Yeah but most people don't want to waste their precious leisure time researching which of the twenty-odd Linux-ey things out there might really run all their media software and hardware without issues. Info from Linux advocates about what to get isn't always trustworthy.
If someone produced a platform called, say, "MediaLinux" that was pretty much guaranteed to be nicely set up as a DVD and photos and videos and CD-ripping platform (and could obviously also run FurryFox Ooo, etc.), and came pre-setup with Wine and a user-friendly database that listed the current Wine compatibility ratings of most popular Windows apps, and what you could and couldn't expect to do on the system, then maybe maybe most home Windows users might have a credible alternative.
But until the Linux community gets organised enough to be able to actually tell Windows users what the sensible alternative IS for home use, then people are going to cling to the known product that they think is the safest option.
I have an old version of Adobe Acrobat, and just over a year ago I thought that I'd consider upgrading to the latest version.
So I downloaded and installed the demo version for the latest release, to see what goodies it had.
Not only did the Acrobat demo regularly try to sneakily "phone home", but when it did, it glitched my system, and caused conflicts with some other background processes that had a similarly cavalier attitude to thinking that THEY owned the computer (hello MS Update).
In the end, the program went into a death spiral where the failed home-phoning glitched the software itself in the middle of file operations, and somehow the thing ended up progressively corrupting its own files, until two weeks into the four-week demo period, the thing stopped twitching and finally died, and I had to unpick the unholy mess it left behind by hand, because even the uninstaller no longer worked.
Now, the sad thing is, for all I know, the program without the "phone home" stuff might well have been stable. I'd already bought an earlier version, and was serious about getting the update, partly from a misplaced sense of customer loyalty, and partly because I thought that the Adobe colour-conversion facilities might be better than on the free or near-free third-party PDF-editor apps.
But what the demo showed me was that (a) Adobe felt entitled to write stuff into their software to do stupid and dangerous things to my PC that I wasn't aware that I'd agreed to, and that I'd normally associate with malware, vaporising any residual sense of loyalty or trust, and (b) that there's no way in hell that I was going to run a piece of dangerous buggy shit like that on any important computer system.
So bye-bye Adobe product sale, and bye-bye any hope of me buying another Adobe product until they can prove that they can be trusted. I don't want that stuff anywhere near my hard drives.
Personally I think that there should be a list of companies whose products shouldn't be allowed onto any government or company (or educational) networks without breaking the network's security certification, and while Adobe are pulling shit like this, they ought to be on the list.
Yep, MS could get into trouble for market abuse for their current inconsistencies over who is "allowed" XP and who isn't.
If they'd simply pulled the plug on XP totally, and said, "that's it, we aren't going to sell XP any more, because it's old and we don't want to be lumbered with the after-sales support forever", then that might be a legitimate manufacturer's decision.
But they didn't do that, because they didn't want to lose the netbook market. So they said that netbook manufacturers could continue to buy, install, and sell-on XP, but laptop manufacturers couldn't. When you say to a company, "We have a product, we're selling it to other people, but we refuse to sell it to you to work with your products, because we now want you to buy a different product from us", then that starts to get dodgy.
It's a bit like if a car-seat manufacturer has two ranges of car seats, their older smaller range and their new wider deluxe range. They want manufacturers to build the wider seats into all new luxury cars that can take them, but if they discontinue the older range, they'll lose the section of the market that supplies cars where the newer seats don't physically fit. So they continue to sell both ranges, but tell manufacturers that they are "banned" from selling the older seats fitted to the larger cars, even if those same cars have been sold fitted with those same seats in the past. That level of interference is getting into "illegal restraint of trade" territory.
The question is, how much control should a dominant component manufacturer have over how their products are used? Should they be allowed to micromanage what people do with their products with these sorts of restrictions and conditions? If a product has already been certified for XP, should they be allowed to then tell a manufacturer that they can still buy copies of XP, but they're are no longer allowed to preinstall them on those particular machines because new MS policy is that those particular customers should be buying something else? Even if this upsets both the suppliers and the customers?
Now to me, it sounds like MS are probably legally in the wrong here (as they have been so many times before when it comes to OEM contracts). And they probably know that they're in the wrong, but figure that the stakes here are so high that they'd rather break the law and worry about the consequences later... after all, none of their suppliers are going to want to sue them for fear of unofficial retaliation.
So this customer has decided, look, this is complete s**t - I should be able to buy the current software that I want on the machine that I want, without my supplier saying that they aren't allowed to do that because of some arbitrary rule imposed illegally on them by MS. So she figures, (a) it's unlawful and unfair, (b) someone should do something about it, (c) the laptop manufacturers won't, (d) she has the receipts that prove that this illegal behaviour by MS has cost her money, and (e) if it's illegal, and she's provably been damaged by it, then she's in a position to take a stand and sue, and maybe have the court ruling force MS to stop breaking the law (as she sees it).
Microsoft don't know what the hell they are doing. Sell stock now.
POSSIBILITY #2
Microsoft DO know what they're doing, and realise that the company is on the edge of total collapse, partly due to their poor image, and they have do do something risky and major to change their business model and their public perception. They're prepared to make a major loss paying people all over the country to tell customers that Win7 isn't total crap, and to show them how well it runs. Things really are that desperately bad, and MS know it, but they figure that they don't have much choice, desperate times call for desperate measures. Sell stock now.
POSSIBILITY #3
MS have spotted a genuine opportunity, not for retailing, but for buying up a load of high-value retail sites in good locations at the bottom of the market. They have a hell of a lot of money in the bank, and they don't really know what to do with it. They don't know how to make money on the web, their acquisition of Yahoo's brains-pool failed, the Office and Windows revenues are set to dry up with no obvious replacement products, and the billions they're spent on R&D haven't really produced anything. And banks aren't as safe as they used to be.
So a Walmart guy comes up to them and says, Dudes, with all that cash there's a real investment opportunity here. Shops are going bust all over, and its a buyers market. With all that cash, you should be able to buy up retail space in strategic locations dirt-cheap, then kit out the stores as MS Stores and run 'em for a few years as a public relations exercise. If the shops do well you make money, if they do badly you write it off as a PR cost. Then when the economy recovers and the retail space prices rise again in a few years, you own all these key sites, and you get to sell them off at a high profit. Or maybe hang onto them and rent them out. You currently have a lot of your core investment in intangibles that're high risk (Win7) or are going to have lost most of their value in five years time. Switch some of your investment portfolio into actual physical things that have a real physical worth, and are almost guaranteed to be worth more in five or ten years time than they are now in the current slump.
That'd be the sales pitch, anyway. How well it works out in practice probably depends on how well the Walmart guy really knows the retail property market. We shall see.
With free upgrade track from Vista (if the rumor is true)
No, Microsoft's free Win7 upgrade scheme was going to be entirely administered by participating OEMs, at their discretion (translation: it's a "final resort" option to allow OEM customer service people to placate angry Vista customers who bought premium hardware with Vista, have "issues" with it, no longer have XP as an option, and who can't be placated any other way). It's an OEM service based on an option agreed between MS and the OEM. It's not an option that MS themselves would be offering to end-users (and "retail Windows pack" customers are specifically excluded from the scheme).
Then again, they always have the option to change their minds and do this. IMO, a free upgrade to Win7 for all Vista customers would be an excellent idea, and would really help restore their reputation. The question is, are Microsoft bright enough to realise it?
According to the original article, Microsoft are =NOT= going to "Offer Free Windows 7 Upgrade To Vista Users".
They are going to set up a scheme whereby some participating OEMs may decide to participate and offer free Windows 7 upgrades to some of their Vista customers, who bought systems with the more expensive Vista versions, at the discretion of the OEM.
It'll be up to the OEMs whether they want to participate, and if they do, which customers they want to offer it to, how the scheme will be run, and who'll be told about it.
People who have Vista Home Basic or Vista Starter won't be eligible for an upgrade under the scheme, and neither apparently will people who bought retail copies of Vista. It's just for selected customers of selected OEMs, who bought Home Premium, Business or Ultimate.
MS themselves won't be directly offering customers anything under the scheme.
What it seems to be is a scheme that will give participating OEMs, who have aggrieved customers who bought premium products and got stuck with Vista, and who can't be placated any other way, the option of a freebie win7 upgrade to shut them up.
That certainly seems to be true in terms of speed.
I was in a shop yesterday and they had a deal on digital cameras. According to the box, the included software needed at least a 1.3Ghz Pentium processor if you were running Vista, but only a 500MHz Pentium if you were running XP/2000.
In the music-production world, the standard advice if you bought a computer with ME preinstalled was to "flush" ME, and go out and buy a copy of 98SE instead. ME simply wasn't stable enough.
Congrats if you had ME working fine on your hardware.
One of the more serious problems with ME is that some of the bugs corrupted the OS'es own system files when you tried to install updates.
Luckily MS identified the problem and issued updates. Snag was, to solve the problem, you had to install the updates, and if you'd just installed the updates, then the act of installing them had already corrupted your system.
I mean, maybe there was some a new installation service pack that let you install ME from scratch with the fixes already in place, or maybe MS issued a revised install image to OEMs... but if you bought a retail version... you were screwed.
I lost count of the number of times I tried reinstalling ME with the upgrades to try to find a sequence that wouldn't pre-f**k the OS.
I finally lost patience when I realised that it was showing multiple copies of system dialog boxes that were supposed to be modal.
Basically, the install that people like me saw was broken. The code was all out of sync with itself. Processes that were automated and that were supposed to wait for something to happen before proceeding would just steam ahead. I think this is why the program installation and OS upgrade stuff usually failed... it'd stack up a queue of tasks that were supposed to be implemented in a strict order, and then the later tasks would start activating before the first tasks were finished. One task would be writing stuff into the registry, and another would barge in and start doing its stuff in the middle.
So the #1 rule on ME was that whenever you saw a dialog that asked you whether you wanted to reboot now and install, you always had to click "later", and then reboot manually, with _never_ more than one program update at a time. Unfortunately, with MS's own OS-update installation stuff, "strictly one update per reboot" wasn't always possible. So while a lot of stuff would install nicely on ME, Microsoft's own stuff often wouldn't. And then you'd be back to a corrupted OS installation again. At this point you'd try to use system restore, and find that system restore had a bug that meant none of your restore points created after a certain date would work. You could download a hotfix that would solve that for future restore points, but that hotfix was one of the ones that corrupted the OS files...
Ugh. It's all coming back.:( I'd blotted this out of my memory as a traumatic experience. I had multiple machines from different manufacturers with different manufacturers' chipsets... they all ran 98 and 98SE well, and all of them had the same basic problems with ME.
Most of the problems seemed to come down to the OS being told to carry out a list of tasks in strict sequential order, and not bothering to wait until one item was completed before starting the next. That's why the "Do you want to reboot now to install the software?" option usually failed... the OS would launch the bit of code that started writing the reboot install files, and then it'd get impatient, and cut the power and reboot before the files had finished been written.
That's what tends to make me tend to believe the story that the problem was due to a last-minute speed-optimisation run before release. It felt like the sort of snafu that would never have survived beta-testing, but might have been introduced at the last moment by someone thinking, "Hey, maybe we can make the boot sequence even faster if we say that the OS doesn't have to wait for one process to finish before it starts the next!", and doing a search-and replace that destroyed ME's ability to lock out one task until another had finished.
The clincher was when I realised that if you had a button that would launch a "system modal" dialog, which was supposed to shut out all other user interaction until that dialog had been dealt with... if you clicked the button several times, q
System restore on ME came with a date bug. Any restore files saved past a certain date couldn't be restored.
So the OS came with a "System Restore" safety net for technical problems, that the user was told about and encouraged to use, but when you needed it... it turned out not to work any more.
There was a certain irony that after all the fuss about the Millennium bug, MS's Millennium OS still managed to have a bad date bug.
If you'd removed SysRest and a bunch of other things, its possible that you fluked on a minimal configuration of ME that was actually okay. But for most people, it was a buggy pile of shit that should never have been released.
One story I've seen (which has a sort of authentic ring to it) is that the last prerelease beta was actually pretty good, but that they then did one last set of optimisations before release, and broke it.
A pilot is supposed to have done this on a test flight while he was flying past a hotel where an aviation conference was taking place.
:)
Got told off for it. Innocently explained that there was no risk, since it was a simple one-gee manoeuvre. Was told politely but firmly Never To Do It Again. Probably through gritted teeth.
Well, I was actually thinking of thinking of Microsoft's old policies, but I know that policies are subject to change.
I've just had a quick look at the MS site:
There's a document called "Windows Life-Cycle Policy", published 2002, updated April 2008, which has a footnote on XP
3 As of April 2008, Microsoft is extending availability of Windows XP Home Edition for OEMs to install on Ultra Low-Cost PCs. The new OEM end date will be the later of either June 30, 2010, or one year after the general availability of the next version of Windows.
, which is interesting but odd, because I thought that a little after that, they announced that XP OEM availability was going to stop during 2008.
So maybe some MS guys haven't read their own policy documents, or maybe policy is being remade on the fly, or maybe the MS site isn't linking to current info. Maybe all three.
The "Microsoft Support Lifecycle Policy FAQ" says, for "Consumer, Hardware, and Multimedia products"
Microsoft will offer Mainstream Support for either a minimum of 5 years from the date of a productâ(TM)s general availability, or for 2 years after the successor product (N+1) is released, whichever is longer. Extended Support is not offered for Consumer, Hardware, and Multimedia products.
Is XP Home a consumer product? If so, it'd seem to be that the five-year limit has already expired, and the two-year cutoff, starting from the release of Vista on Jan 30 2007 (date from Wikipedia), gives an end date for "XP Home" mainstream support of... some time last week.
If "XP Home" isn't eligible for extended support, that would suggest that support for that OS version is now already over.
Of course, if XP Home doesn't count as a consumer product, things are different, and if XP Pro is entitled to an extended support period, then I guess XP home gets to ride on its coat-tails.
For business and developer products, the FAQ says:
Extended Support for the 5 years following Mainstream support or for 2 years after the second successor product (N+2) is released, whichever is longer.
So this would suggest that XP Pro has just switched from mainstream support to extended support, and will stay there for five years? I guess that coincides with your "early 2014" date.
Technically I suppose it might be difficult for MS to support XP Pro without also supporting XP Home, especially if they're still selling Home to OEMs, but given how badly they seem to want to get rid of XP, I dunno. I guess there are probably legal restrictions in the EU that might make it difficult for them to continue selling XP Home OEM for netbooks and other entry-level devices, while also declaring it to be obsolete and unsupported.
Actually, I thought that Apple missed out on the market for school computers for tots. I figured that they could have teamed up with Disney and produced a range of brightly-coloured desktop Imacs with big Mickey Mouse ears, a Disney-themed desktop, and large cartooney-style application window frames.
The sort of thing that you could put in the corner of a creche, or that rich parents could buy for the nursery. Customers could choose between Mickey Mac and Minnie Mac, and there'd be brightly-coloured plastic snap-on cosmetic customisation options.
If MS stops supporting XP, perhaps there's a business opportunity here for third-party support companies to step in and do for XP what they've been doing for Linux.
Part of me hopes that MS do the decent thing and keep supporting XP... but another part of me wonders what would happen if they continued their run of self-destructing business decisions, so we could all watch the current MS car crash continue. It'd be kinda interesting to see MS abandoning XP, and the third-party support community stepping in to fill the void, and maybe doing a decent job of it.
I was thinking of registering DashSlot.com and putting the symbol "-|" in all my SlashDot signoffs, to see how many people got confused and started visiting the wrong site... but someone registered it before me.
The naming system is stupid, and it's getting stupider.
The old "Home/Professional" distinction doesn't work any more as a ranking system. The home user is likely to want lots of bells and whistles like media players and video stuff that the "professional" IT department may not want enabled on their users' desktop machines. The CEO of an "Enterprise" might want a top-end laptop with video conferencing, but most of the "enterprise" employees won't be getting that.
An "Ultimate" edition isn't really an ultimate edition, because there are (hopefully) going to be service packs and updates.
A "starter" edition suggests that the user is going to upgrade that copy of Windows at some point, but the "starter edition" is targeted at entry-level machines that might not be able to grade. And so on.
It's just totally screwed up.
Now, here's a radical idea. If you need to produce different editions for different markets, how about starting by defining those markets, creating different editions market-by-market (like they did with "Windows Server"), and naming each edition after the sector that it's aimed at (like they did with "Windows Server")?
So if you're producing a cut-down Win7 for netbooks, call it "Windows 7.0 for Netbooks".
If you're producing a different cut-down version for basic desktop machines for people working in office environments, call it "Windows 7.0 for Office Desktops".
Don't call the "everything" version "Ultimate" or "Ultimate Premium", call it something more accurate and less shouty, like "Windows 7.0 Complete Edition".
Rather than starting with the OS specifications, then generating lots of different arbitrary cut-down versions, giving them abstract names, and only then starting to think about who's going to want to buy each version, start with the customer. Think about what your customers actually want, and use the naming system to tell them how to get it.
Do you sell to schools and universities using a different pricing plan? Then don't sell them a generic "Home" or "Professional" edition, release "Windows 7.0 for Schools", decide the pricing, and then finalise the specs and the branding differences (different default desktop layout, wallpapers, default shortcuts, colour schemes) later. With customer-centric editions, if you make a mistake in the specs, and decide that "Windows 7.0 for Schools" really needed some feature that you'd left out, you can add it as a freebie update... because the "edition" is no longer defined by the specifications, it's defined by the business model, pricing plan, and targeted customer base. If two versions aimed at different markets end up having almost the same specs, and just different cosmetics and included apps, it doesn't matter.
I mean, hell, if MS want to do a "Windows 7.0 for Callcentres", which is basically "Windows 7.0 for Business" with the video facilities stripped out, then that's fine, isn't it?
IFAICS, the only problem with this approach is that it'd need Microsoft to have people who actually understand the needs of the different customer-bases that they're trying to sell to. Instead of their marketing guys relying on being able to come up with whizzy names that impress the MS board, they'd actually have to do some old-fashioned legwork and get to understand something about their actual market.
They'd have to have some idea what their customers actually wanted.
I was just adding a google sitesearch bar to my website and testing it when this happened, so all the results I saw flagged as dangerous were mine.
This site is potentially harmful? Aiiieeee!
'Twas a great relief to realise some minutes later that it wasn't just me but the entire internet that was helpfully flagged as potentially harmful. Whew.
In the OJ case, even a 1:100,000,000 match was unreliable as evidence that OJ definitely left the blood, because after OJ provided a blood sample, someone from the police department booked the blood out and visited the crime scene with OJ's blood in his pocket, and I think it was supposed to be after that that the incriminating bloodstain (on the gate?) was discovered.
So, given that the LAPD police wanted OJ really badly, what's your estimated likelihood that the person at the crime scene with the container of blood (with motive, means and opportunity) planted that bloodstain?
If you figure that the chances that that blood was planted is one in a hundred, then the estimated certainty of his guilt, based purely on that one blood stain, would have to be downgraded accordingly. If you figure that the blood match is a dead cert, the thing that becomes relevant is the odds that the sample was left maliciously.
If you figure that it's more likely than not that the sample was planted, then the claimable evidence of guilt from that one piece of evidence drops to less than 50:50.
This doesn't mean that he's "really" more likely to be innocent than guilty: it means that on that one piece of evidence he's more likely to be found not guilty. If you factor in other pieces of evidence, the composite "claimable guiltiness" factor could increase again.
However, in the OJ case, once it had been established that the jury couldn't trust the LAPD to handle evidence competently, the rest of the evidence became suspect, too, and the jury had to decide that the case wasn't safe.
Similarly with the LHC safety assessment. If the LHC claim that the certainty that the LHC is safe is overwhelming, then the focus switches to how certain we are that the calculations are meaningful. If their record of safe/unsafe was wrong 50% of the time (if 50% of the things that they said were "safe" blew up), then it would be difficult to say that we knew from their assessment that the odds that it was safe were significantly better than that 50%.
Of course, an "unreliable" assessment of high safety doesn't magically make the plant physically unsafe, but what it does mean is that the claimable safety of the plant, based on that one assessment, as judged by a third party, would be lowered. If more evidence accumulated suggesting that the plant was safe, the "claimable safety" of the plant would rise again.
Of course, the underlying safety of the plant would be the same regardless of what the assessments said, or how good or bad they were... but that underlying data isn't available to us.
A "twat" is supposed to be someone who does something that they think makes them look cool, without realising that they've failed to carry it off, and actually they're just making themselves look a bit daft.
If you wanted to use an insulting word to refer dismissively to people genuinely worried about the LHC, "twat" is simply the wrong word. You might get away with "twat" if the sentence had been "Anyone who says...", but not for "Anyone who thinks..."
OTOH, if you're an LHC scientist, and you try to present a blokey air by calling people who object to your project names, and you use the wrong name, then that might be considered a slightly twatty thing to do.
Uh... the idea that black holes were unstoppable monsters that devoured everything in their path and could never, ever be destroyed didn't come from the "tinfoil hat" brigade... it came from 1960's physics textbooks and peer-reviewed papers.
So the current researchers are saying: "Look, we know that we told you all that stuff about black holes being immortal, and we said that it was mathematically proved, and that the people who were sceptical about the idea were idiots who didn't understand the math... but... um... that was all basically phooey."
"Now we know how things REALLY work, and we can tell you quite confidently that the opposite is true, and people who are sceptical about the new arguments are idiots who don't understand the math."
At this point, the Member of the Public is entitled to raise one eyebrow and say "But isn't that what you told us the last time?"
Slashdot Mirror
Win8=ReactOS?
If someone produced a platform called, say, "MediaLinux" that was pretty much guaranteed to be nicely set up as a DVD and photos and videos and CD-ripping platform (and could obviously also run FurryFox Ooo, etc.), and came pre-setup with Wine and a user-friendly database that listed the current Wine compatibility ratings of most popular Windows apps, and what you could and couldn't expect to do on the system, then maybe maybe most home Windows users might have a credible alternative.
But until the Linux community gets organised enough to be able to actually tell Windows users what the sensible alternative IS for home use, then people are going to cling to the known product that they think is the safest option.
So I downloaded and installed the demo version for the latest release, to see what goodies it had. Not only did the Acrobat demo regularly try to sneakily "phone home", but when it did, it glitched my system, and caused conflicts with some other background processes that had a similarly cavalier attitude to thinking that THEY owned the computer (hello MS Update).
In the end, the program went into a death spiral where the failed home-phoning glitched the software itself in the middle of file operations, and somehow the thing ended up progressively corrupting its own files, until two weeks into the four-week demo period, the thing stopped twitching and finally died, and I had to unpick the unholy mess it left behind by hand, because even the uninstaller no longer worked.
Now, the sad thing is, for all I know, the program without the "phone home" stuff might well have been stable. I'd already bought an earlier version, and was serious about getting the update, partly from a misplaced sense of customer loyalty, and partly because I thought that the Adobe colour-conversion facilities might be better than on the free or near-free third-party PDF-editor apps.
But what the demo showed me was that (a) Adobe felt entitled to write stuff into their software to do stupid and dangerous things to my PC that I wasn't aware that I'd agreed to, and that I'd normally associate with malware, vaporising any residual sense of loyalty or trust, and (b) that there's no way in hell that I was going to run a piece of dangerous buggy shit like that on any important computer system.
So bye-bye Adobe product sale, and bye-bye any hope of me buying another Adobe product until they can prove that they can be trusted. I don't want that stuff anywhere near my hard drives.
Personally I think that there should be a list of companies whose products shouldn't be allowed onto any government or company (or educational) networks without breaking the network's security certification, and while Adobe are pulling shit like this, they ought to be on the list.
If they'd simply pulled the plug on XP totally, and said, "that's it, we aren't going to sell XP any more, because it's old and we don't want to be lumbered with the after-sales support forever", then that might be a legitimate manufacturer's decision.
But they didn't do that, because they didn't want to lose the netbook market. So they said that netbook manufacturers could continue to buy, install, and sell-on XP, but laptop manufacturers couldn't. When you say to a company, "We have a product, we're selling it to other people, but we refuse to sell it to you to work with your products, because we now want you to buy a different product from us", then that starts to get dodgy.
It's a bit like if a car-seat manufacturer has two ranges of car seats, their older smaller range and their new wider deluxe range. They want manufacturers to build the wider seats into all new luxury cars that can take them, but if they discontinue the older range, they'll lose the section of the market that supplies cars where the newer seats don't physically fit. So they continue to sell both ranges, but tell manufacturers that they are "banned" from selling the older seats fitted to the larger cars, even if those same cars have been sold fitted with those same seats in the past. That level of interference is getting into "illegal restraint of trade" territory.
The question is, how much control should a dominant component manufacturer have over how their products are used? Should they be allowed to micromanage what people do with their products with these sorts of restrictions and conditions? If a product has already been certified for XP, should they be allowed to then tell a manufacturer that they can still buy copies of XP, but they're are no longer allowed to preinstall them on those particular machines because new MS policy is that those particular customers should be buying something else? Even if this upsets both the suppliers and the customers?
Now to me, it sounds like MS are probably legally in the wrong here (as they have been so many times before when it comes to OEM contracts). And they probably know that they're in the wrong, but figure that the stakes here are so high that they'd rather break the law and worry about the consequences later ... after all, none of their suppliers are going to want to sue them for fear of unofficial retaliation.
So this customer has decided, look, this is complete s**t - I should be able to buy the current software that I want on the machine that I want, without my supplier saying that they aren't allowed to do that because of some arbitrary rule imposed illegally on them by MS. So she figures, (a) it's unlawful and unfair, (b) someone should do something about it, (c) the laptop manufacturers won't, (d) she has the receipts that prove that this illegal behaviour by MS has cost her money, and (e) if it's illegal, and she's provably been damaged by it, then she's in a position to take a stand and sue, and maybe have the court ruling force MS to stop breaking the law (as she sees it).
POSSIBILITY #1
Microsoft don't know what the hell they are doing. Sell stock now.
POSSIBILITY #2
Microsoft DO know what they're doing, and realise that the company is on the edge of total collapse, partly due to their poor image, and they have do do something risky and major to change their business model and their public perception. They're prepared to make a major loss paying people all over the country to tell customers that Win7 isn't total crap, and to show them how well it runs. Things really are that desperately bad, and MS know it, but they figure that they don't have much choice, desperate times call for desperate measures. Sell stock now.
POSSIBILITY #3
MS have spotted a genuine opportunity, not for retailing, but for buying up a load of high-value retail sites in good locations at the bottom of the market. They have a hell of a lot of money in the bank, and they don't really know what to do with it. They don't know how to make money on the web, their acquisition of Yahoo's brains-pool failed, the Office and Windows revenues are set to dry up with no obvious replacement products, and the billions they're spent on R&D haven't really produced anything. And banks aren't as safe as they used to be.
So a Walmart guy comes up to them and says, Dudes, with all that cash there's a real investment opportunity here. Shops are going bust all over, and its a buyers market. With all that cash, you should be able to buy up retail space in strategic locations dirt-cheap, then kit out the stores as MS Stores and run 'em for a few years as a public relations exercise. If the shops do well you make money, if they do badly you write it off as a PR cost. Then when the economy recovers and the retail space prices rise again in a few years, you own all these key sites, and you get to sell them off at a high profit. Or maybe hang onto them and rent them out. You currently have a lot of your core investment in intangibles that're high risk (Win7) or are going to have lost most of their value in five years time. Switch some of your investment portfolio into actual physical things that have a real physical worth, and are almost guaranteed to be worth more in five or ten years time than they are now in the current slump.
That'd be the sales pitch, anyway. How well it works out in practice probably depends on how well the Walmart guy really knows the retail property market. We shall see.
No, Microsoft's free Win7 upgrade scheme was going to be entirely administered by participating OEMs, at their discretion (translation: it's a "final resort" option to allow OEM customer service people to placate angry Vista customers who bought premium hardware with Vista, have "issues" with it, no longer have XP as an option, and who can't be placated any other way). It's an OEM service based on an option agreed between MS and the OEM. It's not an option that MS themselves would be offering to end-users (and "retail Windows pack" customers are specifically excluded from the scheme).
Then again, they always have the option to change their minds and do this. IMO, a free upgrade to Win7 for all Vista customers would be an excellent idea, and would really help restore their reputation. The question is, are Microsoft bright enough to realise it?
This is genuinely hilarious. You know, I haven't laughed so much all week. Laughter is healthy. I feel good. Thanks MS!
They are going to set up a scheme whereby some participating OEMs may decide to participate and offer free Windows 7 upgrades to some of their Vista customers, who bought systems with the more expensive Vista versions, at the discretion of the OEM.
It'll be up to the OEMs whether they want to participate, and if they do, which customers they want to offer it to, how the scheme will be run, and who'll be told about it.
People who have Vista Home Basic or Vista Starter won't be eligible for an upgrade under the scheme, and neither apparently will people who bought retail copies of Vista. It's just for selected customers of selected OEMs, who bought Home Premium, Business or Ultimate.
MS themselves won't be directly offering customers anything under the scheme.
What it seems to be is a scheme that will give participating OEMs, who have aggrieved customers who bought premium products and got stuck with Vista, and who can't be placated any other way, the option of a freebie win7 upgrade to shut them up.
I was in a shop yesterday and they had a deal on digital cameras. According to the box, the included software needed at least a 1.3Ghz Pentium processor if you were running Vista, but only a 500MHz Pentium if you were running XP/2000.
I thought, ouch.
In the music-production world, the standard advice if you bought a computer with ME preinstalled was to "flush" ME, and go out and buy a copy of 98SE instead. ME simply wasn't stable enough.
One of the more serious problems with ME is that some of the bugs corrupted the OS'es own system files when you tried to install updates.
Luckily MS identified the problem and issued updates. Snag was, to solve the problem, you had to install the updates, and if you'd just installed the updates, then the act of installing them had already corrupted your system.
I mean, maybe there was some a new installation service pack that let you install ME from scratch with the fixes already in place, or maybe MS issued a revised install image to OEMs ... but if you bought a retail version ... you were screwed.
I lost count of the number of times I tried reinstalling ME with the upgrades to try to find a sequence that wouldn't pre-f**k the OS. I finally lost patience when I realised that it was showing multiple copies of system dialog boxes that were supposed to be modal.
Basically, the install that people like me saw was broken. The code was all out of sync with itself. Processes that were automated and that were supposed to wait for something to happen before proceeding would just steam ahead. I think this is why the program installation and OS upgrade stuff usually failed ... it'd stack up a queue of tasks that were supposed to be implemented in a strict order, and then the later tasks would start activating before the first tasks were finished. One task would be writing stuff into the registry, and another would barge in and start doing its stuff in the middle.
So the #1 rule on ME was that whenever you saw a dialog that asked you whether you wanted to reboot now and install, you always had to click "later", and then reboot manually, with _never_ more than one program update at a time. Unfortunately, with MS's own OS-update installation stuff, "strictly one update per reboot" wasn't always possible. So while a lot of stuff would install nicely on ME, Microsoft's own stuff often wouldn't. And then you'd be back to a corrupted OS installation again. At this point you'd try to use system restore, and find that system restore had a bug that meant none of your restore points created after a certain date would work. You could download a hotfix that would solve that for future restore points, but that hotfix was one of the ones that corrupted the OS files ...
Ugh. It's all coming back. :( I'd blotted this out of my memory as a traumatic experience. I had multiple machines from different manufacturers with different manufacturers' chipsets ... they all ran 98 and 98SE well, and all of them had the same basic problems with ME.
Most of the problems seemed to come down to the OS being told to carry out a list of tasks in strict sequential order, and not bothering to wait until one item was completed before starting the next. That's why the "Do you want to reboot now to install the software?" option usually failed ... the OS would launch the bit of code that started writing the reboot install files, and then it'd get impatient, and cut the power and reboot before the files had finished been written.
That's what tends to make me tend to believe the story that the problem was due to a last-minute speed-optimisation run before release. It felt like the sort of snafu that would never have survived beta-testing, but might have been introduced at the last moment by someone thinking, "Hey, maybe we can make the boot sequence even faster if we say that the OS doesn't have to wait for one process to finish before it starts the next!", and doing a search-and replace that destroyed ME's ability to lock out one task until another had finished.
The clincher was when I realised that if you had a button that would launch a "system modal" dialog, which was supposed to shut out all other user interaction until that dialog had been dealt with ... if you clicked the button several times, q
So the OS came with a "System Restore" safety net for technical problems, that the user was told about and encouraged to use, but when you needed it ... it turned out not to work any more.
There was a certain irony that after all the fuss about the Millennium bug, MS's Millennium OS still managed to have a bad date bug.
If you'd removed SysRest and a bunch of other things, its possible that you fluked on a minimal configuration of ME that was actually okay. But for most people, it was a buggy pile of shit that should never have been released.
One story I've seen (which has a sort of authentic ring to it) is that the last prerelease beta was actually pretty good, but that they then did one last set of optimisations before release, and broke it.
You strap them to the wings, obviously. Duh.
Got told off for it. Innocently explained that there was no risk, since it was a simple one-gee manoeuvre. Was told politely but firmly Never To Do It Again. Probably through gritted teeth.
It'll probably never, ever happen again (ever).
Wish I'd seen it.
I've just had a quick look at the MS site:
There's a document called "Windows Life-Cycle Policy", published 2002, updated April 2008, which has a footnote on XP
, which is interesting but odd, because I thought that a little after that, they announced that XP OEM availability was going to stop during 2008.
So maybe some MS guys haven't read their own policy documents, or maybe policy is being remade on the fly, or maybe the MS site isn't linking to current info. Maybe all three.
The "Microsoft Support Lifecycle Policy FAQ" says, for "Consumer, Hardware, and Multimedia products"
Is XP Home a consumer product? If so, it'd seem to be that the five-year limit has already expired, and the two-year cutoff, starting from the release of Vista on Jan 30 2007 (date from Wikipedia), gives an end date for "XP Home" mainstream support of ... some time last week.
If "XP Home" isn't eligible for extended support, that would suggest that support for that OS version is now already over.
Of course, if XP Home doesn't count as a consumer product, things are different, and if XP Pro is entitled to an extended support period, then I guess XP home gets to ride on its coat-tails.
For business and developer products, the FAQ says:
So this would suggest that XP Pro has just switched from mainstream support to extended support, and will stay there for five years? I guess that coincides with your "early 2014" date.
Technically I suppose it might be difficult for MS to support XP Pro without also supporting XP Home, especially if they're still selling Home to OEMs, but given how badly they seem to want to get rid of XP, I dunno. I guess there are probably legal restrictions in the EU that might make it difficult for them to continue selling XP Home OEM for netbooks and other entry-level devices, while also declaring it to be obsolete and unsupported.
Actually, I thought that Apple missed out on the market for school computers for tots. I figured that they could have teamed up with Disney and produced a range of brightly-coloured desktop Imacs with big Mickey Mouse ears, a Disney-themed desktop, and large cartooney-style application window frames.
The sort of thing that you could put in the corner of a creche, or that rich parents could buy for the nursery. Customers could choose between Mickey Mac and Minnie Mac, and there'd be brightly-coloured plastic snap-on cosmetic customisation options.
Part of me hopes that MS do the decent thing and keep supporting XP ... but another part of me wonders what would happen if they continued their run of self-destructing business decisions, so we could all watch the current MS car crash continue. It'd be kinda interesting to see MS abandoning XP, and the third-party support community stepping in to fill the void, and maybe doing a decent job of it.
Eric, -|
If that's the case, does that mean that they drop XP support when Win7 comes out?
And does "No longer supporting" include key validation services?
The old "Home/Professional" distinction doesn't work any more as a ranking system. The home user is likely to want lots of bells and whistles like media players and video stuff that the "professional" IT department may not want enabled on their users' desktop machines. The CEO of an "Enterprise" might want a top-end laptop with video conferencing, but most of the "enterprise" employees won't be getting that.
An "Ultimate" edition isn't really an ultimate edition, because there are (hopefully) going to be service packs and updates.
A "starter" edition suggests that the user is going to upgrade that copy of Windows at some point, but the "starter edition" is targeted at entry-level machines that might not be able to grade. And so on.
It's just totally screwed up.
Now, here's a radical idea. If you need to produce different editions for different markets, how about starting by defining those markets, creating different editions market-by-market (like they did with "Windows Server"), and naming each edition after the sector that it's aimed at (like they did with "Windows Server")?
So if you're producing a cut-down Win7 for netbooks, call it "Windows 7.0 for Netbooks".
If you're producing a different cut-down version for basic desktop machines for people working in office environments, call it "Windows 7.0 for Office Desktops".
Don't call the "everything" version "Ultimate" or "Ultimate Premium", call it something more accurate and less shouty, like "Windows 7.0 Complete Edition".
Rather than starting with the OS specifications, then generating lots of different arbitrary cut-down versions, giving them abstract names, and only then starting to think about who's going to want to buy each version, start with the customer. Think about what your customers actually want, and use the naming system to tell them how to get it.
Do you sell to schools and universities using a different pricing plan? Then don't sell them a generic "Home" or "Professional" edition, release "Windows 7.0 for Schools", decide the pricing, and then finalise the specs and the branding differences (different default desktop layout, wallpapers, default shortcuts, colour schemes) later. With customer-centric editions, if you make a mistake in the specs, and decide that "Windows 7.0 for Schools" really needed some feature that you'd left out, you can add it as a freebie update ... because the "edition" is no longer defined by the specifications, it's defined by the business model, pricing plan, and targeted customer base. If two versions aimed at different markets end up having almost the same specs, and just different cosmetics and included apps, it doesn't matter.
I mean, hell, if MS want to do a "Windows 7.0 for Callcentres", which is basically "Windows 7.0 for Business" with the video facilities stripped out, then that's fine, isn't it?
IFAICS, the only problem with this approach is that it'd need Microsoft to have people who actually understand the needs of the different customer-bases that they're trying to sell to. Instead of their marketing guys relying on being able to come up with whizzy names that impress the MS board, they'd actually have to do some old-fashioned legwork and get to understand something about their actual market.
They'd have to have some idea what their customers actually wanted.
This site is potentially harmful? Aiiieeee!
'Twas a great relief to realise some minutes later that it wasn't just me but the entire internet that was helpfully flagged as potentially harmful. Whew.
So, given that the LAPD police wanted OJ really badly, what's your estimated likelihood that the person at the crime scene with the container of blood (with motive, means and opportunity) planted that bloodstain?
If you figure that the chances that that blood was planted is one in a hundred, then the estimated certainty of his guilt, based purely on that one blood stain, would have to be downgraded accordingly. If you figure that the blood match is a dead cert, the thing that becomes relevant is the odds that the sample was left maliciously.
If you figure that it's more likely than not that the sample was planted, then the claimable evidence of guilt from that one piece of evidence drops to less than 50:50.
This doesn't mean that he's "really" more likely to be innocent than guilty: it means that on that one piece of evidence he's more likely to be found not guilty. If you factor in other pieces of evidence, the composite "claimable guiltiness" factor could increase again.
However, in the OJ case, once it had been established that the jury couldn't trust the LAPD to handle evidence competently, the rest of the evidence became suspect, too, and the jury had to decide that the case wasn't safe.
Similarly with the LHC safety assessment. If the LHC claim that the certainty that the LHC is safe is overwhelming, then the focus switches to how certain we are that the calculations are meaningful. If their record of safe/unsafe was wrong 50% of the time (if 50% of the things that they said were "safe" blew up), then it would be difficult to say that we knew from their assessment that the odds that it was safe were significantly better than that 50%.
Of course, an "unreliable" assessment of high safety doesn't magically make the plant physically unsafe, but what it does mean is that the claimable safety of the plant, based on that one assessment, as judged by a third party, would be lowered. If more evidence accumulated suggesting that the plant was safe, the "claimable safety" of the plant would rise again.
Of course, the underlying safety of the plant would be the same regardless of what the assessments said, or how good or bad they were ... but that underlying data isn't available to us.
If you wanted to use an insulting word to refer dismissively to people genuinely worried about the LHC, "twat" is simply the wrong word. You might get away with "twat" if the sentence had been "Anyone who says ...", but not for "Anyone who thinks ..."
OTOH, if you're an LHC scientist, and you try to present a blokey air by calling people who object to your project names, and you use the wrong name, then that might be considered a slightly twatty thing to do.
So the current researchers are saying: "Look, we know that we told you all that stuff about black holes being immortal, and we said that it was mathematically proved, and that the people who were sceptical about the idea were idiots who didn't understand the math ... but ... um ... that was all basically phooey."
"Now we know how things REALLY work, and we can tell you quite confidently that the opposite is true, and people who are sceptical about the new arguments are idiots who don't understand the math."
At this point, the Member of the Public is entitled to raise one eyebrow and say "But isn't that what you told us the last time?"