"Make that twice as expensive to run Vista inside another OS. Vista Ultimate costs about $400 compared to Premium which is about $250."
As another poster mentioned, that's not true. But even if it were, it's irrelevant. Costing more money to allow virtualization is better than not allowing it at all, particularly when steps are taken to ensure that no one can legally provide it (virtualizing MacOS requires breaking the copy protection and thus the DMCA).
"You'll see OS X Server with virtualisation and unlimited installs of the OS on that VM in the very near future, I'm sure."
No doubt; just like we're going to see an upgradeable mini-tower RSN.
They don't allow it at all, and have announced no plans to ever allow it. As far as I'm concerned, the criticism stands until it's actually been announced.
"So, you have to buy a Mac, but you get practically unlimited installs. Not so with Microsoft."
This is why I specifically say "with respect to virtualization" in my post. We can go round and round in circles over whether or not getting any hardware you like beats unlimited installs on a very narrow range of hardware, but the question of virtualization is very simple; Apple doesn't allow it.
"Apple will let you run OS X on any computer it's licensed for, regardless of what other OS's may also be running on the computer. As long as you can run OS X on that computer, they don't give a shit what you do with it.
Microsoft, on the other hand, says you only have Vista rights if Vista is the primary OS at that time. Or you can pay them much more money to play fairly, despite the fact that you purchased a copy of Vista licensed to run on this particular computer. Microsoft is restricting your ability to use the software you purchased to run on that computer, and only let you do so if they're the software in charge. This is typical Microsoft behavior and has been since day one."
It's quite disingenuous to claim that Apple is being more reasonable with respect to virtualization.
Microsoft: We want more money to let you run Vista under virtualization.
Apple: You may never, under any circumstances, on any hardware, at any time, for any reason, ever run OS X under virtualization. Period.
Microsoft's terms suck, there's no doubt about that. Apple's are worse.
Cellphones output many times the power of a wifi network (since wifi is in an unregulated band the power is limited) and you hold the transmitter right up to your ear. If the link between cancer and cellphones is tenuous, how are we to believe that wifi is terrible?
The definition of "planet" is now based on a body's relationship to its surroundings, and as long as we're clear on that I don't see the problem. The same thing goes for "moon", "island", etc. Without water around it, it's not an island, it's a hill or a mountain.
We can have other terms to describe bodies that don't relate to their surroundings, but it's a nice, concise way to describe a body as a part of a solar system.
"A kernel port of the ZFS code could never go into the Linux kernel due to the license issues between the CDDL and GPL, where as this is perfectly fine in userspace."
That's why it would be cool if Sun released Solaris under the GPL.
If it were released under the GPL, it could be imported into the Linux kernel.
I think this is awesome.
I can't decide whether Sun has balls of spent Uranium or if they're just really disparate. Possibly both. But I really like this, and I hope their services and hardware businesses benefit accordingly.
The flash 9 linux plugin is in beta, and apparently it works fine in a Linux binary browser on FreeBSD with Linux binary compatibility. They didn't mention OpenBSD, but OpenBSD has a fraction of the desktop users.
"Out of curiosity, how are you implementing failover with different external IP addresses? Or is this for outbound connections only, such as internet enabling an office?"
You are correct that failover would be impossible with different external IP addresses. However, the only thing outside the cluster that notices any difference is the switch, which sees the MAC address start appearing on another physical port. CARP (common address redundancy protocol) creates what is basically a shared virtual NIC, with a shared MAC address and a shared IP address(es).
The real MAC address can optionally have an IP address associated with it, but you do need some way of reaching backup nodes for management. It is common to give every a unique address in addition to the shared one on your internal network. For example, internal nodes might see.1 as the gateway, and this would be the shared address, with your firewall nodes using.2 and.3 for their unique addresses. Also, pfsync (the protocol that distributes state information) generally needs its own physical network (you could probably get away without this but it's not a good idea), and the other nodes are also generally reachable through there.
Turns out a specialized OS for a small number of users often ends up being something that can't be easily replaced. PF has availability features no one outside of Cisco can match, and they can't match them for what it costs us to use OpenBSD for the job.
For example, our Internet connection at work is managed by OpenBSD. If I rebooted our firewall, no one would notice, because the backup would kick in and it would preserve state for everything, even pre-existing TCP connections. You could be streaming music and it wouldn't even skip. How can I do that with Linux again?
"I can't run any of the stuff I need to run under OpenBSD, so why the heck should I even care about it?"
Hm. Whenever I have that problem, I just download the Linux version and run it under binary emulation.
That happened to me once on OpenBSD with pf.conf (firewall ruleset). I remembered that it sends you a diff of changes to important files for security reasons, and figured it must have a copy of the old one somewhere or it couldn't do the diff. Turns out it's in/var/backups.
You're right. But, at the moment, something like "my network port stops working sometimes" is only going to make extra work for people doing triage on the bug tracker. You need to know what chipset it is, and what conditions trigger the bug, etc.
And, as I said, if the bug has already manifested itself, that's sufficient for criticism.
Well as it happens, it's not a colo box, it's a workstation, and I took this opportunity to install a larger hard drive, so rolling back would have been a matter of plugging in the other one. It's not a big deal in this case.
And you're right, I'd want to use Dapper or Debian-stable on a colo, and I'd definitely want a management card with KVM/CD redirection capabilities even then.
I'm just pointing out that there are some instances where it would matter.
Like many people, you are completely missing the point.
If something doesn't work, it has already failed to work, and it is legitimate to make that the subject of complaints. Yes, they need bugs submitted to fix things, but that is a separate issue.
Submitting a good bug report is a fair amount of work: you need to check for dupes, lay out the conditions necessary to reproduce it, give details like your hardware, etc. This is a lot more work than the work necessary to, say Google something, or read a man page, yet we already know that this is too much to expect from most users. If Ubuntu wants things to be easy enough for people that don't know how to google something, then they cannot reasonably expect everyone to submit bug reports.
"Bleeding edge vs. Leading edge The Core 2 Duo have been out for months a full month for Laptops. There was time for the processors to be checked out and proven. If it was bleading edge when Intel release the Chip the next day Apple will be shipping with them. (Like dell)."
I wasn't talking about C2D, I was talking about Core Duos, which Apple shipped in January, right after they were launched. Apple didn't give them much time to vet themselves.
Therefore, I think this is just a rationalization, not the real reason. It's false too much for it to be the real motivation.
"So Apple doesn't feet your needs, then don't get it. If you really need something that customized then get a PC. PCs have always been more customizable then Apples anyways. I myself personally have gone threw a complete customizable phase, where I just build a PC to do what I want. But with the direction of my professional work, as well my personal interests I am no longer as concerned about making all the choices. I am willing to put a little blind trust to a company (that has earned my trust) that they will give me a well balanced system."
That's funny 'cause Apple has lost my trust. They had it at one time, but I wasn't able to keep that up given the needs I have that they don't even attempt to me, and the total lack of balance that's evident in so many of their systems.
For example, I recently got a workstation. Yet, I only needed one dual-core CPU, so getting a dual-Xeon system would be a huge waste of money. Where are Apple's well-balanced systems in this case? I see only extremes in two directions. I don't want a maxed out Dell, but I don't want a gimped pseudo-laptop like an iMac or a massively overkilled machine like a Mac Pro either. I want well balanced, and Apple doesn't have it.
"A maxed out Dell Laptop Costs over $6500 that is about 33% faster then a $3000 Apple Laptop. So is it worth it? It is up to you."
If that's what I was talking about, it wouldn't be worth it. Since that's not what I'm talking about, I clearly have to explain further.
I don't want a completely maxed out machine. I want one that has money allocated to things I care about, and away from things I don't care about. This is much more difficult with Apple. The Dell I'd get might cost the same, but have a better screen instead of a higher clock speed on the CPU that I don't need but Apple won't give me a choice.
"I code a lot too. To get the extra text in, I do a Crazy thing like making the font smaller. Or I hook up to an external Display. and I get my Screen with an additional high res screen, OS X makes this easy an convenient for its users. It is not saying a higher res display wouldn't be nicer but loosing an inch in 2 directions in pixels isn't much a loss. Especially the way apple makes workaround for these limitation much easier and convient to manage then Windows or LInux."
When coding, the first thing I do is make the font smaller, but with Apple's DPI, the pixels are too big and the letters become illegible sooner. With a higher resolution screen, I can take this further, and fit more onto a laptop screen when I can't have dual displays like I do at home.
Slashdot Mirror
"Make that twice as expensive to run Vista inside another OS. Vista Ultimate costs about $400 compared to Premium which is about $250."
As another poster mentioned, that's not true. But even if it were, it's irrelevant. Costing more money to allow virtualization is better than not allowing it at all, particularly when steps are taken to ensure that no one can legally provide it (virtualizing MacOS requires breaking the copy protection and thus the DMCA).
"You'll see OS X Server with virtualisation and unlimited installs of the OS on that VM in the very near future, I'm sure."
No doubt; just like we're going to see an upgradeable mini-tower RSN.
They don't allow it at all, and have announced no plans to ever allow it. As far as I'm concerned, the criticism stands until it's actually been announced.
"So, you have to buy a Mac, but you get practically unlimited installs. Not so with Microsoft."
This is why I specifically say "with respect to virtualization" in my post. We can go round and round in circles over whether or not getting any hardware you like beats unlimited installs on a very narrow range of hardware, but the question of virtualization is very simple; Apple doesn't allow it.
"Apple will let you run OS X on any computer it's licensed for, regardless of what other OS's may also be running on the computer. As long as you can run OS X on that computer, they don't give a shit what you do with it.
Microsoft, on the other hand, says you only have Vista rights if Vista is the primary OS at that time. Or you can pay them much more money to play fairly, despite the fact that you purchased a copy of Vista licensed to run on this particular computer. Microsoft is restricting your ability to use the software you purchased to run on that computer, and only let you do so if they're the software in charge. This is typical Microsoft behavior and has been since day one."
It's quite disingenuous to claim that Apple is being more reasonable with respect to virtualization.
Microsoft: We want more money to let you run Vista under virtualization.
Apple: You may never, under any circumstances, on any hardware, at any time, for any reason, ever run OS X under virtualization. Period.
Microsoft's terms suck, there's no doubt about that. Apple's are worse.
Unfortunately, an SSH connection is much more likely to be allowed out than VPN traffic.
"OpenBSD"
The DoD's involvement with OpenBSD did end in tears.
Cellphones output many times the power of a wifi network (since wifi is in an unregulated band the power is limited) and you hold the transmitter right up to your ear. If the link between cancer and cellphones is tenuous, how are we to believe that wifi is terrible?
The definition of "planet" is now based on a body's relationship to its surroundings, and as long as we're clear on that I don't see the problem. The same thing goes for "moon", "island", etc. Without water around it, it's not an island, it's a hill or a mountain.
We can have other terms to describe bodies that don't relate to their surroundings, but it's a nice, concise way to describe a body as a part of a solar system.
You won't sink. The gravity's too weak, remember?
"A kernel port of the ZFS code could never go into the Linux kernel due to the license issues between the CDDL and GPL, where as this is perfectly fine in userspace."
That's why it would be cool if Sun released Solaris under the GPL.
Tension in the cable would slow down the rotation of the Earth slightly.
If it were released under the GPL, it could be imported into the Linux kernel.
I think this is awesome.
I can't decide whether Sun has balls of spent Uranium or if they're just really disparate. Possibly both. But I really like this, and I hope their services and hardware businesses benefit accordingly.
Light's not that bad. What's bad is air-conditioning during the day.
FreeBSD's routing has seen some major updates in the last few years, they're faster than pretty much everyone (even Linux).
At work, our P4 OpenBSD firewalls running OpenBGPD can handle the 100 megabit connection.
I dunno, I think it would be a waste on anything I/O bound.
The flash 9 linux plugin is in beta, and apparently it works fine in a Linux binary browser on FreeBSD with Linux binary compatibility. They didn't mention OpenBSD, but OpenBSD has a fraction of the desktop users.
"Out of curiosity, how are you implementing failover with different external IP addresses? Or is this for outbound connections only, such as internet enabling an office?"
.1 as the gateway, and this would be the shared address, with your firewall nodes using .2 and .3 for their unique addresses. Also, pfsync (the protocol that distributes state information) generally needs its own physical network (you could probably get away without this but it's not a good idea), and the other nodes are also generally reachable through there.
You are correct that failover would be impossible with different external IP addresses. However, the only thing outside the cluster that notices any difference is the switch, which sees the MAC address start appearing on another physical port. CARP (common address redundancy protocol) creates what is basically a shared virtual NIC, with a shared MAC address and a shared IP address(es).
The real MAC address can optionally have an IP address associated with it, but you do need some way of reaching backup nodes for management. It is common to give every a unique address in addition to the shared one on your internal network. For example, internal nodes might see
Turns out a specialized OS for a small number of users often ends up being something that can't be easily replaced. PF has availability features no one outside of Cisco can match, and they can't match them for what it costs us to use OpenBSD for the job.
For example, our Internet connection at work is managed by OpenBSD. If I rebooted our firewall, no one would notice, because the backup would kick in and it would preserve state for everything, even pre-existing TCP connections. You could be streaming music and it wouldn't even skip. How can I do that with Linux again?
"I can't run any of the stuff I need to run under OpenBSD, so why the heck should I even care about it?"
Hm. Whenever I have that problem, I just download the Linux version and run it under binary emulation.
I thought bash was installed by default on Panther onwards?
hehehehe
/var/backups.
That happened to me once on OpenBSD with pf.conf (firewall ruleset). I remembered that it sends you a diff of changes to important files for security reasons, and figured it must have a copy of the old one somewhere or it couldn't do the diff. Turns out it's in
damn
+1 Hero
How did I not know that?
Cheers.
You're right. But, at the moment, something like "my network port stops working sometimes" is only going to make extra work for people doing triage on the bug tracker. You need to know what chipset it is, and what conditions trigger the bug, etc.
And, as I said, if the bug has already manifested itself, that's sufficient for criticism.
heh
Well as it happens, it's not a colo box, it's a workstation, and I took this opportunity to install a larger hard drive, so rolling back would have been a matter of plugging in the other one. It's not a big deal in this case.
And you're right, I'd want to use Dapper or Debian-stable on a colo, and I'd definitely want a management card with KVM/CD redirection capabilities even then.
I'm just pointing out that there are some instances where it would matter.
Like many people, you are completely missing the point.
If something doesn't work, it has already failed to work, and it is legitimate to make that the subject of complaints. Yes, they need bugs submitted to fix things, but that is a separate issue.
Submitting a good bug report is a fair amount of work: you need to check for dupes, lay out the conditions necessary to reproduce it, give details like your hardware, etc. This is a lot more work than the work necessary to, say Google something, or read a man page, yet we already know that this is too much to expect from most users. If Ubuntu wants things to be easy enough for people that don't know how to google something, then they cannot reasonably expect everyone to submit bug reports.
I dunno, I think I might be willing to part with the camera that is banned on several customer's sites.
Also, I don't care if it's 1" thick, particularly if it's going to cause problems with cooling. That is something I can do without.
"Bleeding edge vs. Leading edge The Core 2 Duo have been out for months a full month for Laptops. There was time for the processors to be checked out and proven. If it was bleading edge when Intel release the Chip the next day Apple will be shipping with them. (Like dell)."
I wasn't talking about C2D, I was talking about Core Duos, which Apple shipped in January, right after they were launched. Apple didn't give them much time to vet themselves.
Therefore, I think this is just a rationalization, not the real reason. It's false too much for it to be the real motivation.
"So Apple doesn't feet your needs, then don't get it. If you really need something that customized then get a PC. PCs have always been more customizable then Apples anyways. I myself personally have gone threw a complete customizable phase, where I just build a PC to do what I want. But with the direction of my professional work, as well my personal interests I am no longer as concerned about making all the choices. I am willing to put a little blind trust to a company (that has earned my trust) that they will give me a well balanced system."
That's funny 'cause Apple has lost my trust. They had it at one time, but I wasn't able to keep that up given the needs I have that they don't even attempt to me, and the total lack of balance that's evident in so many of their systems.
For example, I recently got a workstation. Yet, I only needed one dual-core CPU, so getting a dual-Xeon system would be a huge waste of money. Where are Apple's well-balanced systems in this case? I see only extremes in two directions. I don't want a maxed out Dell, but I don't want a gimped pseudo-laptop like an iMac or a massively overkilled machine like a Mac Pro either. I want well balanced, and Apple doesn't have it.
"A maxed out Dell Laptop Costs over $6500 that is about 33% faster then a $3000 Apple Laptop. So is it worth it? It is up to you."
If that's what I was talking about, it wouldn't be worth it. Since that's not what I'm talking about, I clearly have to explain further.
I don't want a completely maxed out machine. I want one that has money allocated to things I care about, and away from things I don't care about. This is much more difficult with Apple. The Dell I'd get might cost the same, but have a better screen instead of a higher clock speed on the CPU that I don't need but Apple won't give me a choice.
"I code a lot too. To get the extra text in, I do a Crazy thing like making the font smaller. Or I hook up to an external Display. and I get my Screen with an additional high res screen, OS X makes this easy an convenient for its users. It is not saying a higher res display wouldn't be nicer but loosing an inch in 2 directions in pixels isn't much a loss. Especially the way apple makes workaround for these limitation much easier and convient to manage then Windows or LInux."
When coding, the first thing I do is make the font smaller, but with Apple's DPI, the pixels are too big and the letters become illegible sooner. With a higher resolution screen, I can take this further, and fit more onto a laptop screen when I can't have dual displays like I do at home.