When they say that it's "4-5x faster" but you're using all that extra speed to cover up for the crummy GPU... that's just pitiful.
Except you're not. The GMA950 is more than capable of driving the OS X GUI while only taking a minor general performance hit compared to a dedicated GPU.
The situation is different for games, but this is a machine that simply isn't meant to be used for gaming. You shouldn't *expect* it to be anything more than barely adequate for playing games.
I'm sure you can see where I'm coming from on this- it was just from my personal experience with a popular brand and I assumed it was commonplace.
All of the things you talk about are the "fault" of the PC distributor, not Microsoft. Although I've never had to deal with it personally, I, too, find the idea of "rescue" CDs that can only be used to reinstall and entire system to be frustrating and broken.
However, these issues are all separate to the topic of discussion, which is pricing of OS X vs pricing of Windows vis-a-vis "upgrade" vs "full".
Grandparent is right, you're wrong. Unix is much more secure than windows, and windows is still catching up by incorporating those same design features, period.
You can wipe the drive completely and do a clean install of OS X from any retail version. Upgrades require that you have a previous version of the operating system already on your drive. Sure, if you get a Macintosh you already have a previous version, but if ever something happens like a hard drive crash, you can just do a clean install.
Clearly I was too subtle when I said "economically speaking"...
You can only run MacOS legally on a Macintosh. Every Macintosh sold comes with a MacOS license rolled into its cost. If you own a Macintosh, you have already paid for the "full" version of MacOS analagoud to an OEM or "full" version of Windows. Every version thereafter is priced as an "upgrade", because you cannot legally run OS X without having already paid for a "full" version.
This is completely, utterly and totally a separate issue to whether or not you can install on a blank hard disk with a store-bought OS X CD or not. The installation procedure is not relevant to how the products are priced. I am not making any comment on the installation procedure, I am making comment on the GPs price comparisons.
I repeat: ECONOMICALLY SPEAKING (ie: relevant to pricing) every retail version of OS X is an "upgrade" - and as such price comparisons are only valid to "upgrade" versions of Windows.
If Apple ever release a version of OS X for non-Apple hardware, *then* it will be valid to compare it to "full" versions of Windows.
While some screamed bloody murder over Apple's apparent downgrade to Intel's integrated graphics chipset, the new MacBook completely outclasses the old iBook on all fronts [...]
Now there's a textbook example of "damning with faint praise" (and I say this as a G4 iBook owner).
The old iBook - indeed all the old G4 based machines - have been "outclassed" for years. It's not hard to make a new machine completely "outclass" an old one when the old one has been obselete for years already.
Precisely the point. Why stick a crappy GPU in with a great CPU? Nvidia and ATI make a variety of laptop chips which run circles around the GMA950. At a comparable or lower price.
The GMA950 is free, it comes with the system chipset. It's hard to get any cheaper than that. A dedicated video chipset would probably add significant (relative) cost to the manufacturing of the i^HMacBook, as it would require additional system board real estate, cooling allowances and the chips (GPU and RAM) themselves.
With that said, I was very disappointed there wasn't a "high end" MacBook with dedicated video to serve as a 12" PB replacement. My initial thoughts were that the black MacBook would fill this niche and that's why it was US$150 more (and a different colour), but closer examination revealed it was just a cynical money-grab.
Intel's usual quality and performance standards go out the window when it comes to their graphics chips.
No, they're simply not *meant* to be anything more than a very basic video card. You're saying because a Honda Civic can't outrun a Nissan Skyline, Honda's "usual quality and performance standards go out the window".
As others have pointed out, the Core Duo only beat out the Athlon64 FX-60 when overclocked. If the chip, when overclocked, was safe for production environments, then the chip would have shipped at a higher default clock speed.
It's not exactly unknown for Intel to have chips reliably capable of higher speeds, but not immediately release them in order to maximise profits.
So, Viiv is not a CPU name, it's like a standards designation. (I think that all started with Centrino -- it wasn't just the CPU, it was the CPU and wireless and such stuff.)
Readers of sufficient vintage may like to consider it an up-to-date version of the "Multimedia PC" (MPC) from 1990 (or thereabouts).
For you youngun's, MPC dictated a few basic features that a PC had to have to be considered capable of "multimedia". From memory, it required an Adlib compatible soundcard, a 1x CDROM and a VGA-capable video card.
When the vast majority of malware requires some sort of user interaction to install, you most certainly *can* blame the user.
Apple, Sun and Linux don't have Windoze problems.
Hardly surprising, considering *collectively* they probably have less than a tenth of the marketshare.
Or, to put it another way, "Apple, Sun and Linux" would have to have ca. ten times as many problems as Windows to appear equally as bad.
In the Apple case, the user is less expected to fiddle and may know less about their computer than any other kind of user. Viruses and worms for any system other than Windoze never last long outside a lab.
Considering that all other platforms have tiny marketshares, dramatically limiting the ability of malware to spread, combined with small, relatively tightly-knit communities, within which news spreads fast and, in the case of "Sun and Linux", typical end users with significantly above-average technical knowledge, that's hardly surprising in the least.
It's not the user's fault that Windows botnets form the backbone of every computer crime network, it's Microsoft's.
Again, as long as the majority of malware requires end user interaction to install, it remains the users' faults.
In terms of internet infrastructure, Unix *is* more popular than Windows.
There are a *lot* of machines out there that aren't "internet infrastructure". Indeed, machines that are "internet infrastructure" are a very small minority.
On the grand scheme of things, servers and the like, Unix & Unix like operating systems > The combined set of Windows operating systems, marketshare-wise.
Are you seriously trying to suggest there are more servers on the internet than clients ?
Or are you just suggesting that there are more unix servers on the 'net than Windows servers ?
Because the former doesn't even pass the laugh test and I fail to see how the latter is relevant. Viruses and other generic malware typically neither target, nor compromise on a wide scale, server machines (Windows *or* unix).
Unix has been around much longer than Windows. Most of the internet's big iron runs on Unix or Unix-like systems. Many of the internet's juiciest targets (largest companies) serve on Unix or Unix-like systems, including Google.
They're generally also run by highly skilled professionals who not only take pro-active steps to prevent compromises from ocurring, but are able to detect and remedy them extremely quickly. Which is why they're not really the targets of people writing viruses, malware and other automated (or semi-automated) exploits.
Desktop Marketshare = Viruses is a popular myth. It belongs where the sun don't shine.
Marketshare is an integral factor of a platforms exposure to malicious code. Both directly and indirectly. This is inescapable fact.
Wherever did you get the idea that commercial software packages couldn't reuse the source code?
Probably his strong grip on reality, something people who insist it's possible to build a business around selling GPLed software generally lack.
I can't think of any widely-used free software licenses that prohibit commercial use.
Of course, in reality, it's rather hard to make money selling software when your first customer can turn around an undercut you on price, or distribute your product to the entire world for free.
Certainly the GNU GPL doesn't, as you can see quite plainly from the number of commercial distributions, many of them (such as Red Hat) selling Linux-based operating systems for the same price as Microsoft Windows.
Red Hat isn't selling Linux, it's selling support contracts.
Xen I can understand being problematic, as it requires a "special" kernel that might not be compatible with your production environment, but I don't really see why you can't use VMWare. A laptop with a dual-core CPU and 4G of RAM should be more than sufficient to run a development environment (if not several) in VMs.
If Microsoft management wants to generate some goodwill, then the management should open-source old versions of Microsoft Windows that are no longer being sold. In this way, people who have the older computers could easily get a copy of the older versions of Windows.
Your argument is broken. All Microsoft would have to do to "generate some goodwill" is to make old versions of Windows freely downloadable. There's not even the slightest justification there for open-sourcing them.
Ultimately though, OSX performance is a success story because on a G3 700mhz with 256M of ram its actually useable. Have you tried running Windows XP on a similar setup? Tried turning all of the eye candy on? Bet you didn't like the way it performed either.
Windows is noticably more responsive on my 550Mhz P3 laptop than my 1Ghz G4 iBook (at least until it starts paging). While I prefer my iBook on the road because it has much better battery life, I prefer to use the old Compaq when there's a powerpoint nearby (unless I'm doing something which is going to be memory and/or CPU intensive, since the iBook has twice the RAM (768 vs 384) and more than twice the processing power).
However, the real problem is that OS X remains sluggish even on much faster machines (I find my mum's G5 iMac too slow to use with anything more than a couple of apps running) and really needs a monster machine (dual G5+) to get even close to "fast".
Windows is *much* nicer to older machines than OS X is and also scales up in performance much better.
OS X *is* "horribly slow". Every benchmark shows this and a few days of serious interactive use will also demonstrate it. There are a lot of nice things about OS X, but performance - especially interactive responsiveness - and multitasking are not high on the list. I also used to think it was simply because of the poor old G4 processor (particularly the memory bandwidth), but when the G5s came out and OS X was *still* chunky to use, that theory was shot.
OS X is a microkernel like Windows NT is a microkernel. Which is to say, its architecture is modular and there's lots of getting passed around, but everything runs in kernel mode for decent performance (although that doesn't seem to help OS X much).
I believe Vista is moving more stuff out of kernel mode, moving it back towards the original versions of NT. So in theory it's getting a bit more microkernel-ish with the next release.
I take your point that IT departments ought to be split between user support and infrastructure support. However, I take exception with the idea that any company with only a couple IT staff is "too small to be of consequence". With the vast majority of companies being too small to be of consequence, doesn't that make them consequential?
If you have *two* IT staff, then you have sufficient for one to be doing basic user support and another to be working along more strategic lines (and, IME, this is the kind of breakdown a typical small IT department has - the "IT Manager" doing the "planning" and the PFYs doing the "rebooting").
In a single-man IT shop, then obviously it isn't going to work, but such small shops are inconsequential.
And I should clarify here that by "inconsequential" I mean in the context of the discussion, not the economy. Such small businesses, pretty much inherently, have very little compartmentalisation and specialisation. Everyone does everything and everyone bothers anyone whenever they need something done.
What I'm getting at is that an overworked IT staffer in MicroBiz is no more replaceable than one in Megacorp500.
I would have to disagree. In all likelihood, the IT staff in MicroBiz has substantial localised knowledge - often the only repository of it - that probably couldn't even be duplicated again, let alone documented efficiently. The IT staffer in MegaCorp is far more likely to be repalceable drone. This is not to say people whose experience is integral to the functioning of the company don't exist in big business, it's just that there's far fewer of them, and general policies and procedures in place to avoid such situations in the first place. Business continuity is critical when your income is being measured in the hundreds of millions range. It's usually an afterthought (as in, "after the bus hit him, we really thought we were boned") when you're in the hundreds of thousands range.
If you treat them badly or overload them with work, they will quit (or grin and bear it), and losing 50-100% of your entire IT staff is much worse than losing 5-10% of the staff.
Unfortunately, IT staff in larger corporations are not integral to their workplaces, even most who think they are. Particularly in big corporations, replacing large chunks of your IT department is (comparitively) not a huge problem. This is one of the reasons staff turnover and promotion-by-new-job tends to be very prevalent in the industry.
The vast majority of IT staff are - realistically - about as valuable and difficult to replace as a secretary. They have cookie-cutter skills that can (and often are, these days) learnt in multi-month "boot camps". They're not like Engineers, Accountants or Lawyers who take *years* to be educated to the point of even basic usefulness, and as long again to really become productive and valuable to the company.
What it requires is some way to minimize the impact of a minimal IT department, I think. Easily-configured networks, plug and play servers, and automatic failure detection and prevention software are all necessary. To some extent these exist, but not to the extent that IT staff can be replaced by them wholesale.
Much as it's going to cost people jobs, I have to agree. A *massive* proportion of the typical IT workers job is made up of ridiculously trivial troubleshooting that most of them can do in their sleep. These sort of trivial tasks simply shouldn't require people to do - and one day they won't.
OTOH, high-level support, developers, system and network admins and the like are going to be in demand for a lot longer yet. It's the low-end desktop and network support - the 'rebooters' - who are going to suffer.
That said, it also means that if something is working well enough and the users are satisfied with the performance of the backbone, then any upgrades or new system implementations are PURELY egotistical masturbation.
Only if you assume the company is completely and utterly static, with no plans for this to change.
Most companies I know of (and have worked for) are interested in expansion and growth - at the very least in profits and productivity, if not size and marketshare. I've never seen any that have no ambitions to improve themselves at all, and if I were employed by such a company, I would certainly be trying to find work elsewhere.
That "egotistical masturbation", as you call it, is there so that when some PHB walks in without any prior warning and says "we've just bought a company that will increase our employee count by 30% and our workload by 50%, you've got $RIDICULOUSLY_SHORT_TIMEFRAME to integrate them with us", you don't spend $RIDICULOUSLY_SHORT_TIMEFRAME working 80-hour weeks with no overtime pay.
What that means, in concrete terms, is that your IT plan which intends to migrate everyone over from the Windows 2003 Active Server server to the Debian Sarge LAMP server that you host in your mom's basement must take a backseat to user requests to reboot their computer.
If the guys working on future planning are the same guys who have to reboot end user workstations, then your company probably has bigger problems to worry about, or is too small to be of consequence.
It boils down to the fact that IT is a loss for the company. It is a net loser which produces nothing that makes money.
While this is true in a strictly economical sense, if a company's IT department is not continuously working on ways to use technology to improve efficiency, then your IT department is not doing the job it should be. That's what all that "mental masturbation" is supposed to be doing - coming up with better ways to use technology to lower costs and/or increase productivity.
Once the system is stabilized, then user requests must again take priority over the IT plan.
That is entirely dependent on the "user requests". Although, again, in any remotely sanely run IT department, these two tasks should never conflict.
Oil companies are refusing to do anything to lower pump prices because it's not in the best interest of the shareholders, despite it being in the best interest of society as a whole.
No, it's not. Low pump prices encourage excessive usage (which has a whole bunch of negative direct and indirect flow-on effects, and very few positive ones) and discourage R&D into, and production of, alternative fuels.
(Incidentally, if you're American, you have no concept of what the words "high fuel prices" mean.)
Personally, I have no issue with petrol prices here in Australia (AU$1.35ish/L), and quite frankly wouldn't have a huge problem with them going up more (I wouldn't expect it would would my lifestyle re: vehicle usage until it hit at least $2/L). Every time I think about some North-shore soccer mum paying AU$120-odd to fill up her X5 she uses to drive the few kilometres between home, school and the supermarket, it makes me smile.
I see no evidence in the article that ISPs are going to be looking at packet payloads (let alone entire sessions). Indeed, the technical requirements of doing so, alone, make the very suggestion ridiculous.
The ISPs attempting to leech profits intend to do so based on content as well as origin/destination. Once they make decisions based on content they should be liable for all the content.
Traffic is being prioritised based on source and destination IPs and, possibly, ports. That's not filtering based on content. There is no way you can meaningfully derive the content of any data transmission based on IPs and port numbers. All you can do is - very generally - determine the type of traffic it is, which is precisely what these ISPs want to do, so they can squeeze more money out of customers using more of one type of data than the other.
Again, the original analogy is broken. It is like trying to say that if toll roads base their toll charges on the source, destination and type of vehicle (which many of them do), they are liable for the goods being transported in those vehicles (which they most certainly are not).
I am no fan of discriminatory traffic shaping, however, there are some technical justifications and a hell of a lot of business justifications for doing so. Outside of irrational idealism, it's hard to see any arguments against it.
With that said, I think ISPs are vastly underestimating which service - theirs or Google's - consumers consider to be more important.
The fact is that Microsoft software is insecure, overpriced, bloated crap.
Compared to what ?
Oh yes, you can get your work done with it, and in many cases it is easier because it's all made by the same guys and they know all the secret tricks.
Which "secret tricks" are you referring to ? Is this another rehash on the good old "secret APIs" myth ?
Microsoft software is demonstrably less secure than, well, pretty much everything else.
Demonstratable how ? Please don't tell me you think comparing how frequently security problems occur between products with such disparate marketshare actually has any meaning ?
It is also highly unstable. I crash windows software (including the OS) all the damned time, and on good hardware.
Then something about your computer configuration is broken, and you should get it fixed. However, I'm guessing you don't have the slightest interest in getting it fixed, because then you wouldn't have anything to reinforce your stereotyping.
If you run windows, you are not in control of your computer. Microsoft is. Short of attacking the system with a debugger, you really have no way of knowing what Microsoft is doing with your data. For all you know, the indexing service caches all your passwords and sneaks them into your crash reports as a snippet of binary data.
I could say exactly the same thing about all the OSS we use. For all *I* know, SSH is sending my passwords to some random mailing list everytime I type one in.
Linux is better than Windows in most departments already and it's getting better faster.
Maybe in the fantasies of a Slashdot poster, but out in the real world, real people don't think so. AD, SMS, MOM, Exchange, Office, Terminal Services, application integration, collaboration. These are cornerstones of Enterprise IT infrastructure and there's very little of it "Linux" even manages to do as well, let alone better.
Slashdot Mirror
Except you're not. The GMA950 is more than capable of driving the OS X GUI while only taking a minor general performance hit compared to a dedicated GPU.
The situation is different for games, but this is a machine that simply isn't meant to be used for gaming. You shouldn't *expect* it to be anything more than barely adequate for playing games.
Anti-spyware (and antivirus) software isn't protecting from defects in the code, it's protecting from defects in the user.
All of the things you talk about are the "fault" of the PC distributor, not Microsoft. Although I've never had to deal with it personally, I, too, find the idea of "rescue" CDs that can only be used to reinstall and entire system to be frustrating and broken.
However, these issues are all separate to the topic of discussion, which is pricing of OS X vs pricing of Windows vis-a-vis "upgrade" vs "full".
For example ?
Clearly I was too subtle when I said "economically speaking"...
You can only run MacOS legally on a Macintosh. Every Macintosh sold comes with a MacOS license rolled into its cost. If you own a Macintosh, you have already paid for the "full" version of MacOS analagoud to an OEM or "full" version of Windows. Every version thereafter is priced as an "upgrade", because you cannot legally run OS X without having already paid for a "full" version.
This is completely, utterly and totally a separate issue to whether or not you can install on a blank hard disk with a store-bought OS X CD or not. The installation procedure is not relevant to how the products are priced. I am not making any comment on the installation procedure, I am making comment on the GPs price comparisons.
I repeat: ECONOMICALLY SPEAKING (ie: relevant to pricing) every retail version of OS X is an "upgrade" - and as such price comparisons are only valid to "upgrade" versions of Windows.
If Apple ever release a version of OS X for non-Apple hardware, *then* it will be valid to compare it to "full" versions of Windows.
Now there's a textbook example of "damning with faint praise" (and I say this as a G4 iBook owner).
The old iBook - indeed all the old G4 based machines - have been "outclassed" for years. It's not hard to make a new machine completely "outclass" an old one when the old one has been obselete for years already.
The GMA950 is free, it comes with the system chipset. It's hard to get any cheaper than that. A dedicated video chipset would probably add significant (relative) cost to the manufacturing of the i^HMacBook, as it would require additional system board real estate, cooling allowances and the chips (GPU and RAM) themselves.
With that said, I was very disappointed there wasn't a "high end" MacBook with dedicated video to serve as a 12" PB replacement. My initial thoughts were that the black MacBook would fill this niche and that's why it was US$150 more (and a different colour), but closer examination revealed it was just a cynical money-grab.
Intel's usual quality and performance standards go out the window when it comes to their graphics chips.
No, they're simply not *meant* to be anything more than a very basic video card. You're saying because a Honda Civic can't outrun a Nissan Skyline, Honda's "usual quality and performance standards go out the window".
Economically speaking, every retail version of OS X is an upgrade. For a valid comparison, you need to compare upgrade pricing for Windows.
Clearly, you're not a businessman.
It's not exactly unknown for Intel to have chips reliably capable of higher speeds, but not immediately release them in order to maximise profits.
Readers of sufficient vintage may like to consider it an up-to-date version of the "Multimedia PC" (MPC) from 1990 (or thereabouts).
For you youngun's, MPC dictated a few basic features that a PC had to have to be considered capable of "multimedia". From memory, it required an Adlib compatible soundcard, a 1x CDROM and a VGA-capable video card.
When the vast majority of malware requires some sort of user interaction to install, you most certainly *can* blame the user.
Apple, Sun and Linux don't have Windoze problems.
Hardly surprising, considering *collectively* they probably have less than a tenth of the marketshare.
Or, to put it another way, "Apple, Sun and Linux" would have to have ca. ten times as many problems as Windows to appear equally as bad.
In the Apple case, the user is less expected to fiddle and may know less about their computer than any other kind of user. Viruses and worms for any system other than Windoze never last long outside a lab.
Considering that all other platforms have tiny marketshares, dramatically limiting the ability of malware to spread, combined with small, relatively tightly-knit communities, within which news spreads fast and, in the case of "Sun and Linux", typical end users with significantly above-average technical knowledge, that's hardly surprising in the least.
It's not the user's fault that Windows botnets form the backbone of every computer crime network, it's Microsoft's.
Again, as long as the majority of malware requires end user interaction to install, it remains the users' faults.
There are a *lot* of machines out there that aren't "internet infrastructure". Indeed, machines that are "internet infrastructure" are a very small minority.
On the grand scheme of things, servers and the like, Unix & Unix like operating systems > The combined set of Windows operating systems, marketshare-wise.
Are you seriously trying to suggest there are more servers on the internet than clients ?
Or are you just suggesting that there are more unix servers on the 'net than Windows servers ?
Because the former doesn't even pass the laugh test and I fail to see how the latter is relevant. Viruses and other generic malware typically neither target, nor compromise on a wide scale, server machines (Windows *or* unix).
Unix has been around much longer than Windows. Most of the internet's big iron runs on Unix or Unix-like systems. Many of the internet's juiciest targets (largest companies) serve on Unix or Unix-like systems, including Google.
They're generally also run by highly skilled professionals who not only take pro-active steps to prevent compromises from ocurring, but are able to detect and remedy them extremely quickly. Which is why they're not really the targets of people writing viruses, malware and other automated (or semi-automated) exploits.
Desktop Marketshare = Viruses is a popular myth. It belongs where the sun don't shine.
Marketshare is an integral factor of a platforms exposure to malicious code. Both directly and indirectly. This is inescapable fact.
Do you drive a modern car ? Travel in large aircraft ? Use banks ? Own a gaming console ?
Ever take the elevator to the 50th floor ? How about make calls on a mobile phone ?
Probably his strong grip on reality, something people who insist it's possible to build a business around selling GPLed software generally lack.
I can't think of any widely-used free software licenses that prohibit commercial use.
Of course, in reality, it's rather hard to make money selling software when your first customer can turn around an undercut you on price, or distribute your product to the entire world for free.
Certainly the GNU GPL doesn't, as you can see quite plainly from the number of commercial distributions, many of them (such as Red Hat) selling Linux-based operating systems for the same price as Microsoft Windows.
Red Hat isn't selling Linux, it's selling support contracts.
Xen I can understand being problematic, as it requires a "special" kernel that might not be compatible with your production environment, but I don't really see why you can't use VMWare. A laptop with a dual-core CPU and 4G of RAM should be more than sufficient to run a development environment (if not several) in VMs.
Your argument is broken. All Microsoft would have to do to "generate some goodwill" is to make old versions of Windows freely downloadable. There's not even the slightest justification there for open-sourcing them.
Windows is noticably more responsive on my 550Mhz P3 laptop than my 1Ghz G4 iBook (at least until it starts paging). While I prefer my iBook on the road because it has much better battery life, I prefer to use the old Compaq when there's a powerpoint nearby (unless I'm doing something which is going to be memory and/or CPU intensive, since the iBook has twice the RAM (768 vs 384) and more than twice the processing power).
However, the real problem is that OS X remains sluggish even on much faster machines (I find my mum's G5 iMac too slow to use with anything more than a couple of apps running) and really needs a monster machine (dual G5+) to get even close to "fast".
Windows is *much* nicer to older machines than OS X is and also scales up in performance much better.
OS X *is* "horribly slow". Every benchmark shows this and a few days of serious interactive use will also demonstrate it. There are a lot of nice things about OS X, but performance - especially interactive responsiveness - and multitasking are not high on the list. I also used to think it was simply because of the poor old G4 processor (particularly the memory bandwidth), but when the G5s came out and OS X was *still* chunky to use, that theory was shot.
OS X is a microkernel like Windows NT is a microkernel. Which is to say, its architecture is modular and there's lots of getting passed around, but everything runs in kernel mode for decent performance (although that doesn't seem to help OS X much).
I believe Vista is moving more stuff out of kernel mode, moving it back towards the original versions of NT. So in theory it's getting a bit more microkernel-ish with the next release.
If you have *two* IT staff, then you have sufficient for one to be doing basic user support and another to be working along more strategic lines (and, IME, this is the kind of breakdown a typical small IT department has - the "IT Manager" doing the "planning" and the PFYs doing the "rebooting").
In a single-man IT shop, then obviously it isn't going to work, but such small shops are inconsequential.
And I should clarify here that by "inconsequential" I mean in the context of the discussion, not the economy. Such small businesses, pretty much inherently, have very little compartmentalisation and specialisation. Everyone does everything and everyone bothers anyone whenever they need something done.
What I'm getting at is that an overworked IT staffer in MicroBiz is no more replaceable than one in Megacorp500.
I would have to disagree. In all likelihood, the IT staff in MicroBiz has substantial localised knowledge - often the only repository of it - that probably couldn't even be duplicated again, let alone documented efficiently. The IT staffer in MegaCorp is far more likely to be repalceable drone. This is not to say people whose experience is integral to the functioning of the company don't exist in big business, it's just that there's far fewer of them, and general policies and procedures in place to avoid such situations in the first place. Business continuity is critical when your income is being measured in the hundreds of millions range. It's usually an afterthought (as in, "after the bus hit him, we really thought we were boned") when you're in the hundreds of thousands range.
If you treat them badly or overload them with work, they will quit (or grin and bear it), and losing 50-100% of your entire IT staff is much worse than losing 5-10% of the staff.
Unfortunately, IT staff in larger corporations are not integral to their workplaces, even most who think they are. Particularly in big corporations, replacing large chunks of your IT department is (comparitively) not a huge problem. This is one of the reasons staff turnover and promotion-by-new-job tends to be very prevalent in the industry.
The vast majority of IT staff are - realistically - about as valuable and difficult to replace as a secretary. They have cookie-cutter skills that can (and often are, these days) learnt in multi-month "boot camps". They're not like Engineers, Accountants or Lawyers who take *years* to be educated to the point of even basic usefulness, and as long again to really become productive and valuable to the company.
What it requires is some way to minimize the impact of a minimal IT department, I think. Easily-configured networks, plug and play servers, and automatic failure detection and prevention software are all necessary. To some extent these exist, but not to the extent that IT staff can be replaced by them wholesale.
Much as it's going to cost people jobs, I have to agree. A *massive* proportion of the typical IT workers job is made up of ridiculously trivial troubleshooting that most of them can do in their sleep. These sort of trivial tasks simply shouldn't require people to do - and one day they won't.
OTOH, high-level support, developers, system and network admins and the like are going to be in demand for a lot longer yet. It's the low-end desktop and network support - the 'rebooters' - who are going to suffer.
Only if you assume the company is completely and utterly static, with no plans for this to change.
Most companies I know of (and have worked for) are interested in expansion and growth - at the very least in profits and productivity, if not size and marketshare. I've never seen any that have no ambitions to improve themselves at all, and if I were employed by such a company, I would certainly be trying to find work elsewhere.
That "egotistical masturbation", as you call it, is there so that when some PHB walks in without any prior warning and says "we've just bought a company that will increase our employee count by 30% and our workload by 50%, you've got $RIDICULOUSLY_SHORT_TIMEFRAME to integrate them with us", you don't spend $RIDICULOUSLY_SHORT_TIMEFRAME working 80-hour weeks with no overtime pay.
What that means, in concrete terms, is that your IT plan which intends to migrate everyone over from the Windows 2003 Active Server server to the Debian Sarge LAMP server that you host in your mom's basement must take a backseat to user requests to reboot their computer.
If the guys working on future planning are the same guys who have to reboot end user workstations, then your company probably has bigger problems to worry about, or is too small to be of consequence.
It boils down to the fact that IT is a loss for the company. It is a net loser which produces nothing that makes money.
While this is true in a strictly economical sense, if a company's IT department is not continuously working on ways to use technology to improve efficiency, then your IT department is not doing the job it should be. That's what all that "mental masturbation" is supposed to be doing - coming up with better ways to use technology to lower costs and/or increase productivity.
Once the system is stabilized, then user requests must again take priority over the IT plan.
That is entirely dependent on the "user requests". Although, again, in any remotely sanely run IT department, these two tasks should never conflict.
No, it's not. Low pump prices encourage excessive usage (which has a whole bunch of negative direct and indirect flow-on effects, and very few positive ones) and discourage R&D into, and production of, alternative fuels.
(Incidentally, if you're American, you have no concept of what the words "high fuel prices" mean.)
Personally, I have no issue with petrol prices here in Australia (AU$1.35ish/L), and quite frankly wouldn't have a huge problem with them going up more (I wouldn't expect it would would my lifestyle re: vehicle usage until it hit at least $2/L). Every time I think about some North-shore soccer mum paying AU$120-odd to fill up her X5 she uses to drive the few kilometres between home, school and the supermarket, it makes me smile.
Indeed. Heaven forbid someone be able to make money from selling software.
I see no evidence in the article that ISPs are going to be looking at packet payloads (let alone entire sessions). Indeed, the technical requirements of doing so, alone, make the very suggestion ridiculous.
The ISPs attempting to leech profits intend to do so based on content as well as origin/destination. Once they make decisions based on content they should be liable for all the content.
Traffic is being prioritised based on source and destination IPs and, possibly, ports. That's not filtering based on content. There is no way you can meaningfully derive the content of any data transmission based on IPs and port numbers. All you can do is - very generally - determine the type of traffic it is, which is precisely what these ISPs want to do, so they can squeeze more money out of customers using more of one type of data than the other.
Again, the original analogy is broken. It is like trying to say that if toll roads base their toll charges on the source, destination and type of vehicle (which many of them do), they are liable for the goods being transported in those vehicles (which they most certainly are not).
I am no fan of discriminatory traffic shaping, however, there are some technical justifications and a hell of a lot of business justifications for doing so. Outside of irrational idealism, it's hard to see any arguments against it.
With that said, I think ISPs are vastly underestimating which service - theirs or Google's - consumers consider to be more important.
Compared to what ?
Oh yes, you can get your work done with it, and in many cases it is easier because it's all made by the same guys and they know all the secret tricks.
Which "secret tricks" are you referring to ? Is this another rehash on the good old "secret APIs" myth ?
Microsoft software is demonstrably less secure than, well, pretty much everything else.
Demonstratable how ? Please don't tell me you think comparing how frequently security problems occur between products with such disparate marketshare actually has any meaning ?
It is also highly unstable. I crash windows software (including the OS) all the damned time, and on good hardware.
Then something about your computer configuration is broken, and you should get it fixed. However, I'm guessing you don't have the slightest interest in getting it fixed, because then you wouldn't have anything to reinforce your stereotyping.
If you run windows, you are not in control of your computer. Microsoft is. Short of attacking the system with a debugger, you really have no way of knowing what Microsoft is doing with your data. For all you know, the indexing service caches all your passwords and sneaks them into your crash reports as a snippet of binary data.
I could say exactly the same thing about all the OSS we use. For all *I* know, SSH is sending my passwords to some random mailing list everytime I type one in.
Linux is better than Windows in most departments already and it's getting better faster.
Maybe in the fantasies of a Slashdot poster, but out in the real world, real people don't think so. AD, SMS, MOM, Exchange, Office, Terminal Services, application integration, collaboration. These are cornerstones of Enterprise IT infrastructure and there's very little of it "Linux" even manages to do as well, let alone better.