That's not correct. Nobody owns all of their own fiber, no matter how big they are. Not even the Tier 1's. They may own some of it, but the majority is leased. In fact, when you purchase fiber between two locations, even though you purchase it from one party, they may in turn own some segments, while leasing various segments from multiple different providers. Even though the bulk of the costs of fiber are one time, the high capital costs result in a leasing model to cover the costs over a long period of time.
People have been arguing as you are that x86's bloated CISC instruction set was inferior to a cleaner RISC architecture for the last 20+ years. Nobody has ever proven that the elegance of the instruction set matters with hard data though.
What evidence we do have goes against that argument.
The only evidence that we have is that the benefits of commoditization and economies of scale often outweigh any architectural advantages. The fact that x86 incorporated many elements of RISC would also demonstrate its value.
Apple machines used a cleaner RISC architecture for a while in the desktop space. They never performed any better than equivalent x86 based machines, and in the end Apple abandoned RISC and moved to x86.
Manufacturing processes simply trumped architectural differences. PowerPC's have never been manufactured on anywhere near the scale of x86.
Intel came out with a cleaner RISC based instruction set that that the Itanium line uses. If x86 was really as bad as you say, Itanium chips would be running circles around the x86 based server chips provided by both Intel and AMD. That isn't happenning.
Itanium is EPIC, not CISC. It is the exact opposite of RISC. It may not be running circles around x86, but that may be due to compilers not yet being advanced enough to take full advantage of the architecture. We may still see this change in the future.
You've been able to use the linux adobe flash plugin with a native browser for quite some time now. Works just as well as it does in Linux (not necessarily saying much). The FreeBSD handbook is your friend:
I would beg to differ. The package management is just as good if not better than what's available in Linuxland, so there's no great difficulty in setting it up as a good desktop system.
Having excellent support for many non-x86 platforms, as well as having a small footprint make it a great choice for older hardware. I currently have it installed on on my old UltraSparc and Alpha workstations.
OpenBSD contributes more than just OpenSSH to other OS'es. Aside from pushing hardware manufacturers to open up their documentation, they've also reverse engineered drivers that have made their way into the other BSD's and even Linux (remember the whole Atheros? issue last year). Whereas many Linux distributions and the other BSD's have made compromises with proprietary drivers and binary blobs, OpenBSD still pushes for true open source.
PF and CARP also make OpenBSD a superior router platform to any IPTables based setup any day. You may be surprised how popular it is in the data centre.
Unlike Minix, OpenBSD's niche has a place in real world usage.
Cogent is a Tier 1 network service provider (weather or not Sprint and L3 want it to be).
It's apparent from this announcement that Cogent and Sprint have never had settlement free peering in place, aside from the 3 month trial. Therefore, Cogent has never been a Tier 1 network service provider, as clearly they do not have settlement free peering to all other Tier 1's.
Cogent offers great service at an unbeatable price (4-5 USD per Mbps as opposed to the 15-20 or so Sprint and competitors are charging).
They do offer service, but it's definitely not great. Nobody who takes their uptime seriously would ever single-home on Cogent.
Cogent is the type of NSP we want as a Tier 1. A very strong backer of Net Neutrality, and no intention of trying to get into the entertainment business unlike Verizon, AT&T, etc. Cogent has a goal of offering the best service at the lowest price (the end result being realistically moving the US forward in terms of available bandwidth).
That's a pretty noble goal, but they're far from meeting it. They offer the worst service at the lowest price. You're simply getting what you pay for.
As much as Sprint would like to position itself as a provider for Cogent, it's not. Sprint is a peer for Cogent with Cogent being an equivalent size of the current Sprint network, and larger than many of Sprints other peers.
Size is not the only criteria for peering. Peering is based on mutual benefit. Many larger content providers will exchange traffic with smaller eyeball networks as they do see mutual benefit from peering (i.e. not having to pay their upstreams). Clearly, Sprint does not see the benefit of peering with Cogent, and they have no obligation to peer with them.
The idea that Sprint doesn't get as much out of peering with Cogent as Cogent does peering with Sprint is absurd and PR propaganda to try and look like this was anything other than a power-play to keep a competitor at bay.
As Cogent obviously does not peer with Sprint, your assertion that Sprint gets as much out of peering with Cogent as Cogent gets from Sprint, especially in light of the great disparity in traffic exchange, is completely absurd.
Nobody who initiates a settlement-free peering agreement intends to pay for a transit connection if the trial period doesn't work out. If there was a serious chance that Cogent could be convinced to buy service from Sprint, Sprint would never have agreed to peering since the two types of connection are mutually incompatible.
From the sounds of it, Cogent was previously purchasing transit from Sprint. It's not presumptious to resume the prior relationship, after the 3 month trial did not work out.
Then too, Sprint probably decided to block Cogent's routes from the rest of their peers. Otherwise I expect Cogent would simply have routed around the downed connections. That was a step too far, one which makes Sprint a dangerous selection as your sole carrier as well.
That's not how Internet routing works. Nobody is blocking Cogent's routes. The reason why Cogent doesn't have any other routes to Sprint is because they are attempting to get all of their traffic through peering, and do not have any transit providers.
In peering, you only advertise routes for your own network. As such, none of the other networks that Cogent peers with would advertise Sprint's routes to them. This is why without a direct connection to Sprint, Cogent does not have any routes to Sprint, and does not have any means of routing around the down connections.
No it doesn't. Your assuming time doesn't matter and that the hardware follows the documentation.
Developers are willing to contribute their time. This is the whole point of open source. If the hardware doesn't follow the documentation, then the documentation should be better.
Time does matter because if a person releases a product to market without a pre-existing Linux driver then Linux people can't use it until developers purchase it and begin hacking on it.
That's rather short sighted. Sure, the driver comes out a few months earlier, but when the company stops supporting that product, who's to maintain the code when nobody has a clue what it does? Without documentation, the drivers are meaningless. Perpetuity is a more than worthwhile trade-off for a few months.
Without assistance from the people who actually worked on developing the hardware then your going to end up doing a lot of trial and error to figure out what is wrong.
And there is somehow less trial and error when having to reverse engineer a driver?
Your assuming that such documentation exists or that it's even possible for that manufacturer to create, and that they can afford to create such documentation. Not everybody has all the time and money in the world to create extensive documentation for their products.
Are you joking? Documentation is essential for any engineering process, be it hardware or software. If they don't have such documentation, there is a serious flaw in their internal processes. If such a company really didn't have any such documentation, they would be doing themselves a huge favour by writing some, as their product quality would see substantial improvement.
Not everybody has the budget and experience that Intel and AMD have...
So this is why small Taiwanese manufacturers like Areca, who are relatively new in the market, are doing a much better job with documentation and open source drivers than old giants like Adaptec or LSI?
The reality of the situation that is unless you have the attention of OEMs and have people that are willing to work on the inside with the manufacturers to work on documentation and drivers then isn't going to get the same level of attention that even Linux gets.
That's exactly what people are trying to accomplish here.
Because of the realities surrounding developing hardware having working, open source, Linux drivers is the best documentation that your going to get, and in fact are often superior.
Absolute nonsense. The only reason why the drivers would be the best documentation we can get, is because people like you are willing to accept it.
With 10Gb circuits increasing in popularity and replacing OC48's, we're probably still in good shape, until 2012. Perhaps, by then the Cisco CRS-1 won't seem so overpriced.
In the meantime, everyone can buy Juniper:)
Rather than real estate, you will usually be renting a locked cabinet with network and power drops.
Also, infrastructure is very expensive. Redundant generators, UPS'es, and network connections are all major costs, not to mention the staff you would need to support it all. It usually ends up being much cheaper just to rent a cabinet or two in a data center. There are actually much fewer data centers than you would be lead to believe by different hosting companies' website. The vast majority are renting space in somebody else's data center, which they are trying to pass off as their own.
One thing that a lot of people may not realize, is that data centers should be near other data centers or bandwidth carriers, usually in an internet building or carrier hotel. You want to have easy access to your upstreams and peers, so you're not paying a fortune to run links to them.
Now, you can just have a fiber circuit that runs to a carrier hotel, but then you're adding an additional point of failure, and uptime is everything in this business. You can mitigate that somewhat by having redundant fiber circuits that take different paths, which can often be hard to find and/or difficult/expensive to run.
ServerBeach is self-managed so it doesn't directly compete with Rackspace. Peer 1 Dedicated Hosting is where you'd go for managed hosting from the same company as ServerBeach.
* Somebody put BSD code into GPL code, and did not keep the original copyright tags. A totally isolated incident. Whether or not it's isolated, it's clear from all the recent discussion that many people have the misconception that this is permissible under the BS license.
* Althought the GPL coder was technically wrong, there was no harm done, and the situation has been fixed.
Incorrect. Reyk's code is still being distributed by MadWifi with the GPL license wrapped around his code, when he has explicitly stated that this is against his wishes, and most likely illegal.
* The BSD community has been having a screeming bitch-fest for weeks, making all kinds of insane accusations and threats. The BSD community is trying to correct common misconceptions by the GPL community, and also illustrate the uncooperative nature of the GPL license, and the practice of licensing modifications to BSD code under the GPL license.
* Although the BSD community has no problem at with BSD code hidden in a msft binary, they get their panties all in a wad about BSD code put into Linux. That is because MS has complied with the BSD license whereas the GPL developer in question has not. MS also doesn't assert itself as open source, or complain about the lack of cooperation with open source efforts. As the GPL community does, and is behaving starkly in contrast with these intentions, they are guilty of hypocrisy in this regard whereas MS is not.
* Theo de Raadt is so bitchy and irratating that even his fellow BSD zealots can't stand him much of the time. And even though Theo is clearly unqualified about legal matters, much of the BSD is getting behind on this. Theo is first and foremost a developer, and is highly respected in that regard. He is not concerned with his popularity rating or politics, and neither is most of the OpenBSD community. He is very vocal about his personal opinions, but everyone is entitled to their opinions. As his opinions are more likely to contribute than to take away from his work, they are not matters of concern for OpenBSD users.
Being as literate as you are, you should have read by now that it was only Sam Leffler's code that was dual-licensed, and Reyk's code was NOT wherein lies the problem.
Removing the one and only license that Reyk chose for his code is a legal problem.
If strings is too obscure and not native enough for you, how about notepad.exe? Anything capable of reading text within a binary file will work.
You will be using BSD software whether you like it or not, unless you stop using the Internet.
I am now convinced that Linux zealots are *less* knowledgeable than BS zealots, or even apple zealots. I will still use Linux when it's the right tool for the job, even though the community seems bloody ignorant.
That ZDNet article is completely flawed, and should be ignored. It's rated -15 for a reason.
For one, it assumes that a RAID 5 array will become corrupt due to bad sectors. Only the files that are on those sectors will be corrupted, and since they are on multiple drives, the parity drive will allow you to recover that data.
In addition, it ignores the fact that the hard disks will simply mark those sectors as bad and will no longer use them. Not to mention, drives these days come with spare sectors that automatically replace bad ones.
Drive capacity and the probability of a RAID 5 array becoming corrupted have no correlation whatsoever.
Just because it only costs $250-$300 to manufacture doesn't mean that's all the phone costs. There are a multitude of other costs involved here, including research & development, warranty/maintenance, support, distribution, sales, marketing, yada yada yada. Find some accurate estimates on what the total real cost for final delivery is to the end user, and that it's substantially less than $500 and you might have a point.
Why does nobody understand that the code that was originally written under the BSD license never had the GPL license added to it by the author? Reyk, the author, has explicitly said this, and Theo has openly vocalized this.
The reason that the code is dual-licensed is because the original code is BSD licensed and ONLY the MODIFICATIONS to it are GPL licensed. The original code itself is still under the BSD license, and not dual licensed by the original author.
Also, just because companies can use BSD code in their closed-source proprietary projects, and not distribute the source code doesn't mean that they can just remove the BSD license. If they choose to distribute the code, they must include the original copyright and BSD license just as the license says.
Not really. Although Intel uses less power under load, AMD uses less power at idle. Given that most servers, and even more so desktops spend more time idling than under load, in most cases AMD ends up using less power overall.
"The disadvantage of which is that if there's a vulnerability in the base system, you probably end up updating much more than if there's a vulnerability in one of the packages in, say, Debian's base system. This is what finally turned me away from OpenBSD."
Not true, you can always download a specific patch and compile the affected binaries only.
"It's mostly GNOME and KDE that look like they are Windows clones, and those exist equally well for GNU/Linux and NetBSD."
It's not just the desktop environment that targets Windows; but the entire design philosophy and goals. Linux aims to be everything to everyone. Linus Torvalds himself states that he is content so long as things are good enough. The BSD's on the other hand strive for technical excellence in a focused area.
"On the other hand, GNU/Linux may owe a lot of its mind share to the political cause that many people associate with it."
Agreed. It's a shame that politics and not merit that drives the success of an operating system.
How did this post get modded insightful? He completely missed the point of the parent: i.e. even if one company witholds a patch, someone else can still develop the patch separately and distribute it just the same. In order for all of this to be an issue,
1. A vulnerability exists that the general public is not aware of. 2. Somebody that produces a closed source proprietary derivative product is aware of this vulnerability, does not disclose it, and patches it. 3. Yet another person is aware of this vulnerability, and exploits it.
It would be extremely improbable that all these requirements are met, especially since the person in 2. is unlikely to receive any benefit for doing so, whereas they would receive plenty of recognition for exposing said vulnerability. Besides, if such a person or entity were to exist, they could just as easily omit the patch in their source while distributing patched binaries. Not to mention, OpenBSD has a better security track record than any other open or closed source operating system anyway, so it goes to show that the license is hardly any hindrance.
An BSD licensed project is not going to be adversely affected just because a closed proprietary derivative is produced. Development goes on whether somebody decides to fork it and not give any code back, or not use it at all.
Lee Ka Shing exceeds all of these at 13.0 Billion, if you mean Asian by residence. He is in turn exceeded by Lakshmi Mittal with 25.0 Billion, if you go by citizenship.
Just because the BSD license doesn't force companies to give back, doesn't mean they can't do it anyway.
For a business that uses OpenBSD code, it would just make good business sense to support the project at a fraction of what it would cost to develop the same code in-house. It is ridiculous that Sun wouldn't even cover the travel expenses of an OpenBSD developer to go their conference, because the value of the developer's hours would have far exceeded such travel expenses. That's just simply bad business.
Uhh... which currently supported version of Windows can even reasonably run on a 386/25?
You can suspect and imagine all you want. EXPERIENCE however, would tell you otherwise.
1. But the OSes that actually USE OpenSSH DO include it, which is the point. Lack of builtin features to update installed apps is the fault of the operating system, not the individual app.
2. You seem to not recognize that OpenSSH would be updated alongside with the rest of the system on RedHat or other operating systems that include it, and the admin would not need to spend any time on OpenSSH specifically. Not to mention, any admin that needs to update a typical app on 500 desktops by hand is completely worthless anyway, and you would have a lot more security problems than what implementation of SSH you are running.
3. Auto-update mechanisms can introduce vulnerabilities as well. So, it's not a given that an auto-update mechanism is a good idea. The cons could very well outweigh the pros.
MSIE is a bug-ridden nightmare for a slew of problems, beyond just ego. Sure, ego can be a problem with open source. OpenSSH, however, was a poor example for you to make this point. Not to say that certain developers on the OpenSSH team are without egos, but the security of their software has not suffered because of it.
"Everyone thinks that they know better and don't bother listening to anyone else's opinion. It frequently comes back to bite them in the backside." You should heed your own advice.
Slashdot Mirror
That's not correct. Nobody owns all of their own fiber, no matter how big they are. Not even the Tier 1's. They may own some of it, but the majority is leased. In fact, when you purchase fiber between two locations, even though you purchase it from one party, they may in turn own some segments, while leasing various segments from multiple different providers. Even though the bulk of the costs of fiber are one time, the high capital costs result in a leasing model to cover the costs over a long period of time.
People have been arguing as you are that x86's bloated CISC instruction set was inferior to a cleaner RISC architecture for the last 20+ years. Nobody has ever proven that the elegance of the instruction set matters with hard data though.
What evidence we do have goes against that argument.
The only evidence that we have is that the benefits of commoditization and economies of scale often outweigh any architectural advantages. The fact that x86 incorporated many elements of RISC would also demonstrate its value.
Apple machines used a cleaner RISC architecture for a while in the desktop space. They never performed any better than equivalent x86 based machines, and in the end Apple abandoned RISC and moved to x86.
Manufacturing processes simply trumped architectural differences. PowerPC's have never been manufactured on anywhere near the scale of x86.
Intel came out with a cleaner RISC based instruction set that that the Itanium line uses. If x86 was really as bad as you say, Itanium chips would be running circles around the x86 based server chips provided by both Intel and AMD. That isn't happenning.
Itanium is EPIC, not CISC. It is the exact opposite of RISC. It may not be running circles around x86, but that may be due to compilers not yet being advanced enough to take full advantage of the architecture. We may still see this change in the future.
You've been able to use the linux adobe flash plugin with a native browser for quite some time now. Works just as well as it does in Linux (not necessarily saying much). The FreeBSD handbook is your friend:
http://www.freebsd.org/doc/handbook/desktop-browsers.html
Ummm... where's this fantasy version of the US you speak of?
You mean the one that voted for George W. Bush, twice?
I would beg to differ. The package management is just as good if not better than what's available in Linuxland, so there's no great difficulty in setting it up as a good desktop system.
Having excellent support for many non-x86 platforms, as well as having a small footprint make it a great choice for older hardware. I currently have it installed on on my old UltraSparc and Alpha workstations.
OpenBSD contributes more than just OpenSSH to other OS'es. Aside from pushing hardware manufacturers to open up their documentation, they've also reverse engineered drivers that have made their way into the other BSD's and even Linux (remember the whole Atheros? issue last year). Whereas many Linux distributions and the other BSD's have made compromises with proprietary drivers and binary blobs, OpenBSD still pushes for true open source.
PF and CARP also make OpenBSD a superior router platform to any IPTables based setup any day. You may be surprised how popular it is in the data centre.
Unlike Minix, OpenBSD's niche has a place in real world usage.
You mean, Hotmail used to run FreeBSD before Microsoft bought it, and for the 4+ years it took them to migrate it over to Windows without failing?
Hotmail itself has never run on Linux. It may however have some of its content delivered by Akamai's CDN, which does run Linux (but not Apache).
Cogent is a Tier 1 network service provider (weather or not Sprint and L3 want it to be).
It's apparent from this announcement that Cogent and Sprint have never had settlement free peering in place, aside from the 3 month trial. Therefore, Cogent has never been a Tier 1 network service provider, as clearly they do not have settlement free peering to all other Tier 1's.
Cogent offers great service at an unbeatable price (4-5 USD per Mbps as opposed to the 15-20 or so Sprint and competitors are charging).
They do offer service, but it's definitely not great. Nobody who takes their uptime seriously would ever single-home on Cogent.
Cogent is the type of NSP we want as a Tier 1. A very strong backer of Net Neutrality, and no intention of trying to get into the entertainment business unlike Verizon, AT&T, etc. Cogent has a goal of offering the best service at the lowest price (the end result being realistically moving the US forward in terms of available bandwidth).
That's a pretty noble goal, but they're far from meeting it. They offer the worst service at the lowest price. You're simply getting what you pay for.
As much as Sprint would like to position itself as a provider for Cogent, it's not. Sprint is a peer for Cogent with Cogent being an equivalent size of the current Sprint network, and larger than many of Sprints other peers.
Size is not the only criteria for peering. Peering is based on mutual benefit. Many larger content providers will exchange traffic with smaller eyeball networks as they do see mutual benefit from peering (i.e. not having to pay their upstreams). Clearly, Sprint does not see the benefit of peering with Cogent, and they have no obligation to peer with them.
The idea that Sprint doesn't get as much out of peering with Cogent as Cogent does peering with Sprint is absurd and PR propaganda to try and look like this was anything other than a power-play to keep a competitor at bay.
As Cogent obviously does not peer with Sprint, your assertion that Sprint gets as much out of peering with Cogent as Cogent gets from Sprint, especially in light of the great disparity in traffic exchange, is completely absurd.
Nobody who initiates a settlement-free peering agreement intends to pay for a transit connection if the trial period doesn't work out. If there was a serious chance that Cogent could be convinced to buy service from Sprint, Sprint would never have agreed to peering since the two types of connection are mutually incompatible.
From the sounds of it, Cogent was previously purchasing transit from Sprint. It's not presumptious to resume the prior relationship, after the 3 month trial did not work out.
Then too, Sprint probably decided to block Cogent's routes from the rest of their peers. Otherwise I expect Cogent would simply have routed around the downed connections. That was a step too far, one which makes Sprint a dangerous selection as your sole carrier as well.
That's not how Internet routing works. Nobody is blocking Cogent's routes. The reason why Cogent doesn't have any other routes to Sprint is because they are attempting to get all of their traffic through peering, and do not have any transit providers. In peering, you only advertise routes for your own network. As such, none of the other networks that Cogent peers with would advertise Sprint's routes to them. This is why without a direct connection to Sprint, Cogent does not have any routes to Sprint, and does not have any means of routing around the down connections.
With 10Gb circuits increasing in popularity and replacing OC48's, we're probably still in good shape, until 2012. Perhaps, by then the Cisco CRS-1 won't seem so overpriced. In the meantime, everyone can buy Juniper :)
Rather than real estate, you will usually be renting a locked cabinet with network and power drops.
Also, infrastructure is very expensive. Redundant generators, UPS'es, and network connections are all major costs, not to mention the staff you would need to support it all. It usually ends up being much cheaper just to rent a cabinet or two in a data center. There are actually much fewer data centers than you would be lead to believe by different hosting companies' website. The vast majority are renting space in somebody else's data center, which they are trying to pass off as their own.
One thing that a lot of people may not realize, is that data centers should be near other data centers or bandwidth carriers, usually in an internet building or carrier hotel. You want to have easy access to your upstreams and peers, so you're not paying a fortune to run links to them.
Now, you can just have a fiber circuit that runs to a carrier hotel, but then you're adding an additional point of failure, and uptime is everything in this business. You can mitigate that somewhat by having redundant fiber circuits that take different paths, which can often be hard to find and/or difficult/expensive to run.
ServerBeach is self-managed so it doesn't directly compete with Rackspace. Peer 1 Dedicated Hosting is where you'd go for managed hosting from the same company as ServerBeach.
* Somebody put BSD code into GPL code, and did not keep the original copyright tags. A totally isolated incident.
Whether or not it's isolated, it's clear from all the recent discussion that many people have the misconception that this is permissible under the BS license.
* Althought the GPL coder was technically wrong, there was no harm done, and the situation has been fixed.
Incorrect. Reyk's code is still being distributed by MadWifi with the GPL license wrapped around his code, when he has explicitly stated that this is against his wishes, and most likely illegal.
* The BSD community has been having a screeming bitch-fest for weeks, making all kinds of insane accusations and threats.
The BSD community is trying to correct common misconceptions by the GPL community, and also illustrate the uncooperative nature of the GPL license, and the practice of licensing modifications to BSD code under the GPL license.
* Although the BSD community has no problem at with BSD code hidden in a msft binary, they get their panties all in a wad about BSD code put into Linux.
That is because MS has complied with the BSD license whereas the GPL developer in question has not. MS also doesn't assert itself as open source, or complain about the lack of cooperation with open source efforts. As the GPL community does, and is behaving starkly in contrast with these intentions, they are guilty of hypocrisy in this regard whereas MS is not.
* Theo de Raadt is so bitchy and irratating that even his fellow BSD zealots can't stand him much of the time. And even though Theo is clearly unqualified about legal matters, much of the BSD is getting behind on this.
Theo is first and foremost a developer, and is highly respected in that regard. He is not concerned with his popularity rating or politics, and neither is most of the OpenBSD community. He is very vocal about his personal opinions, but everyone is entitled to their opinions. As his opinions are more likely to contribute than to take away from his work, they are not matters of concern for OpenBSD users.
Is that about right?
No. See above.
Being as literate as you are, you should have read by now that it was only Sam Leffler's code that was dual-licensed, and Reyk's code was NOT wherein lies the problem.
Removing the one and only license that Reyk chose for his code is a legal problem.
If strings is too obscure and not native enough for you, how about notepad.exe? Anything capable of reading text within a binary file will work.
You will be using BSD software whether you like it or not, unless you stop using the Internet.
I am now convinced that Linux zealots are *less* knowledgeable than BS zealots, or even apple zealots. I will still use Linux when it's the right tool for the job, even though the community seems bloody ignorant.
That ZDNet article is completely flawed, and should be ignored. It's rated -15 for a reason.
For one, it assumes that a RAID 5 array will become corrupt due to bad sectors. Only the files that are on those sectors will be corrupted, and since they are on multiple drives, the parity drive will allow you to recover that data.
In addition, it ignores the fact that the hard disks will simply mark those sectors as bad and will no longer use them. Not to mention, drives these days come with spare sectors that automatically replace bad ones.
Drive capacity and the probability of a RAID 5 array becoming corrupted have no correlation whatsoever.
Just because it only costs $250-$300 to manufacture doesn't mean that's all the phone costs. There are a multitude of other costs involved here, including research & development, warranty/maintenance, support, distribution, sales, marketing, yada yada yada. Find some accurate estimates on what the total real cost for final delivery is to the end user, and that it's substantially less than $500 and you might have a point.
Why does nobody understand that the code that was originally written under the BSD license never had the GPL license added to it by the author? Reyk, the author, has explicitly said this, and Theo has openly vocalized this. The reason that the code is dual-licensed is because the original code is BSD licensed and ONLY the MODIFICATIONS to it are GPL licensed. The original code itself is still under the BSD license, and not dual licensed by the original author. Also, just because companies can use BSD code in their closed-source proprietary projects, and not distribute the source code doesn't mean that they can just remove the BSD license. If they choose to distribute the code, they must include the original copyright and BSD license just as the license says.
Not really. Although Intel uses less power under load, AMD uses less power at idle. Given that most servers, and even more so desktops spend more time idling than under load, in most cases AMD ends up using less power overall.
"The disadvantage of which is that if there's a vulnerability in the base system, you probably end up updating much more than if there's a vulnerability in one of the packages in, say, Debian's base system. This is what finally turned me away from OpenBSD." Not true, you can always download a specific patch and compile the affected binaries only. "It's mostly GNOME and KDE that look like they are Windows clones, and those exist equally well for GNU/Linux and NetBSD." It's not just the desktop environment that targets Windows; but the entire design philosophy and goals. Linux aims to be everything to everyone. Linus Torvalds himself states that he is content so long as things are good enough. The BSD's on the other hand strive for technical excellence in a focused area. "On the other hand, GNU/Linux may owe a lot of its mind share to the political cause that many people associate with it." Agreed. It's a shame that politics and not merit that drives the success of an operating system.
How did this post get modded insightful? He completely missed the point of the parent: i.e. even if one company witholds a patch, someone else can still develop the patch separately and distribute it just the same. In order for all of this to be an issue,
1. A vulnerability exists that the general public is not aware of.
2. Somebody that produces a closed source proprietary derivative product is aware of this vulnerability, does not disclose it, and patches it.
3. Yet another person is aware of this vulnerability, and exploits it.
It would be extremely improbable that all these requirements are met, especially since the person in 2. is unlikely to receive any benefit for doing so, whereas they would receive plenty of recognition for exposing said vulnerability. Besides, if such a person or entity were to exist, they could just as easily omit the patch in their source while distributing patched binaries. Not to mention, OpenBSD has a better security track record than any other open or closed source operating system anyway, so it goes to show that the license is hardly any hindrance.
An BSD licensed project is not going to be adversely affected just because a closed proprietary derivative is produced. Development goes on whether somebody decides to fork it and not give any code back, or not use it at all.
Lee Ka Shing exceeds all of these at 13.0 Billion, if you mean Asian by residence. He is in turn exceeded by Lakshmi Mittal with 25.0 Billion, if you go by citizenship.
Just because the BSD license doesn't force companies to give back, doesn't mean they can't do it anyway.
For a business that uses OpenBSD code, it would just make good business sense to support the project at a fraction of what it would cost to develop the same code in-house. It is ridiculous that Sun wouldn't even cover the travel expenses of an OpenBSD developer to go their conference, because the value of the developer's hours would have far exceeded such travel expenses. That's just simply bad business.
Uhh... which currently supported version of Windows can even reasonably run on a 386/25? You can suspect and imagine all you want. EXPERIENCE however, would tell you otherwise.
1. But the OSes that actually USE OpenSSH DO include it, which is the point. Lack of builtin features to update installed apps is the fault of the operating system, not the individual app. 2. You seem to not recognize that OpenSSH would be updated alongside with the rest of the system on RedHat or other operating systems that include it, and the admin would not need to spend any time on OpenSSH specifically. Not to mention, any admin that needs to update a typical app on 500 desktops by hand is completely worthless anyway, and you would have a lot more security problems than what implementation of SSH you are running. 3. Auto-update mechanisms can introduce vulnerabilities as well. So, it's not a given that an auto-update mechanism is a good idea. The cons could very well outweigh the pros. MSIE is a bug-ridden nightmare for a slew of problems, beyond just ego. Sure, ego can be a problem with open source. OpenSSH, however, was a poor example for you to make this point. Not to say that certain developers on the OpenSSH team are without egos, but the security of their software has not suffered because of it. "Everyone thinks that they know better and don't bother listening to anyone else's opinion. It frequently comes back to bite them in the backside." You should heed your own advice.