Agree ME was a consumer version of windows - hence it was an "accident" that I omitted it.
2000 was not. I doubt you'd have found it pre-installed on almost any PC from a major OEM. I will agree that it was much more useful as a consumer OS than previous NT incarnations.
My post was not a comment on sane desktop management or what the best desktop OS was at any point in time. It was merely a list of the branding issues MS had for just their consumer OS version - the one that comes pre-installed on 95% of all the PCs sold anywhere. New marketing team, new numbering/lettering/dating/whatever scheme.
Have to agree. I work in a fortune 500 company and stuff like email/fileshares/network has outages on at least this kind of frequency, and I imagine it costs way more for a company to do it alone than to outsource it to Google.
I think the Google outages just get way more publicity. If some Exchange outage results in 10% of my company not getting emails for an hour, there is a good chance the other 90% won't hear about it at all, and many of the 10% won't notice either (away in meetings, and of course the client-server design of Outlook is a bit more failure-tolerant). Heck - a few years ago a location with 10k employees lost power for almost an entire day due to some kind of utilities mess-up.
Outages on large enterprise systems happen a lot more often than people realize.
If you want to find a consumer version of Windows that doesn't get support updates you'll have to go back to Windows ME. Even XP STILL gets updates (though not for much longer).
Nobody is suggesting Android phones should get security updates forever. However, there is a big difference between 10 years from the last sale and dropping support while the phones are still in the stores.
It would be nice if they just defined a standard stable driver ABI, like most proprietary OSes, or required the drivers to be open source, like Linux. You really need one or the other.
Imagine if MS release a special version of Windows for every model of PC out there, and they never updated any of them. No business would want to touch that. Instead MS defines a standard driver ABI and they've only changed it a few times in the last two decades (3.1->9X/NT, 9x->XP, then the 64 bit switch around Vista), and then they leave the drivers up to the device vendors and provide OS updates for a decade after they stop selling it. The result is that if you bought a printer in 2001 on XP, you can still use the printer on XP with security updates today, and you could probably at least go to a more current 32-bit Windows version without losing your printer and buy another 5-10 years if you had to (and there is a decent chance you'll find a 64-bit driver in which case you can just go to the current version and get about 15 years out of it).
Sure, Windows support isn't perfect, but we're comparing months to years here - Android comes nowhere near to even RedHat/Canonical support timelines, let alone what MS/Apple deliver.
With android there is exactly one phone available at any time that will get updates for 1.5 years from date of FIRST sale, and it isn't available on every carrier.
If you want a Nexus phone TODAY on Verizon, for example, your only choice is a Galaxy Nexus, and I doubt you'll get an update after this year for that phone (it is possible, but who can say).
The whole strength of Android is supposed to be the incredible diversity. All that diversity goes away if you start out by tossing all but one model of phone at any given time. Even the Nexus phones get fairly minimal support - the Nexus One only got updates for only a few months after it was last sold as far as I'm aware (not sure if there were any post-GB security updates, but it did not get ICS 11 months after it was discontinued). I would consider strong support to be all security updates for at least two years after the last day sold. Microsoft delivers security updates for 10 years after an OS is last sold, but nobody else really comes close to that.
There are many contracts you aren't given the option to sign, like signing yourself into slavery, or signing away your rights if you're injured due to neglect (oh, every ski slope will ask you to sign one of those, but the only effect they really have is to discourage lawsuits - they have little legal power).
The major carriers have similar terms and conditions on their contracts and are near-monopolies.
All of this points to a need tor regulation. Nobody is saying carriers can't sell brand A vs B phones. They're just saying that they should receive security updates only for the duration of these contracts (which would mean two years from the LAST one sold).
The problem is that CM doesn't support phones for that long either. Take the G2 - the last official CM release for that was Gingerbread, and Cyanogen himself owns one of those. The Nexus S, which has very similar hardware, is on Jelly Bean.
CM support is hit-and-miss, and it is hard to predict whether it will exist or for how long. While the CM team is doing great work, it really isn't a substitute for vendor-supported security patches.
That seems like a reasonable minimum. The sad thing is that most flagship Android phones are updated for about six months after they FIRST go on sale, no two years after they LAST go on sale. Even the Nexus phones are only updated for 1.5 years after they FIRST go on sale.
The Nexus One was sold until it was replaced by the S in November 2010. It did not receive ICS a year later. It did get GB a month later, but I'm not sure if it got any security updates after that.
The previous Google-branded phone was the ADP. It was sold until Jan 2010 (I believe - certainly no other Google-branded phone was available before then), and it was already out of date having missed Eclair three months before.
The only way you even get 1.5 years of updates on a Nexus-branded phone is to buy it on launch date. If you bought a Nexus 4 today I'd be skeptical that you'd get much more than a year of updates.
Believe it or not at my company's mail room they have an X-ray machine and a glovebox. They don't use them on every package, but if they get one that looks suspicious they will use them. From what I understand this is pretty common post-9/11. I doubt they have any way to test anything they find in the packages, but if the x-ray looked like a bomb or the letter in the glove box contained powder they'd likely abandon the building and call the police.
A trader, who is allowed to execute trades, and has all the machinery to do so at his disposal, sends out orders to buy one billion dollars of stock and relies on the settling period, which I believe is three days (may be five).
Yes, we know what he was trying to do. The question is how the firm let it happen in the first place.
If I go into the expense system at my employer and punch in that I spent one billion dollars out-of-pocket on inkjet toner the system will not just mail me a check for a billion dollars. Such a transaction (even if legitimate) would go all the way up to the CEO for approval, along with checks for compliance with purchasing rules/etc. The system is likely capable of printing any figure on the check the company wants, but it is designed with safeguards so that ordinary employees can't go spending arbitrary amounts of money. If I said I bought $50 worth of printer toner then most likely it would just mail me a check and if I didn't turn in the receipts they'd deal with it after-the-fact.
The problem with trading firms is that they spend other-people's-money. That means that they only care about having procedures on paper - they could really care less about what happens to their clients. Oh, sure, they want to preserve their reputations, but only so that they can make more money off of other-people's-money. If this guy had made millions instead of losing it he'd have gotten a promotion (as long as the firm got their cut).
The F-35 is also needed for the Navy. I agree you could ditch 1/3 versions from a US perspective, but would that really save much money? The Air Force variant of the plane is likely the cheapest anyway (though why they can't just use the carrier version I do not know).
I am pretty impressed that they managed to drive the cost of the "cheap" plane above that of the "expensive" one...
The big improvement post-XP was 64-bit support. That started out really rough, but at this point it is fairly mature. Considering I'm running 8GB on my Windows box right now, I don't think I could really go back to XP. I stuck with XP up until 7 (and held off on the hardware upgrade until 7 worked out the kinks in Vista).
Win2K was before the consumer/NT kernel merge. It was pretty close to XP, however.
You didn't really get that many people who migrated from Win 9X to Win2K. There was a vocal minority who held the opinion you bring up, but you won't find too many average Joes who ever ran Win2K.
ME was an accident, but this was CONSUMER versions of windows, and NT/2000 weren't really consumer-oriented. XP was when the two branches were essentially merged, which is of course why it was such a big improvement for everybody.
MS doesn't even need software people on the board.
All they need are board members who ask for a laptop running the new version of Windows and take a few days to try it out.
Only a complete idiot could have actually used a Win8 demo and concluded that they weren't going to have problems getting people to accept it.
Most likely the board just showed up to watch the presentation on the upcoming EPS numbers, signed off, collected their $200k or whatever, and went home.
I'm not a windows guy. My laptop is a macbook pro and my day to day workstation is debian. However, I recently built a windows gaming computer and I like windows 8. Is it different? Yes. Does it have a learning curve? Yes. In the end it's stable, solid, easy to use, and looks nice.
Here is the problem - you don't already use Windows. 95% of the reason you buy Windows n+1 is because you already have a huge stack of Windows n software and you don't want to throw it away and buy all new software.
That is what Windows 8 is throwing away. If you don't already use Windows 7/etc then of course it isn't a big deal, because the learning curve is an issue anyway and you have to buy all new stuff anyway.
The reason MS has 90% of the market is because you can install 10-year-old software on it and it works just fine. I'm not aware of ANY other OS that delivers that level of compatibility, unless you're talking about mainframes.
I've posted this sort of thing elsewhere on slashdot, but since it isn't showing up in Google I'll summarize.
Tech companies do well under their founders, and that is because those founders were essentially selected by the market. Their boards are hands-off - either because they lack power (founder has too many shares), or know better than to use it.
Founders don't stick around forever, eventually their hand-groomed successor takes over. Usually they do well, though Ballmer clearly is an exception.
Once the board really gets power then everything is run by MBAs and the company goes downhill fast.
All that said, companies would do well to have boards that actually understand the business the company is in. Sure, engineers with good management skills are somewhat rare, but for what board members get paid I think they could be found.
Honestly, if they just shipped Windows 7 service pack 2 and called it Windows 8 chances are that MS would be doing a lot better.
Oh, sure, maybe in 10 years they're still facing big challenges on the desktop, but that doesn't mean that they need to destroy billions of dollars in shareholder value NOW by pretending they're a start-up with no established base. That is the kind of thinking they need on their phones (but with tie-ins to things like Office, Outlook, Exchange, Sharepoint, etc), not on their already-dominant desktop OS.
For whatever reason they're letting nuts-and-bolts thinking dominate their marketing and not the other way around. MS keeps living in some dream-world where they can stop supporting legacy hardware, only support an OS for two years like they're Ubuntu, change the APIs and development tools every three years, and so on. They forget that the companies that buy their OS do so because they can upgrade their OS once a decade, run that ERP system they spend 5 billion dollars on for two decades, and not have to completely retool their entire enterprise in a week when they roll out an upgrade of their OS.
The fact that your IE6-only application STILL can be run on a supported version of Windows is the reason that MS is still in business. Sure, companies understand nothing lasts forever, but they'll take 10 years over 2 any day of the week. Consumers love their new and shiny tablets, but just wait and see how happy they are when they're no longer new and shiny and yet they still have to replace them every two years because the OS is obsolete and the battery no longer works. Imagine proposing a two year desktop replacement cycle at work, and then realize that those corporate-discounted desktops and laptops are still WAY cheaper than the short-term hardware that is being proposed to replace them.
Some will no doubt say that supporting all those legacy technologies costs MS millions of dollars a year. This is VERY short-sighted thinking. If you turn it around what you're saying is that by spending a few million dollars a year you can cement your dominance in a market that delivers BILLIONS of dollars in revenue annually.
The other big item is compatibility. In Win8 non-Metro apps are basically considered second-class citizens. The whole reason people run WIndows is because it runs all the software they ALREADY have. When was the last time MS completely eliminated an API? Have they ever done this? I wouldn't be surprised if I could fire up a dos prompt and run Sidekick in it (Win95 certainly supported this).
For the longest time the reason to not switch away from Windows was that you didn't have to throw away all your existing investments. MS has been trying to ditch all the legacy support for the last few years, and what they don't realize is that their legacy support is their main differentiator. The IE6-only application is the whole reason businesses buy Windows, as much as it makes all the ivory tower types rip their hair out (and for good reason).
When MS comes out with a new product that treats existing software as second-class citizens and makes users learn a whole new UI they suddenly put themselves on a level playing field with all the other new things out there. Before grandma had to decide between a fancy new tablet that worked differently, or another PC that worked exactly the same as their last one. Now grandma has to choose between the fancy new tablet and another PC, BOTH of which are equally disadvantaged by learning curve and new software investment. As a result, Windows has to compete on a level playing field against the entire market, and it should be no wonder that it is losing.
The irony is that if Ubuntu had stuck with the old Gnome 2 interface (at least in terms of superficial appearance) they might be mopping up on people who just want something that looks like what they're already used to.
Slashdot Mirror
Agree ME was a consumer version of windows - hence it was an "accident" that I omitted it.
2000 was not. I doubt you'd have found it pre-installed on almost any PC from a major OEM. I will agree that it was much more useful as a consumer OS than previous NT incarnations.
My post was not a comment on sane desktop management or what the best desktop OS was at any point in time. It was merely a list of the branding issues MS had for just their consumer OS version - the one that comes pre-installed on 95% of all the PCs sold anywhere. New marketing team, new numbering/lettering/dating/whatever scheme.
I missed ME, and I appreciate the joke, but this was focused on consumer releases. CE/NT were not consumer desktop releases.
Well, it could be worse...
Have to agree. I work in a fortune 500 company and stuff like email/fileshares/network has outages on at least this kind of frequency, and I imagine it costs way more for a company to do it alone than to outsource it to Google.
I think the Google outages just get way more publicity. If some Exchange outage results in 10% of my company not getting emails for an hour, there is a good chance the other 90% won't hear about it at all, and many of the 10% won't notice either (away in meetings, and of course the client-server design of Outlook is a bit more failure-tolerant). Heck - a few years ago a location with 10k employees lost power for almost an entire day due to some kind of utilities mess-up.
Outages on large enterprise systems happen a lot more often than people realize.
If you want to find a consumer version of Windows that doesn't get support updates you'll have to go back to Windows ME. Even XP STILL gets updates (though not for much longer).
Nobody is suggesting Android phones should get security updates forever. However, there is a big difference between 10 years from the last sale and dropping support while the phones are still in the stores.
It would be nice if they just defined a standard stable driver ABI, like most proprietary OSes, or required the drivers to be open source, like Linux. You really need one or the other.
Imagine if MS release a special version of Windows for every model of PC out there, and they never updated any of them. No business would want to touch that. Instead MS defines a standard driver ABI and they've only changed it a few times in the last two decades (3.1->9X/NT, 9x->XP, then the 64 bit switch around Vista), and then they leave the drivers up to the device vendors and provide OS updates for a decade after they stop selling it. The result is that if you bought a printer in 2001 on XP, you can still use the printer on XP with security updates today, and you could probably at least go to a more current 32-bit Windows version without losing your printer and buy another 5-10 years if you had to (and there is a decent chance you'll find a 64-bit driver in which case you can just go to the current version and get about 15 years out of it).
Sure, Windows support isn't perfect, but we're comparing months to years here - Android comes nowhere near to even RedHat/Canonical support timelines, let alone what MS/Apple deliver.
With android there is exactly one phone available at any time that will get updates for 1.5 years from date of FIRST sale, and it isn't available on every carrier.
If you want a Nexus phone TODAY on Verizon, for example, your only choice is a Galaxy Nexus, and I doubt you'll get an update after this year for that phone (it is possible, but who can say).
The whole strength of Android is supposed to be the incredible diversity. All that diversity goes away if you start out by tossing all but one model of phone at any given time. Even the Nexus phones get fairly minimal support - the Nexus One only got updates for only a few months after it was last sold as far as I'm aware (not sure if there were any post-GB security updates, but it did not get ICS 11 months after it was discontinued). I would consider strong support to be all security updates for at least two years after the last day sold. Microsoft delivers security updates for 10 years after an OS is last sold, but nobody else really comes close to that.
There are many contracts you aren't given the option to sign, like signing yourself into slavery, or signing away your rights if you're injured due to neglect (oh, every ski slope will ask you to sign one of those, but the only effect they really have is to discourage lawsuits - they have little legal power).
The major carriers have similar terms and conditions on their contracts and are near-monopolies.
All of this points to a need tor regulation. Nobody is saying carriers can't sell brand A vs B phones. They're just saying that they should receive security updates only for the duration of these contracts (which would mean two years from the LAST one sold).
The problem is that CM doesn't support phones for that long either. Take the G2 - the last official CM release for that was Gingerbread, and Cyanogen himself owns one of those. The Nexus S, which has very similar hardware, is on Jelly Bean.
CM support is hit-and-miss, and it is hard to predict whether it will exist or for how long. While the CM team is doing great work, it really isn't a substitute for vendor-supported security patches.
That seems like a reasonable minimum. The sad thing is that most flagship Android phones are updated for about six months after they FIRST go on sale, no two years after they LAST go on sale. Even the Nexus phones are only updated for 1.5 years after they FIRST go on sale.
The Nexus One was sold until it was replaced by the S in November 2010. It did not receive ICS a year later. It did get GB a month later, but I'm not sure if it got any security updates after that.
The previous Google-branded phone was the ADP. It was sold until Jan 2010 (I believe - certainly no other Google-branded phone was available before then), and it was already out of date having missed Eclair three months before.
The only way you even get 1.5 years of updates on a Nexus-branded phone is to buy it on launch date. If you bought a Nexus 4 today I'd be skeptical that you'd get much more than a year of updates.
Believe it or not at my company's mail room they have an X-ray machine and a glovebox. They don't use them on every package, but if they get one that looks suspicious they will use them. From what I understand this is pretty common post-9/11. I doubt they have any way to test anything they find in the packages, but if the x-ray looked like a bomb or the letter in the glove box contained powder they'd likely abandon the building and call the police.
A trader, who is allowed to execute trades, and has all the machinery to do so at his disposal, sends out orders to buy one billion dollars of stock and relies on the settling period, which I believe is three days (may be five).
Yes, we know what he was trying to do. The question is how the firm let it happen in the first place.
If I go into the expense system at my employer and punch in that I spent one billion dollars out-of-pocket on inkjet toner the system will not just mail me a check for a billion dollars. Such a transaction (even if legitimate) would go all the way up to the CEO for approval, along with checks for compliance with purchasing rules/etc. The system is likely capable of printing any figure on the check the company wants, but it is designed with safeguards so that ordinary employees can't go spending arbitrary amounts of money. If I said I bought $50 worth of printer toner then most likely it would just mail me a check and if I didn't turn in the receipts they'd deal with it after-the-fact.
The problem with trading firms is that they spend other-people's-money. That means that they only care about having procedures on paper - they could really care less about what happens to their clients. Oh, sure, they want to preserve their reputations, but only so that they can make more money off of other-people's-money. If this guy had made millions instead of losing it he'd have gotten a promotion (as long as the firm got their cut).
The F-35 is also needed for the Navy. I agree you could ditch 1/3 versions from a US perspective, but would that really save much money? The Air Force variant of the plane is likely the cheapest anyway (though why they can't just use the carrier version I do not know).
I am pretty impressed that they managed to drive the cost of the "cheap" plane above that of the "expensive" one...
The big improvement post-XP was 64-bit support. That started out really rough, but at this point it is fairly mature. Considering I'm running 8GB on my Windows box right now, I don't think I could really go back to XP. I stuck with XP up until 7 (and held off on the hardware upgrade until 7 worked out the kinks in Vista).
Win2K was before the consumer/NT kernel merge. It was pretty close to XP, however.
You didn't really get that many people who migrated from Win 9X to Win2K. There was a vocal minority who held the opinion you bring up, but you won't find too many average Joes who ever ran Win2K.
I intentionally traced the non-NT chain, as those WERE the consumer versions of Windows.
NT didn't have all the crashing issues, but good luck finding printer drivers or especially games that worked on it.
Good point - replay is much harder if the connection is bidirectional allowing for challenge/response.
ME was an accident, but this was CONSUMER versions of windows, and NT/2000 weren't really consumer-oriented. XP was when the two branches were essentially merged, which is of course why it was such a big improvement for everybody.
MS doesn't even need software people on the board.
All they need are board members who ask for a laptop running the new version of Windows and take a few days to try it out.
Only a complete idiot could have actually used a Win8 demo and concluded that they weren't going to have problems getting people to accept it.
Most likely the board just showed up to watch the presentation on the upcoming EPS numbers, signed off, collected their $200k or whatever, and went home.
I'm not a windows guy. My laptop is a macbook pro and my day to day workstation is debian. However, I recently built a windows gaming computer and I like windows 8. Is it different? Yes. Does it have a learning curve? Yes. In the end it's stable, solid, easy to use, and looks nice.
Here is the problem - you don't already use Windows. 95% of the reason you buy Windows n+1 is because you already have a huge stack of Windows n software and you don't want to throw it away and buy all new software.
That is what Windows 8 is throwing away. If you don't already use Windows 7/etc then of course it isn't a big deal, because the learning curve is an issue anyway and you have to buy all new stuff anyway.
The reason MS has 90% of the market is because you can install 10-year-old software on it and it works just fine. I'm not aware of ANY other OS that delivers that level of compatibility, unless you're talking about mainframes.
I've posted this sort of thing elsewhere on slashdot, but since it isn't showing up in Google I'll summarize.
Tech companies do well under their founders, and that is because those founders were essentially selected by the market. Their boards are hands-off - either because they lack power (founder has too many shares), or know better than to use it.
Founders don't stick around forever, eventually their hand-groomed successor takes over. Usually they do well, though Ballmer clearly is an exception.
Once the board really gets power then everything is run by MBAs and the company goes downhill fast.
All that said, companies would do well to have boards that actually understand the business the company is in. Sure, engineers with good management skills are somewhat rare, but for what board members get paid I think they could be found.
Honestly, if they just shipped Windows 7 service pack 2 and called it Windows 8 chances are that MS would be doing a lot better.
Oh, sure, maybe in 10 years they're still facing big challenges on the desktop, but that doesn't mean that they need to destroy billions of dollars in shareholder value NOW by pretending they're a start-up with no established base. That is the kind of thinking they need on their phones (but with tie-ins to things like Office, Outlook, Exchange, Sharepoint, etc), not on their already-dominant desktop OS.
For whatever reason they're letting nuts-and-bolts thinking dominate their marketing and not the other way around. MS keeps living in some dream-world where they can stop supporting legacy hardware, only support an OS for two years like they're Ubuntu, change the APIs and development tools every three years, and so on. They forget that the companies that buy their OS do so because they can upgrade their OS once a decade, run that ERP system they spend 5 billion dollars on for two decades, and not have to completely retool their entire enterprise in a week when they roll out an upgrade of their OS.
The fact that your IE6-only application STILL can be run on a supported version of Windows is the reason that MS is still in business. Sure, companies understand nothing lasts forever, but they'll take 10 years over 2 any day of the week. Consumers love their new and shiny tablets, but just wait and see how happy they are when they're no longer new and shiny and yet they still have to replace them every two years because the OS is obsolete and the battery no longer works. Imagine proposing a two year desktop replacement cycle at work, and then realize that those corporate-discounted desktops and laptops are still WAY cheaper than the short-term hardware that is being proposed to replace them.
Hate to self-reply, but just one more thought:
Some will no doubt say that supporting all those legacy technologies costs MS millions of dollars a year. This is VERY short-sighted thinking. If you turn it around what you're saying is that by spending a few million dollars a year you can cement your dominance in a market that delivers BILLIONS of dollars in revenue annually.
You hit the nail on the head.
The other big item is compatibility. In Win8 non-Metro apps are basically considered second-class citizens. The whole reason people run WIndows is because it runs all the software they ALREADY have. When was the last time MS completely eliminated an API? Have they ever done this? I wouldn't be surprised if I could fire up a dos prompt and run Sidekick in it (Win95 certainly supported this).
For the longest time the reason to not switch away from Windows was that you didn't have to throw away all your existing investments. MS has been trying to ditch all the legacy support for the last few years, and what they don't realize is that their legacy support is their main differentiator. The IE6-only application is the whole reason businesses buy Windows, as much as it makes all the ivory tower types rip their hair out (and for good reason).
When MS comes out with a new product that treats existing software as second-class citizens and makes users learn a whole new UI they suddenly put themselves on a level playing field with all the other new things out there. Before grandma had to decide between a fancy new tablet that worked differently, or another PC that worked exactly the same as their last one. Now grandma has to choose between the fancy new tablet and another PC, BOTH of which are equally disadvantaged by learning curve and new software investment. As a result, Windows has to compete on a level playing field against the entire market, and it should be no wonder that it is losing.
The irony is that if Ubuntu had stuck with the old Gnome 2 interface (at least in terms of superficial appearance) they might be mopping up on people who just want something that looks like what they're already used to.