When you're talking about MACHINE and OS downtimes, then you MUST be talking about MACHINE and OS uptimes/availability, NOT application Availability.
I don't know how much more clearly I can make it.
I've had several nines of uptime on Windows Server without a budget that accommodates purchasing failovers for everything, but because updates and server downtime are scheduled to a time of my choosing, I can "claim", according to him, that it's got better uptime than it really has?
Of course not. Linux server installations, really, truly does have better uptime than most Windows Server(tm) installations. And if he hadn't made completely bullshit claims to back it up, I'd not have complained.
But then he brings up all this entirely OS agnostic bullshit that Windows and Linux can do perfectly well and claims it in the Linux Kernel's favor. I can virtualize too. I can migrate VMs too, if I had the budget (*shakes fist*) for more servers. I'm even well aware Linux can do what I'm asking for less upfront cost and likely, over several years, a lower TCO.
Those are legitimate bragging points for Linux, all.
But claiming X nines of application uptime after factoring in things like virtualization, failovers, clusters, caches and other completely OS agnostic capabilities that there are a good half dozen vendors for as if it were an inherent Linux "feature" that you can't get elsewhere... That's just BS.
IANAL, but anything plainly viewable from public property is not considered private.
On the other hand, if Google had developed a portable camera that can see through walls, blinds, hedges and clothes, and started driving that around public property, I think the locals might have a leg to stand on. For that matter, I'd like to see how close they'd get to Langley, or Fort Meade for example.
Interestingly, if everyone has access to said technology, it's no longer exotic or invasion of privacy. There's a novel by Arthur C. Clarke and Stephen Baxter, "The Light of Other Days," in which technology to remotely view any location on earth becomes widespread, convenient, and eventually, integral to modern life. When anyone can watch you anywhere, no matter what you're doing, does privacy even make sense?
Food for thought, as well as a rebuttal to you begging the question (that it is illegal to take pictures from a public road of private property.)
You're drastically changing the definition of downtime in order to support your initial premise, making this into an argument of semantics instead of anything else. Of course for arbitrary definitions of downtime, I can claim anything and everything I run has no downtime. I do IT myself and have never had any "downtime" according to your definition, but there certainly has been "my time" taken away because of having to spend "my time" after hours in order to prevent "downtime" according to your definition.
Of course, I would be a huge liar to say, "I run Windows Server 2003 R2 with zero downtime for months at a time." That's of course a load of BS that no one would believe, and I don't understand why you feel privileged to be able to say the same thing about a Linux distro of choice but you consider it entirely valid.
So in the future, for the sake of intelligent discussion, clarify what you mean by downtime before getting into a debate and then explaining, oh no, you didn't mean THAT downtime, of course you didn't mean the widespread understanding of what downtime is, you meant YOUR definition of downtime.
A VM is merely a way of abstracting the hardware layer, but it doesn't mean you have 100% uptime. The VM, in order to be upgraded, has to experience downtime. And even if you have fast VM switching between two hosts (a rather fickle technology) that only gives you the ability to upgrade an arbitrary host, but does not let you upgrade the VM without taking it down.
So, no, you still miss the point or don't understand what VMs do if you think that a virtual machine means you never have to worry about kernel security patches.
So your copout is, ironically, to say that updates aren't needed. Rather than mention that there actually are tools that let you arbitrarily swap kernels while rebooting. (This is NEVER recommended for production systems, of course.)
Would it be cool if Windows -could- do the same thing? Yeah, sure. But I'm guessing without putting up 30+ warning prompts, most admins wouldn't heed the security and stability issues of swapping a running kernel and guess what? Microsoft would be blamed for having an instable OS.
What a troll, you clearly didn't spend much time with Vista.
Look to the left for file data? Now you look to the bottom. Or you use the details mode. There are so many ways of finding that information. Hundreds of MBs? DWM used around 30MB I found. Windows 7 uses around 23MB for the window manager. And while Vista on a low resource system was, admittedly, pretty bad. On a high resource system, it more intelligently used extra CPU cycles, extra RAM, and most computer's largely idle built in graphics to do useful work.
And the administrator tools! My god, you can't begin to compare the tools in XP and Vista. Every administrator tool got better, command line tools appeared for even the most esoteric functions. Command line differencing or full system image backups using a snapshotting filesystem the user has control over (wbadmin and vssadmin respectively, allow control of those features.) Task scheduler gained a truly ridiculous number of capabilities. I have my computer wake me up to music, and at the same time, the computer wakes itself up. Out of sleep mode. Computer management, disk management, the firewall, the performance monitor, the reliability monitor, the security policies,... every administrative aspect became vastly better.
I just, I can't keep listening to people troll Vista like it's an administrative nightmare. If anything, it gives IT people vastly more control over their OS than they've had with XP. Ever since I've installed Vista I've had free, automated system backups, reliable sleep and wake, a snappier display because I use discrete graphics cards, I've had better control over what I can do with regards to guest accounts (SteadyState + UAC + fine grained ACLs means so, so much.)
And lastly, the DRM. I've never encountered it. Not once. Music from iTunes, AFAIK, doesn't use it. They have their own draconian DRM that they implement. None of my ripped music has ever been abridged by Vista or 7. None of my videos have ever been paused or downgraded in resolution, and that includes HD quality video. The DRM is solely for those who choose to implement it in software, to provide them with that guarantee if they choose to. I don't agree with that choice, but I'd rather the OS let me watch the damn thing (and let me exploit the analog hole) than have to watch it on a piece of hardware I have no control over (most Blu-Ray players and movies with HDCP enabled.)
Frankly, what Vista screwed up on were a few operations (file copies, how did they F that up?) and public relations. When that troll of an article on DRM came out, they should have responded with an open letter the next day. Instead, they've never addressed those concerns. They needed to be on top of things, and they essentially ignored widespread public criticism. It's not hard to understand why Vista would be so disliked if they never attempted to alleviate or address concerns.
I hope this changes with 7. Certainly, the Engineering Windows 7 blog has done that so far. We'll see if that continues post release.
LGA-1366 is promised to continue at least on to the 6-core, 32nm "Gulftown." At the very least, Gulftown will be at least another 50% improvement for highly threaded applications.
And that'll work fine in Vista, because that's not Program Files. Like in Linux, there are a few privileged, protected folders, and pretty much everything else, including mountable r/w media like flash drives, is essentially an extension of one's home directory. (Unless you do something fancy with the permissions, of course.)
I disagree, purchasing an i7 now as opposed to a Core 2 Quad makes a great deal more sense as long as you're the sort of person who upgrades their computers themselves. If you don't ever swap a CPU, if you don't care about memory latency, or a whole mess of other things, then, really it doesn't matter what you buy. Buy a Phenom II rig, they're certainly Good Enough(tm) for whatever you want to do.
On the other hand, if you want to be able to upgrade your computer and you do want that extra performance over the competition, buying i7 now is the only thing that makes sense. The LGA-1366 is the new "workstation" class Intel socket. There will be a lower pin count variant for desktops, and there will be a higher pin count variant for servers and multiple-processor systems. So if you need higher performance now, and you want to be able to upgrade, you either get an AM3 board (Phenom II) or you get an LGA-1366 board (i7.)
IBM's contributions to the open source community are much, much greater than any actual donation of developer time and money.
They've defended Linux and the GPL and sought to protect the Linux kernel from frivolous patent claims by using what is surely the largest warchest of patents ever accumulated. (Honestly, they're worse than the US or Russia during the cold war, as far as acquiring weapons goes.) It speaks a lot that they're willing to go up to bat for an ideal and a bunch of nameless, faceless developers who have all individually contributed a bit of code here or there, but lack the ability to defend that code as patent unencumbered in a court room.
When he says the conclusion will be wrong, he is correct. You would turn out to be "flat dead wrong" according to all the much better models we have for longer term climate change. His model would find every year a period of global cooling and global warming. A better model would involve the temperatures worldwide and their far-reaching effects.
So even if this hypothetical model turned up the correct result, the model is not accurate, because the basis of that reasoning is inherently flawed. The conclusion may be correct, but even a broken clock is, well, you know.
Intel's new Core i7 920 can be overclocked to 3.2GHz with the stock cooler (and without voiding the warranty,) and go much higher on marginally more expensive coolers. 3.8GHz easily on a $40 heatsink and fan, 4.0 GHz if you opt for the extremely high end heatsinks, 4.2 GHz I've seen on water running stable. And on more exotic coolers, you can hit 4.5-5.0 GHz.
I imagine that is similar to his situation. The Core i7 costs $290 right now or so, and can easily beat the older Core 2 based $1550 processor you found in most benchmarks. Especially anything that hits the memory, as my i7 can hit 18GB/s (big B) to RAM. That's two and a half times what Intel's dual-QX9775 "Skulltrail" platform can do.
My entire machine cost lest than that QX9775, by the way.
In case you missed the recent news, you can really impress her if you say during a prominent trial, "No, I don't need any compensation. But you could send my wife some flowers." Or mix it up, say chocolate, for example.
With open source libraries, you generally have to find the wheel before you can reuse it.
Often people end up reinventing the wheel because they (a.) couldn't find one someone else made, (b.) found one, but it wasn't under licensing terms that they could use with their project, or (c.) found one, but the project lost its way and ended up incomplete with a lead developer who may well have been hit by a bus.
Not saying closed source libraries are more helpful, plentiful, or accessible, but open source is not the panacea that zealots on Slashdot would like it to be.
I make a competing calculator (hypothetically). I want an icon on the desktop for the Windows Calculator, Maxima, Octave and Mathematica.
I also (again, hypothetically) make a notepad replacement. I want my product, Notepad++, Wordpad, Microsoft Word, and a half dozen scintilla-based knockoffs.
I also hypothetically make an alternative desktop shell. Because Microsoft FORCES you to use theirs, before you even get to see all of the five BILLION other fucking icons, I want a screen to pop up with only a mouse, and a choice of shells. Mine, which doesn't support UAC, separation of privileges, explorer shells (which will confuse the heck out of people,) explorer extensions (bye-bye TortoiseSVN, TortoiseHG, etc,) or other features. Also included should be shells that barely work.
And finally, after booting into Windows becomes a clusterfuck of choosing about eighteen trillion defaults, I as a developer expect my users to have a relatively stable and ubiquitous set of APIs available.
Oh wait, we threw that out the window.
Fuck.
Here's an idea. Let Microsoft keep doing what they're doing and easily choose between default programs, and even allow those programs to prompt the user to alter their default. Because any other option is fraught with favoritism and is just going to cram OEM desktops with more bullshit than ever before, and make the idea of targeting the Windows desktop from a developer or support perspective laughable.
There are so many ways to disable automatic updating, it's not even funny. Some disable all Windows Update functionality, some disable automatic updates, some disable automatic downloads.
Slashdot Mirror
When you're talking about MACHINE and OS downtimes, then you MUST be talking about MACHINE and OS uptimes/availability, NOT application Availability.
I don't know how much more clearly I can make it.
I've had several nines of uptime on Windows Server without a budget that accommodates purchasing failovers for everything, but because updates and server downtime are scheduled to a time of my choosing, I can "claim", according to him, that it's got better uptime than it really has?
Of course not. Linux server installations, really, truly does have better uptime than most Windows Server(tm) installations. And if he hadn't made completely bullshit claims to back it up, I'd not have complained.
But then he brings up all this entirely OS agnostic bullshit that Windows and Linux can do perfectly well and claims it in the Linux Kernel's favor. I can virtualize too. I can migrate VMs too, if I had the budget (*shakes fist*) for more servers. I'm even well aware Linux can do what I'm asking for less upfront cost and likely, over several years, a lower TCO.
Those are legitimate bragging points for Linux, all.
But claiming X nines of application uptime after factoring in things like virtualization, failovers, clusters, caches and other completely OS agnostic capabilities that there are a good half dozen vendors for as if it were an inherent Linux "feature" that you can't get elsewhere... That's just BS.
IANAL, but anything plainly viewable from public property is not considered private.
On the other hand, if Google had developed a portable camera that can see through walls, blinds, hedges and clothes, and started driving that around public property, I think the locals might have a leg to stand on. For that matter, I'd like to see how close they'd get to Langley, or Fort Meade for example.
Interestingly, if everyone has access to said technology, it's no longer exotic or invasion of privacy. There's a novel by Arthur C. Clarke and Stephen Baxter, "The Light of Other Days," in which technology to remotely view any location on earth becomes widespread, convenient, and eventually, integral to modern life. When anyone can watch you anywhere, no matter what you're doing, does privacy even make sense?
Food for thought, as well as a rebuttal to you begging the question (that it is illegal to take pictures from a public road of private property.)
You're drastically changing the definition of downtime in order to support your initial premise, making this into an argument of semantics instead of anything else. Of course for arbitrary definitions of downtime, I can claim anything and everything I run has no downtime. I do IT myself and have never had any "downtime" according to your definition, but there certainly has been "my time" taken away because of having to spend "my time" after hours in order to prevent "downtime" according to your definition.
Of course, I would be a huge liar to say, "I run Windows Server 2003 R2 with zero downtime for months at a time." That's of course a load of BS that no one would believe, and I don't understand why you feel privileged to be able to say the same thing about a Linux distro of choice but you consider it entirely valid.
So in the future, for the sake of intelligent discussion, clarify what you mean by downtime before getting into a debate and then explaining, oh no, you didn't mean THAT downtime, of course you didn't mean the widespread understanding of what downtime is, you meant YOUR definition of downtime.
A VM is merely a way of abstracting the hardware layer, but it doesn't mean you have 100% uptime. The VM, in order to be upgraded, has to experience downtime. And even if you have fast VM switching between two hosts (a rather fickle technology) that only gives you the ability to upgrade an arbitrary host, but does not let you upgrade the VM without taking it down.
So, no, you still miss the point or don't understand what VMs do if you think that a virtual machine means you never have to worry about kernel security patches.
So your copout is, ironically, to say that updates aren't needed. Rather than mention that there actually are tools that let you arbitrarily swap kernels while rebooting. (This is NEVER recommended for production systems, of course.)
http://www.randombugs.com/linux/patching-linux-kernel-rebooting.html
About a minute of google searching gave me that.
Would it be cool if Windows -could- do the same thing? Yeah, sure. But I'm guessing without putting up 30+ warning prompts, most admins wouldn't heed the security and stability issues of swapping a running kernel and guess what? Microsoft would be blamed for having an instable OS.
Even ksplice doesn't always work, anyhow.
See, digital signatures WORK! Proof that they'll solve all of our software update and malware pro...
Wait, conficker was written by the bad guys?
Dammit.
What a troll, you clearly didn't spend much time with Vista.
Look to the left for file data? Now you look to the bottom. Or you use the details mode. There are so many ways of finding that information. Hundreds of MBs? DWM used around 30MB I found. Windows 7 uses around 23MB for the window manager. And while Vista on a low resource system was, admittedly, pretty bad. On a high resource system, it more intelligently used extra CPU cycles, extra RAM, and most computer's largely idle built in graphics to do useful work.
And the administrator tools! My god, you can't begin to compare the tools in XP and Vista. Every administrator tool got better, command line tools appeared for even the most esoteric functions. Command line differencing or full system image backups using a snapshotting filesystem the user has control over (wbadmin and vssadmin respectively, allow control of those features.) Task scheduler gained a truly ridiculous number of capabilities. I have my computer wake me up to music, and at the same time, the computer wakes itself up. Out of sleep mode. Computer management, disk management, the firewall, the performance monitor, the reliability monitor, the security policies, ... every administrative aspect became vastly better.
I just, I can't keep listening to people troll Vista like it's an administrative nightmare. If anything, it gives IT people vastly more control over their OS than they've had with XP. Ever since I've installed Vista I've had free, automated system backups, reliable sleep and wake, a snappier display because I use discrete graphics cards, I've had better control over what I can do with regards to guest accounts (SteadyState + UAC + fine grained ACLs means so, so much.)
And lastly, the DRM. I've never encountered it. Not once. Music from iTunes, AFAIK, doesn't use it. They have their own draconian DRM that they implement. None of my ripped music has ever been abridged by Vista or 7. None of my videos have ever been paused or downgraded in resolution, and that includes HD quality video. The DRM is solely for those who choose to implement it in software, to provide them with that guarantee if they choose to. I don't agree with that choice, but I'd rather the OS let me watch the damn thing (and let me exploit the analog hole) than have to watch it on a piece of hardware I have no control over (most Blu-Ray players and movies with HDCP enabled.)
Frankly, what Vista screwed up on were a few operations (file copies, how did they F that up?) and public relations. When that troll of an article on DRM came out, they should have responded with an open letter the next day. Instead, they've never addressed those concerns. They needed to be on top of things, and they essentially ignored widespread public criticism. It's not hard to understand why Vista would be so disliked if they never attempted to alleviate or address concerns.
I hope this changes with 7. Certainly, the Engineering Windows 7 blog has done that so far. We'll see if that continues post release.
LGA-1366 is promised to continue at least on to the 6-core, 32nm "Gulftown." At the very least, Gulftown will be at least another 50% improvement for highly threaded applications.
And that'll work fine in Vista, because that's not Program Files. Like in Linux, there are a few privileged, protected folders, and pretty much everything else, including mountable r/w media like flash drives, is essentially an extension of one's home directory. (Unless you do something fancy with the permissions, of course.)
I disagree, purchasing an i7 now as opposed to a Core 2 Quad makes a great deal more sense as long as you're the sort of person who upgrades their computers themselves. If you don't ever swap a CPU, if you don't care about memory latency, or a whole mess of other things, then, really it doesn't matter what you buy. Buy a Phenom II rig, they're certainly Good Enough(tm) for whatever you want to do.
On the other hand, if you want to be able to upgrade your computer and you do want that extra performance over the competition, buying i7 now is the only thing that makes sense. The LGA-1366 is the new "workstation" class Intel socket. There will be a lower pin count variant for desktops, and there will be a higher pin count variant for servers and multiple-processor systems. So if you need higher performance now, and you want to be able to upgrade, you either get an AM3 board (Phenom II) or you get an LGA-1366 board (i7.)
IBM's contributions to the open source community are much, much greater than any actual donation of developer time and money.
They've defended Linux and the GPL and sought to protect the Linux kernel from frivolous patent claims by using what is surely the largest warchest of patents ever accumulated. (Honestly, they're worse than the US or Russia during the cold war, as far as acquiring weapons goes.) It speaks a lot that they're willing to go up to bat for an ideal and a bunch of nameless, faceless developers who have all individually contributed a bit of code here or there, but lack the ability to defend that code as patent unencumbered in a court room.
Vista and Win 7 does this too by default. Called Hybrid Sleep.
Still about 15 seconds after installing Office and other applications.
That i7 sips down more juice than my Lenovo X60 laptop (screen and all,) so what I do is I just sleep my computer and then carry it UPS and all.
Kinda sucks that the battery weighs 20 pounds and only lasts five minutes.
When he says the conclusion will be wrong, he is correct. You would turn out to be "flat dead wrong" according to all the much better models we have for longer term climate change. His model would find every year a period of global cooling and global warming. A better model would involve the temperatures worldwide and their far-reaching effects.
So even if this hypothetical model turned up the correct result, the model is not accurate, because the basis of that reasoning is inherently flawed. The conclusion may be correct, but even a broken clock is, well, you know.
20%? You need to more closely examine the benchmarks.
Intel's new Core i7 920 can be overclocked to 3.2GHz with the stock cooler (and without voiding the warranty,) and go much higher on marginally more expensive coolers. 3.8GHz easily on a $40 heatsink and fan, 4.0 GHz if you opt for the extremely high end heatsinks, 4.2 GHz I've seen on water running stable. And on more exotic coolers, you can hit 4.5-5.0 GHz.
I imagine that is similar to his situation. The Core i7 costs $290 right now or so, and can easily beat the older Core 2 based $1550 processor you found in most benchmarks. Especially anything that hits the memory, as my i7 can hit 18GB/s (big B) to RAM. That's two and a half times what Intel's dual-QX9775 "Skulltrail" platform can do.
My entire machine cost lest than that QX9775, by the way.
In case you missed the recent news, you can really impress her if you say during a prominent trial, "No, I don't need any compensation. But you could send my wife some flowers." Or mix it up, say chocolate, for example.
With open source libraries, you generally have to find the wheel before you can reuse it.
Often people end up reinventing the wheel because they (a.) couldn't find one someone else made, (b.) found one, but it wasn't under licensing terms that they could use with their project, or (c.) found one, but the project lost its way and ended up incomplete with a lead developer who may well have been hit by a bus.
Not saying closed source libraries are more helpful, plentiful, or accessible, but open source is not the panacea that zealots on Slashdot would like it to be.
Notepad++ is a -good- scintilla knockoff, and it'd be an insult to its author(s) to lump it in with that lot.
I make a competing calculator (hypothetically). I want an icon on the desktop for the Windows Calculator, Maxima, Octave and Mathematica.
I also (again, hypothetically) make a notepad replacement. I want my product, Notepad++, Wordpad, Microsoft Word, and a half dozen scintilla-based knockoffs.
I also hypothetically make an alternative desktop shell. Because Microsoft FORCES you to use theirs, before you even get to see all of the five BILLION other fucking icons, I want a screen to pop up with only a mouse, and a choice of shells. Mine, which doesn't support UAC, separation of privileges, explorer shells (which will confuse the heck out of people,) explorer extensions (bye-bye TortoiseSVN, TortoiseHG, etc,) or other features. Also included should be shells that barely work.
And finally, after booting into Windows becomes a clusterfuck of choosing about eighteen trillion defaults, I as a developer expect my users to have a relatively stable and ubiquitous set of APIs available.
Oh wait, we threw that out the window.
Fuck.
Here's an idea. Let Microsoft keep doing what they're doing and easily choose between default programs, and even allow those programs to prompt the user to alter their default. Because any other option is fraught with favoritism and is just going to cram OEM desktops with more bullshit than ever before, and make the idea of targeting the Windows desktop from a developer or support perspective laughable.
So Microsoft should be forced to sell XP indefinitely and provide support for it indefinitely?
GLaDOS
There are so many ways to disable automatic updating, it's not even funny. Some disable all Windows Update functionality, some disable automatic updates, some disable automatic downloads.
Obama was elected by converter boxes living in watershed lands that were passed over by the census?