same shit with the Power 6. Great chip, coming never to a desktop near you. These specialised high end products are just not of mass interest if for no other reason than price. Which Sun has never cared about. They sell to the high performance market, which (outside of trivially distributed applications like serving Web pages or rendering CG) NEEDS the kind of horsepower than Suns can crank out.
Even if that cheap server breaks right at the end of its warranty, it is still a money saver, a big one. This is almost always a red herring. Price of a system is rarely a company's most significant cost (within an order of magnitude) when you're dealing with high performance computing. It's the people and the data vendor relationships that usually cost you the bulk of your outlay. Hardware of just about any sort is fairly cheap by comparison.
I understand the market for enterprise systems, I also understand that it is small. Hrm... small? Well, not really. It's got a low number of players, but those players have a voracious appetite for processing power.
Don't get me wrong. Most of a large financial house doesn't need a Sun server. However, those who do (e.g. quants) really do. Same goes for government, biotech, military contractors, etc.
Most of them seem to be off-cycle, though Again, I think the list you're looking at isn't "known extinction events" so much as "top extinction events." This research involved a very specific, repeating pattern of relatively small extinctions.
not only that but mass extinctions happened a lot earlier than that and with a far less predictable pattern. which leaves us to wonder why this cycle is this recent? why isn't there a cycle like this stretching back over a billion years? I haven't read the original paper, and the article is thin on details, so I'm not sure exactly how many events they considered... HOWEVER, I do not think you're correct about the conditions being static across spans of billions of years.
Our sun (Sol) is a member of a cluster of stars that were birthed by a nebula of gas and dust around the same time. That cluster (like all stellar nurseries within a galactic disk) tended to break apart as time went on, due to the difference in orbital speeds around the center of the galaxy (regions of the nebula closer to the center move faster, and regions further out, slower). This nebula and the proximity of other stellar systems almost certainly provided some shielding from dangerous intergalactic muons, and the whole nebula would have started with a similar orbital eccentricity as Sol. So, over millions of years, as the nebula was pulled apart by galactic tides, our protection has thinned. The upsetting part (if you get upset about events that could affect us in tens of millions of years) is that it probably has more thinning to go, and our exposure to these extra-galactic particles is probably increasing each cycle.
It's a perfectly reasonable hypothesis if one accepts the premise that mass extinctions have an approximately 62 million year period. From Wikipedia, the last 6 extinction events happened 65 million years ago, 200 million years ago, 251 million years ago, 360 million years ago, 444 million years ago, and 488 million years ago. You're looking at the largest of the extinction events. This theory is attempting to explain a particular set of events which result in only an approx. 10% drop in biodiversity, and which are about 60ish million years apart.
The KT event, for example, had a much larger impact on biodiversity but happened off-cycle, and is pretty clearly the result of a specific meteor strike that we already know about.
Other events may have been volcanic or meteoric or the result of something we didn't know about.
All extinction events being triggered by only one type of external condition was never very likely.
Only 7 million years from now, for all you long range planners. [...] It's a perfectly reasonable hypothesis, though it'll be a while before we can test it. I don't think it'll be as long as you think. Within 100 years we will probably have the ability to send very small probes out of our solar system at speeds which measure substantial fractions of the speed of light. At that point, we can start sending out probes to analyze the galactic "weather" of regions that the Earth will occupy further down the line. It's still a slow process (requiring decades to centuries for results), but it's probably not as lengthy a process as you're thinking it is.
I asked if *you* could use it more pretentiously. It goes without saying that Eric Raymond can use it more pretentiously! Oh, you were trolling. I understand.
The same is true within the plethora of Windows variants and Linux variants. This does not add value to any of those OSes beyond what one would already expect. BSD *is* a Unix derivative (and, at this point, Unix is also a BSD derivative, ever since System 5 rolled in BSD 4 features), so it makes perfect sense for a BSD-based system (MacOS/X) to aspire to compliance with other OSes of its ilk.
You seem to be making an assumption, however, that portability across "species" of OS lineage has some value. I'm really not sure that it does beyond the basic level of API standards compliance that any modern POSIX system (be it Unix, Linux or even Windows) maintains. Modern OSes are just too different, and an effort to create universal portability will ultimately result in very poor utilization of at least n-1 of the OSes out there (c.f. Java). It won't be too long, I think, before we're ready for a new round of standardization around core OS features to layer on top of POSIX, as there's certainly now a new high-water-mark for "stuff you expect to work the same everywhere," but much of the OS still falls well above that mark and remains a moving target as OSes continue to evolve. IMHO portability adds value to Unix certified OSes for all sorts of reasons. Within the Unix family, sure. Linux has many Unixisms, but it's a fundamentally different OS with a different audience (some overlap, of course), and a fundamentally different set of objectives. I see no value in a Linux system trying to be Unix, and more than there would be value in a Unix system seeking a Linux-related certification.
I have ported applications from one Unix certified OS to another, I have also ported applications from Unix certified OSes to Linux and the latter proposition can involve considerably more work. Of course it does! Just as porting to Windows or VMS would. You're porting to a different OS family. You should expect that to be hard work. However, the differences between the approaches of each OS yields a wealth of advantages. An OS flavor evolves to best support the audience for that OS flavor. Unix has an audience which overlaps with Linux, but is not the same, and that brings it a unique set of priorities.
Optimally all of the Unix and Unix-like OSes like Linux should comply with a single global standard giving one the ability to migrate from a low to medium end OS like Linux to a high end OS like many of the Unix certified ones are with almost zero effort. And, of course, here you show your biases. By classifying Linux as a "low to medium end OS," you clearly demonstrate that you entered the conversation with a deeply held notion that Linux is not a peer in the OS marketplace. This is demonstrably false, as the number of high-performance, high-uptime requirement environments maintained by IBM's Global Services could tell you in a heartbeat. I'll give you a hint: AIX isn't in the top spot.
I work in an environment where 100 Linux boxes is considered a "small resource" and annual allowances for downtime are measured in heartbeats. Linux is not the ideal choice for every mission critical environment (again, each OS has its domain), but in many it's the only rational choice.
For all that I despise the tactic, I have to admit that it's a clever little hack. Yeesh, could you use "hack" more pretentiously? Pretentiously?
I mean the word in exactly the same way that I mean it when I say that the GPL is one of the cleverest hacks ever perpetrated on the copyright system. It uses the system itself as leverage to bypass the design goals of the system, which is not only smart, but hadn't been done in the 200 years of US copyright law that preceded it. c.f. the Jargon File's "The Meaning of 'Hack'":
Hacking might be characterized as 'an appropriate application of ingenuity'. Whether the result is a quick-and-dirty patchwork job or a carefully crafted work of art, you have to admire the cleverness that went into it.
I've posted the easy, but radical solution before: allow patents on just about anything, and make them all last at least 20 years, but grant VERY few per year (order of 1000); do not allow transfer of patent ownership; and set a small cap on the number of people allowed per patent as owners (e.g. max of 4 inventors), all of whom must be people, not corporations. With the addition of a per-discipline review board of peers, this would turn patents into a strong incentive for innovation once again.
The harder, but less radical solution would be:
Step 1: Patents are effective on their date of granting, not filing.
Step 2: Once a patent is issued, place it into a peer review process that is semi-open to the public (e.g. anyone with passable credentials in a field can get in as a peer reviewer), via the Web.
Step 3: Any patent which is brought into question by the peer reviewers is revoked, pending possible appeal by the patent holder via the courts. This places the legal burden on the holder to defend a patent that didn't stand up to peer review.
Step 4: Shorten patent duration on a per-discipline basis, with the duration in each discipline being roughly 3-4 times the amount of time that it takes to bring a new product to market, with a cap of 20 years. In software, this would mean about a 6 year patent duration (18 month product cycle times 4).
But Apple's chief patent counsel said the US patent system was 'not broken' and 'not in crisis,' calling it 'the best in the world.'"
Something from the US, "the best in the world"? How dare they say such a thing!
This is infinitely worse than anything Microsoft has ever done. OK, I'm confused. How is Google doing the wrong thing, here (per your subject, "It's Official: Apple & Google are Evil")?
Google is coming out against the current patent system, and you attack them. I'm sure if they came out for it, you would attack them. What could they possibly say that would make you happy?... On second thought, never mind. Making happy those who will take any opportunity to attack a company that tries to do the right thing is probably always a bad idea. Let them stew in their misery.
They key issue, can anyone implement the standard directly without payments, without agreements without any restrictions? MSFT can very well say, there is no payment but all implementors should sign some agreement with us. Then there could be a clause that could revoke the agreement. You're not thinking deviously enough. What they REALLY want to do is have all of the most popular Web data formats require the use of their patents, and then issue a blanked right to use those patents for free to anyone... but in a way that's not GPLv3 compatible.
This is Microsoft's dream because you can't contest it in court. The agreement you're violating if you mix this technology with GPLv3 code is NOT the agreement with Microsoft, but the GPLv3! You would have to sue the FSF in order to use Microsoft's image format in your GPLv3 code.
For all that I despise the tactic, I have to admit that it's a clever little hack.
Unfortunately am in Canada and Bittorrent has been banned by the Internet Police over here, so we're not allowed to download files. Wow. That's totally freaking insane!
Next time I download an OS via BT, I'll think of you... I'm really sorry, man.
That said, I'm a little surprised and disappointed at Slashdot's reaction to this documentary. Someone does a documentary about file sharing, puts it up on BT and we attack them for it... sad. I would have thought we'd be glad to see that someone is finally starting to smell the new media. Do they want their documentary seen? Of course, they do, but if this works out, you know there will be thousands of fairly smart people thinking, "OK, now how do we make that a business?"
The age of the fight between content creators and peer-to-peer sharing needs to come to an end. The age of the peer-to-peer media empires is long overdue.
Are you aware that the Open Group Unix specifications go a lot further than POSIX? Yes, which is why I made the distinction between merely being POSIX compliant and being Unix certified. The two have radically different scopes.
That said, my bit at the end about the day having dawned where POSIX might need a next pass was aimed at a very post-Unix world where the layer above POSIX that's reasonably standard across Unix-like OSes at this point involves things like networking tools, graphics and other things that were never part of POSIX.
A Unix certification is a bit more than a moniker. It means that the level of software portability between Unix 03 compliant systems is guaranteed to be very high. That may not be important to you but to companies/corporations seeking to reduce costs and development times and to achieve the maximum level of reliability and portability in their business critical software a Unix 03 certification has meaning. All this means is that, within the plethora of Unix variants, there's high portability.
The same is true within the plethora of Windows variants and Linux variants. This does not add value to any of those OSes beyond what one would already expect. BSD *is* a Unix derivative (and, at this point, Unix is also a BSD derivative, ever since System 5 rolled in BSD 4 features), so it makes perfect sense for a BSD-based system (MacOS/X) to aspire to compliance with other OSes of its ilk.
You seem to be making an assumption, however, that portability across "species" of OS lineage has some value. I'm really not sure that it does beyond the basic level of API standards compliance that any modern POSIX system (be it Unix, Linux or even Windows) maintains. Modern OSes are just too different, and an effort to create universal portability will ultimately result in very poor utilization of at least n-1 of the OSes out there (c.f. Java). It won't be too long, I think, before we're ready for a new round of standardization around core OS features to layer on top of POSIX, as there's certainly now a new high-water-mark for "stuff you expect to work the same everywhere," but much of the OS still falls well above that mark and remains a moving target as OSes continue to evolve.
Also keep in mind that although no Linux or BSD flavor other than OS X has gone for actual certification apparently many Linux distributions for example still make sure they are more or less Unix compliant and they do it using Open Group test suites. So even if no Linux distro has officially applied for certification it looks to me as if they are keeping their options open. Linux developers like to make sure that their software is compatible with as many applicable standards as is reasonable. I don't think anyone wants a Linux variant that's shoe-horned into meeting the full breadth of Unix certification, though. If you really needed Unix, you'd just get Unix, not Linux.
Append only files have not been required in my experience. What is required is that there be no ability to overwrite a previously written file by the team that is sending the log data. The one is a way to get the other... at least partially. Physical and electronic security and partitioning of roles gets you the rest of the way.
This can be done a number of ways, but the easiest method is to transmit the data in a way that the server chooses the filename, not the client. I'm not sure how filenames enter into it, since you don't give the application people access to the log host anyway.
syslog works for most data, but not all. Linux is one of the only Unix based systems that puts sulog through syslog. The failed logins log is much more difficult, as is the wtmp data. There is a "syslog Working Group" that's working on that and other problems. I don't know if syslogng supports any of their proposals yet, though.
As to writing periodically to a optical media, I wouldn't worry quite so much about that. It's very important to be able to talk about your risk exposure profile. When you can say that the exposure to electronic subversion of the logs (regardless of how hard you make that, via electronic security) ends when the data is written to optical disk, you can make a much stronger case for the data being functionally write-once.
I would instead worry more about the encrypting all that security data while in network transit. That would only be necessary if you transmit sensitive data in the logs. For example, if a healthcare company wrote client data to their in-house application server's logs, then the logs would have to be subject to the same security constraints as every other piece of sensitive data. This is as far as I know, but my PCI exposure is tangential, and I haven't read the requirements first-hand.
Authenticity is also an issue to be concerned with. How do you know that the event that got inserted into the log really came from that box, and not some random other server? Typically, you don't care (as long as you have the valid entries, any invalid ones are typically just noise), sometimes you do. I'm not aware of any PCI requirement for authentication in general, but for some purposes that may be there. I do think that syslogng might provide a means of minimal authentication, but I'm not certain about that.
When working with PCI, know which DSS you are on, 1.0 or 1.1. A fair point.
Ah, sure! but as soon as OpenBSD starts porting the broadcom driver in CVS, the linux guys all scream foul play! You're clearly not responding to what I said, since I made no judgment about anyone's behavior. I was only pointing out that we are, in fact, talking about the Linux driver developers, not the BSD driver developers.
Optical is the right choice here, but you need to understand the PCI requirements and their most common interpretation VERY clearly. What you will probably end up with is something like this:
Logs are written over the network (e.g. syslog)
Logging host, which is locked down, and has no access from the infrastructure that it's performing logging for other than the incoming log data itself.
Logging host writes the logs locally to files which are marked as append-only by the OS (Linux can do this)
The logs are then written periodically (e.g. once per hour) to optical media.
Add redundant logging hosts to taste (3 is a nice number for validation purposes).
don't get defensive about your favorite company. Google's not my favorite company... though if pressed, I might finger them as my favorite for-profit, public corporation. Red Hat would come in a close second; IBM third. These companies have pushed the envelope of open source software development more than any other public corporations. This funnels the massive resources of the stock market directly into improving the software that we Slashdotters feel is so important. What's more, all three have gone to great lengths to preserve open source software development on the political and legal frontiers.
I don't trust Google per se, but I'd say I have more trust in them and their motivations, purely based on their prior actions, than I do in any other public corporation out there.
Slashdot Mirror
Dark energy is a mathematical placeholder name. There is an observed force which we can measure, but which we have no tested model to explain. We call this force dark energy.
When you say, "maybe dark energy," you demonstrate that you don't know what that phrase means. That's like saying, "maybe the solution to the problem is x!" X is just a variable name, not an answer to a question. Even maybe we have to revise theories in astrophysics because we were wrong on something... Which happens all the time as our ability to measure and test the universe around us expands. This is an expected consequence of having more information. Someday, we'll marvel at how little we knew "back then" (e.g. today). For now, we have some very good ideas of how the universe in our local vicinity works, but no one expects to not be surprised by something new. sigh, why do scientists think they are right now when their forbears were wrong? Why do you think that scientists are some alien species that don't understand basic logic? Of course astrophysics know that they have some things wrong today, but this is how we learn. We build solid ground upon which to base further ideas, and we constantly assail these ideas and their underpinnings in order to determine which parts are reliable enough to continue to bear the weight of many other theories. Speaking of Astrophysics, if we can look into the sky and only see x millions of years back based off of light years, That's kind of broken statement. Let's try again, shall we? We can measure distance (in ways that range from simple triangulation to measuring red-shift). We know that light travels a certain distance in a certain amount of time. We therefore know how long light from an object would have traveled in order to get to us.
Now that's not quite "seeing x millions of years back," but it's close enough that I understand (I think) where you're going. how do we know that we are not seeing the opposite side of the big bang curve? What is the "big bang curve?" Do you mean, "how do we know that we're not seeing light that started out at a time before the big bang?"
Well there are several easy reasons for that: 1) The big bang started as a singularity. You can't measure or view anything through a singularity. It's a cosmic wall through which no information can pass 2) If that were true, then the expansion of the universe would change as we looked out into deep space, and those distant objects would be moving toward us. This is not the case.
Of course, your question (at least, as I understand it) assumes that the big bang was "preceded" by a big crunch (the universe collapsing into a singularity). That may or may not be true, and we have no way to prove that it is or isn't, since we can't extract information about what happened before the singularity.
Here we are -> ( *Bang* )
More dumb observations later.
There's also the possibility that the background radiation from highly energetic galaxies prevented life from developing. That is, rather than a catastrophic event repeated at increasing period, the universe was simply hostile to our form of life as a rule. This might be borne out by recent evidence that small periods of Earthly extinction may be the result of Sol's eccentric orbit around the galactic nucleus taking it above the galactic plane, and away from the protective dust that surrounds us.
If the radiation we are exposed to from other galaxies were much more energetic when the universe was smaller by, say, a billion years, then it's quite possible that life which we would recognize simply could not have formed before we did, much less survive long enough for a gamma burst to take them out.
Then again, both could be true.
Now, I'm no Sun fan boy. I work with Linux exclusively these days because that's the world I live in, and it's what my application wants, but if I were in the business of high-powered computation and I needed a box to run a parallel, non-distributed application, I would run it on a Sun. There was a very short stint with IA64 as king for huge RAM jobs (128Gig+ in the box) for the last testing stages, but x86_64 overtook the cpus in speed awhile ago, and now you can get 256Gig+ hosts (we've recently ordered a few of those). We've got a small amount of 32bit linux boxes left too.. mostly for support of older tools. I think you live in the world where very fast, single-threaded, memory-hungry computation is king. In that world, Intel running Linux might be the right call. Just don't make the mistake of assuming that all the world is your back yard.
If we do not do these things, then we have no right to complain about the rampant abuse of the consumer. After all, if we don't care enough to take these steps, who will?
Don't get me wrong. Most of a large financial house doesn't need a Sun server. However, those who do (e.g. quants) really do. Same goes for government, biotech, military contractors, etc.
Our sun (Sol) is a member of a cluster of stars that were birthed by a nebula of gas and dust around the same time. That cluster (like all stellar nurseries within a galactic disk) tended to break apart as time went on, due to the difference in orbital speeds around the center of the galaxy (regions of the nebula closer to the center move faster, and regions further out, slower). This nebula and the proximity of other stellar systems almost certainly provided some shielding from dangerous intergalactic muons, and the whole nebula would have started with a similar orbital eccentricity as Sol. So, over millions of years, as the nebula was pulled apart by galactic tides, our protection has thinned. The upsetting part (if you get upset about events that could affect us in tens of millions of years) is that it probably has more thinning to go, and our exposure to these extra-galactic particles is probably increasing each cycle.
The KT event, for example, had a much larger impact on biodiversity but happened off-cycle, and is pretty clearly the result of a specific meteor strike that we already know about.
Other events may have been volcanic or meteoric or the result of something we didn't know about.
All extinction events being triggered by only one type of external condition was never very likely.
You seem to be making an assumption, however, that portability across "species" of OS lineage has some value. I'm really not sure that it does beyond the basic level of API standards compliance that any modern POSIX system (be it Unix, Linux or even Windows) maintains. Modern OSes are just too different, and an effort to create universal portability will ultimately result in very poor utilization of at least n-1 of the OSes out there (c.f. Java). It won't be too long, I think, before we're ready for a new round of standardization around core OS features to layer on top of POSIX, as there's certainly now a new high-water-mark for "stuff you expect to work the same everywhere," but much of the OS still falls well above that mark and remains a moving target as OSes continue to evolve. IMHO portability adds value to Unix certified OSes for all sorts of reasons. Within the Unix family, sure. Linux has many Unixisms, but it's a fundamentally different OS with a different audience (some overlap, of course), and a fundamentally different set of objectives. I see no value in a Linux system trying to be Unix, and more than there would be value in a Unix system seeking a Linux-related certification. I have ported applications from one Unix certified OS to another, I have also ported applications from Unix certified OSes to Linux and the latter proposition can involve considerably more work. Of course it does! Just as porting to Windows or VMS would. You're porting to a different OS family. You should expect that to be hard work. However, the differences between the approaches of each OS yields a wealth of advantages. An OS flavor evolves to best support the audience for that OS flavor. Unix has an audience which overlaps with Linux, but is not the same, and that brings it a unique set of priorities. Optimally all of the Unix and Unix-like OSes like Linux should comply with a single global standard giving one the ability to migrate from a low to medium end OS like Linux to a high end OS like many of the Unix certified ones are with almost zero effort. And, of course, here you show your biases. By classifying Linux as a "low to medium end OS," you clearly demonstrate that you entered the conversation with a deeply held notion that Linux is not a peer in the OS marketplace. This is demonstrably false, as the number of high-performance, high-uptime requirement environments maintained by IBM's Global Services could tell you in a heartbeat. I'll give you a hint: AIX isn't in the top spot.
I work in an environment where 100 Linux boxes is considered a "small resource" and annual allowances for downtime are measured in heartbeats. Linux is not the ideal choice for every mission critical environment (again, each OS has its domain), but in many it's the only rational choice.
I mean the word in exactly the same way that I mean it when I say that the GPL is one of the cleverest hacks ever perpetrated on the copyright system. It uses the system itself as leverage to bypass the design goals of the system, which is not only smart, but hadn't been done in the 200 years of US copyright law that preceded it. c.f. the Jargon File's "The Meaning of 'Hack'": Hacking might be characterized as 'an appropriate application of ingenuity'. Whether the result is a quick-and-dirty patchwork job or a carefully crafted work of art, you have to admire the cleverness that went into it.
I've posted the easy, but radical solution before: allow patents on just about anything, and make them all last at least 20 years, but grant VERY few per year (order of 1000); do not allow transfer of patent ownership; and set a small cap on the number of people allowed per patent as owners (e.g. max of 4 inventors), all of whom must be people, not corporations. With the addition of a per-discipline review board of peers, this would turn patents into a strong incentive for innovation once again.
The harder, but less radical solution would be:
Step 1: Patents are effective on their date of granting, not filing.
Step 2: Once a patent is issued, place it into a peer review process that is semi-open to the public (e.g. anyone with passable credentials in a field can get in as a peer reviewer), via the Web.
Step 3: Any patent which is brought into question by the peer reviewers is revoked, pending possible appeal by the patent holder via the courts. This places the legal burden on the holder to defend a patent that didn't stand up to peer review.
Step 4: Shorten patent duration on a per-discipline basis, with the duration in each discipline being roughly 3-4 times the amount of time that it takes to bring a new product to market, with a cap of 20 years. In software, this would mean about a 6 year patent duration (18 month product cycle times 4).
Something from the US, "the best in the world"? How dare they say such a thing!
This is infinitely worse than anything Microsoft has ever done. OK, I'm confused. How is Google doing the wrong thing, here (per your subject, "It's Official: Apple & Google are Evil")?
Google is coming out against the current patent system, and you attack them. I'm sure if they came out for it, you would attack them. What could they possibly say that would make you happy?
This is Microsoft's dream because you can't contest it in court. The agreement you're violating if you mix this technology with GPLv3 code is NOT the agreement with Microsoft, but the GPLv3! You would have to sue the FSF in order to use Microsoft's image format in your GPLv3 code.
For all that I despise the tactic, I have to admit that it's a clever little hack.
Next time I download an OS via BT, I'll think of you... I'm really sorry, man.
That said, I'm a little surprised and disappointed at Slashdot's reaction to this documentary. Someone does a documentary about file sharing, puts it up on BT and we attack them for it... sad. I would have thought we'd be glad to see that someone is finally starting to smell the new media. Do they want their documentary seen? Of course, they do, but if this works out, you know there will be thousands of fairly smart people thinking, "OK, now how do we make that a business?"
The age of the fight between content creators and peer-to-peer sharing needs to come to an end. The age of the peer-to-peer media empires is long overdue.
That said, my bit at the end about the day having dawned where POSIX might need a next pass was aimed at a very post-Unix world where the layer above POSIX that's reasonably standard across Unix-like OSes at this point involves things like networking tools, graphics and other things that were never part of POSIX.
The same is true within the plethora of Windows variants and Linux variants. This does not add value to any of those OSes beyond what one would already expect. BSD *is* a Unix derivative (and, at this point, Unix is also a BSD derivative, ever since System 5 rolled in BSD 4 features), so it makes perfect sense for a BSD-based system (MacOS/X) to aspire to compliance with other OSes of its ilk.
You seem to be making an assumption, however, that portability across "species" of OS lineage has some value. I'm really not sure that it does beyond the basic level of API standards compliance that any modern POSIX system (be it Unix, Linux or even Windows) maintains. Modern OSes are just too different, and an effort to create universal portability will ultimately result in very poor utilization of at least n-1 of the OSes out there (c.f. Java). It won't be too long, I think, before we're ready for a new round of standardization around core OS features to layer on top of POSIX, as there's certainly now a new high-water-mark for "stuff you expect to work the same everywhere," but much of the OS still falls well above that mark and remains a moving target as OSes continue to evolve. Also keep in mind that although no Linux or BSD flavor other than OS X has gone for actual certification apparently many Linux distributions for example still make sure they are more or less Unix compliant and they do it using Open Group test suites. So even if no Linux distro has officially applied for certification it looks to me as if they are keeping their options open. Linux developers like to make sure that their software is compatible with as many applicable standards as is reasonable. I don't think anyone wants a Linux variant that's shoe-horned into meeting the full breadth of Unix certification, though. If you really needed Unix, you'd just get Unix, not Linux.
I don't trust Google per se, but I'd say I have more trust in them and their motivations, purely based on their prior actions, than I do in any other public corporation out there.