Slashdot Mirror
Preventing Another Vista-like Release With Windows 7
CRE writes "An article at the OS News site details how Microsoft could best avoid Windows 7 becoming another Vista-esque release. The author advises Microsoft to basically split Windows in two. Windows 7 would be a new operating system based on the proven Windows NT kernel, but with a completely new user interface, with backwards compatibility provided by VMs. In addition, to please business customers and other people concerned with backwards compatibility, Microsoft should create 'Windows Legacy', basically the current Windows, which will receive only security and bug fixes. Relatedly, APCMag is reporting that Microsoft has moved Julie Larson-Green (the driving force behind Office 2007's Ribbon UI) over to work on Windows 7's interface."
Horrible idea, would never be put into practice. MS already spent years merging the 9x consumer brand into the NT-based line. There's no reason they would then spit it again and have to deal with two not fully compatible platforms, requiring a separate support base for each one.
Slashdot: Playing Favorites Since 1997
Keep It Simple Stupid the problem with Vista was that Microsoft wanted to make the Ultimate Operating System, that would put all other Operating Systems to shame (And give to us all the features they promiced us in Windows 95)... But with all the problems with such a large project then ended up with an OS that is arguable slightly better then their old one. I have tried myself to do ambisious projects and they always go over budget and over time, and end up having to do a lot of cuts. I learned not to go crazy and make the ultimate just get it to work correctly and impove on the other one, That way everyone is happy.
If something is so important that you feel the need to post it on the internet... It probably isn't that important.
It's called a service pack. And you can slipstream them right into the install (new XP discs sold today include SP2). There is no need to split it into a different product.
As long as Microsoft can maintain a quick pace of innovation, Linux will always be chasing behind it.
Once the problem becomes well defined and stable, Linux will catch up and O/S will commoditize.
The longer the release cycles- and the more windows UI changes with those releases, the more likely people will change to linux. I'm ready except for Everquest. Everything else is open source on my boxes now.
She was like chocolate when she drank... semi-sweet at first and then increasingly bitter.
...Microsoft has moved Julie Larson-Green (the driving force behind Office 2007's Ribbon UI) over to work on Windows 7's...
Oh, no...
As for the future Windows, I say build it to be a VM store, capable of taking on the personality of any VM---allowing you to have new fancy features as well as the legacy Windows (heck, maybe they should include everything, all the way to DOS, Win3.1, etc.). You don't really `need' an OS (assuming they figure out ways of enabling you to efficiently use the hardware from VM)---you might have a `primary' image that you use all the time, and a buncha others provided for compatibility with previous versions.
"If anything can go wrong, it will." - Murphy
How about releasing a single OS that scales suitably and automatically to the users' dynamic needs, rather than piling options on the user who neither knows nor cares what the options do.
"Make it go."
Can we get a "-1 Wrong" moderation option?
Deliver all the features you promise, and a few extras, when the time is right.
It's that simple.
Whether or not the Vista release was successful or not is generally troll bait but from my personal perspective it had none of the things I wanted and featured many things I didn't. I certainly won't be touching it until well after SP1 and even then only if there are several great games for me to play. It was a release "failure" to people like me who expected some goodies and a new Windows iteration but Microsoft delivered a more restrictive operating system. No thanks!
crazy dynamite monkey
If they're planning on making the next Windows UI mirror Office 2007 then count me on the list of people likely to never buy it. The Office 2007 UI is horrible and badly done. Never before with MS products have I felt the desire to kill someone after using a software. Well except for that time I tried using MS Plus but that's a whole nother article.
My sig of choice is Marlboro
I don't think the answer is to provide yet another windows version. Having a "new" and "legacy" version is only going to make the problem worse. Imagine if Microsoft came out now and said they were going to support XP for 5 more years with fixes as the "legacy" version. Now no one will be forced to upgrade. Many people like XP better but accept the fact that eventually they will need to go to Vista. For Microsoft, they want to keep people at their latest version because it is easier to support the newer and hopefully "better" code than the old.
I think a a better way is to do smaller incremental releases. Sure MS may only want to make people drop the 200$ every 4-5 years, but they could make make their service packs yearly and include more new features (similar to XP SP 2). Then when it comes to the next Windows release it hopefully will not be such a drastic change for users.
Mac users are a loyal fan base making huge OS Changes that breaks all sorts of compatability less damaging. People don't use windows because they are loyal to microsoft, or even like windows, they do so because all their software runs on it. Braking all that compatability would cause many users to rethink what OS they would rather use, free OS like Linux, OS with a good HUI like OS X, or perhaps try some other OS's if all their software goes down the drain then they will most likely feel a grudge toward Microsoft for obsoleting their software investment, and look somewhere else. By doing so I would figure that Linux could rise to about 30% market share, Macintosh would be about 20% and 45% towards windows and 5% going to other OS's from 95% market share to 45% would kill Microsoft or at leat cause a ton of problems for them... Or carry on as they did before and allow a slow leak in market share over decades which would lead to the normalize rate much further in the future vs. Jumping to it right away.
If something is so important that you feel the need to post it on the internet... It probably isn't that important.
Now if it actually incorporates VMs for backward compatibility built on the proven NT kernel, I think they may be onto something. What I see as the problem here is that Microsoft is going it alone like it has so many more times in the past. If this were a joint-venture between M$ and VMWare or some other company of that ilk, I could forsee this being a successful product.
Unfortunately, M$ won't do that and this product will be hyped to the max and actually provide a lackluster experience for users.
The game.
Take the Linux Kernel, Run every program as root, install a bunch of 3rd party drivers for cheap hardware that may or may not function properly. Have this in a distribution that is widly spread so about 90% of all people are using... See how good Linux holds up.
The Windows NT Kernel is actually a very good kernel. It is the fact that the rest of the OS is designed in a way that cause problems to occure.
If something is so important that you feel the need to post it on the internet... It probably isn't that important.
Vista wasn't ill received because of the incompatibility. Plain and simple, it was not a step forwards. It wasn't something that improved your working, playing or surfing experience in any way. If anything, it was a step backwards.
Add various real and perceived problems with privacy, the data hunger of MS, the dread of DRM/TCP and other rather negative reviews, and you see the reason why Vista wasn't the next Win95 hype.
The problem is that XP already has everything the user wants. It can play games, it's compatible with almost any current hardware right out of the box, there is no USB (WinNT) or WiFi (2k) that would require him to update, whatever hardware he wants to plug in, XP can take care of it. Whatever software he wants to run, XP can do it. DirectX10-only games are still far from reaching the shelves, and no business software that I'm aware of requires Vista. The user interface of XP has all the main features that make working, surfing and playing in Windows enjoyable, and all the kinks and wrinkles were also taken care of by third party software vendors (where "vendors" does not necessarily mean you had to pay anything for the soft).
Basically, the reason why Vista didn't sell like hot cakes was simple: It was not needed.
We used to have a Bill of Rights. Now, with the rights gone, all we have left is the bill.
with backwards compatibility provided by VMs
That just gave me an interesting idea: Why doesn't MS ship fully functional versions of previous OS's, wrapped in a VM, with newer versions? What would they lose? I know I'd be far less worried about upgrading to Vista if I knew I could load up a built-in VM of DOS 6.0 or Win98SE or WinXP and play all my favorite shareware games from the '90s as easily as the latest-and-greatest. Same goes for here at work...it would be nice to know that some of our older software could just be loaded in a VM until the vendors catch up with Vista. As long as they maintain security on the sandbox itself, they wouldn't need to worry overmuch about keeping the old OS up to date, and it's not like people would be buying Vista just to exclusively use it to run XP, but it would make for a much more obvious upgrade path than the current hard cutoff in backwards compatibility.
Unpleasantries.
Like MSFT couldn't make their own version of wine in half the time, or even use a VM. OS X did. the OS 9 for PPC VM was advanced enough to play OS 9 games in. Yes I did. Are you saying MSFT isn't smart enough to make such a compatiblity layer? Not even the MSFT basher in me believes that.
i thought once I was found, but it was only a dream.
The issue here is proprietary lock-in. If MS would fix all the architectural problems of MS Windows, it would basically be a new OS. It could keep parts of the kernel, but the userland interfaces would change so much that only VMs could keep compatibility -- and with them comes a huge resources consumption boost on an already heavy architecture. But resources are not the main issue: it is that the new applications would be so different from old ones that vendors would most likely do something cross platform and MS would loose proprietary lock-in.
Also, it would take so long that GNU/Linux would have a huge window of opportunity, with the added benefit of low resources usage and true backwards compatibility.
Finally, it would be so different from MS Windows and so much like GNU/Linux or the Hurd that people would see the king is naked.
Leandro Guimarães Faria Corcete DUTRA
DA, DBA, SysAdmin, Data Modeller
GNU Project, Debian GNU/Lin
I have been saying essentially the same thing for years and years since the discussions about Win32 weaknesses that cannot be fixed without a restructuring of the API.
The Win32 API is a complete mess. Backward compatibility is important to be sure. But the future of stable operating systems is also an issue. Apple couldn't have been bolder in their move to create OSX. They created an entirely new OS and provided some really buggy means to run OS9 software... believe me, it certainly sucked but it generally "worked." It was more than enough motivation for people to migrate to the OSX versions of the same packages they've been using, but for those not willing to make the move for whatever reason, they were able to limp by.
Applying the same idea to a new Microsoft OS would probably work better. Virtualization environments on the PC have come a long way in a relatively short time. One might even suggest that it's fairly mature technology. (I'm not quite ready to say that myself though.) But to provide backward compatibility through virtualization while at the same time creating something like "Win64" and making it completely new, more modern and at the same time tossing backward compatibility out the window (figuratively speaking) would probably bring new life into the "struggling under its own weight" OS and the company who makes it.
To have a right to do a thing is not at all the same as to be right in doing it
Its worse than that for Microsoft. The cost to develop a new OS has increased exponentially (with the complexity) since their 3.1/95 days. That trend isn't going to reverse, and it is going to become impossible for Microsoft to innovate and profit from the OS alone. That is why widespread support for ODF can break them, and why they are fighting it so hard.
The OSS model is working a lot better at spreading out the complexity and costs of innovating within an OS. Its simply a more sustainable "business" model than Microsoft's.
But what makes you think Windows is less stable and less secure than *Nix or OSX? Other than people and their dogs running Windows as administrators (that's more an education problem vs Windows security), Windows is not less secure than *Nix or OSX. In fact, things like file system security is better than *nix, IMO. Windows ACLs just own, it's a breeze to use them versus the obscure *Nix FS security.
And for stability? The only time my Windows box crashed was because of piece of shit ATI drivers. People need to get away from the Windows 9x crashing every 3 minutes mentality. XP is rock solid (didn't drive Vista enough to tell on it's stability). I am currently running a VWare GSX server on Windows 2003, the only time I reboot is to install OS patches. Crashes so far: zero, nada, zilch. Been running it for 7+ months. Hardware: Do it your own el-cheapo components.
The vast majority of Windows crashes are due to defective hardware and/or drivers. Ever installed an unstable driver on Linux? Ever had a hardware failure on OSX?
Two versions: server and workstation. That's it. No more "ultimate" or "home" or any other stripped down versions.
For Server: no client access licenses. When you buy a copy of the server software, you can have as many clients as you want. Each server version is capable of everything, including clustering, load balancing, and everything else.
For Workstation: one interface. It could be new or old, whatever, but exactly one. If it's new, we all need to learn the new version. Don't like that? Get Linux or a Mac.
Finally, both server and workstation should support a single hardware compatibility list. If your hardware isn't on the list, you can't load it; update the list monthly through Windows Update. There is Driver Signing already, but you can get around it by ignoring the warnings. Eliminate getting around the warnings.
...but, by introducing different product lines in their OS, Microsoft will only confuse the customer, and they're way too smart and customer oriented to allow something like that to happen.
What they really ought to do is something more like what Apple did with the Classic Mode environment for supporting OS 9 applications, which ran within OS X. Thing is, MS will probably have to support theirs indefinitely, while Apple was able to successfully kill Classic Mode within about 5 years.
You see? You see? Your stupid minds! Stupid! Stupid!
Want to hear the voice of GOD? cat
You are confused.
The platform that made them #1 was MS-DOS.
The rest was just inertia and network effects.
A Pirate and a Puritan look the same on a balance sheet.
The problems with Windows are all at the Win32 layer. This is a huge problem for Microsoft, since their biggest competitive advantage is backwards compatibility. There is a lot of Win32 software around. Hardly anyone runs Windows because they like Windows, they run Windows because they need to be able to use some bit of Win32 (or, in some cases, Win16) software, and trust Windows to do this more than WINE.
Ever had a hardware failure on OSX? You only get those when there's a bogon overload in the RDF generator.I am TheRaven on Soylent News
Shiny happy Windows style installers have been available for Linux since when the consumer version of Windows was still just a glorified MS-DOS shell.
The notion that there is some lack of "InstallShield for Linux" is one of the more absurd bits of FUD that's cropped over the years (now decades).
A Pirate and a Puritan look the same on a balance sheet.
While the FUD machine has done an admiral job at making Vista seem like a steaming pile, that's all it has been: FUD.
I've been using Vista since November of 2006, essentially days after it was released to MSDN, and it is without a doubt better than XP. The improvements are both obvious and subtle. I'm not going to list them all here, because others have done a good job already.
So if Vista is superior to XP technically, which was deemed by most as a great success, then Vista being a failure must be attributed to sales data. Many early reports showed Vista having poor sales, but those reports were flawed due to the fact that they compared the launch of Vista to the launch of XP. Vista launched Jan. 29th, long after the holiday season was over, where as XP enjoyed the entire holiday season to boosts its sales.
Once this was corrected, reports showed that Vista was selling on pace with XP. Indeed, as of March 2007, Vista's sales were double that of XPs.
In addition, despite being released to consumers and businesses separately, Vista's sales were only 4% behind XP, which was released to both simultaneously. In other words, Vista beat expectations by a long shot.
So it must be that sales of Vista have stagnated since March... opps, that's not true either. Apparently, Vista sold so well that it offset the massive hit Microsoft took as part of extending the Xbox 360 warranty to 3 years.
And then there is the wonderful story that Vista has somehow boosted XP sales, which is completely silly. It didn't boost XP sales. There was a larger proportion of XP sales than were expected, but the breakdown is about 80% Vista, 20% XP. Part of this is thanks to the FUD machine (good job guys) prompting some large OEMs, like Dell, to offer XP on lower end machines. Microsoft underestimated the FUD machine's ability to influence the market. (By the way, there were 7% more XP sales than were expected. Hardly a tidal wave of XP purchases.)
Sorry guys. I know you desperately want to believe that Vista is a failure, both technically and in terms of sales. But you're wrong on both accounts. 2 years from now, when 90% of PCs are running Vista, you'll probably still claim it's a failure, although you'll fall back to the technical side of things.
I'll be sure to bookmark my post and repeatedly link to it in all those flame wars.
As XP is the dominant OS out there, I point out that Vista so far is selling better than XP at the same point into XP's life. XP also had most of the same complaints now leveled at vista. I suspect by the time windows 7 (with the inevitiable delays) comes out, most will have an attitude of I'll buy 7 when you pry vista out of my cold dead hands.
You seem to be under the impression, there is competition and if MSFT does not do what is best for the customers, they will desert it in droves. Time and again MSFT has proved that its customer base is loyal to a fault and is a sucker for punishment. Now go back to the drawing board and come up with a plan for Windows-7 that will force all the weary recently upgraded to Vista finally dudes to plunck down more money to upgrade to Windows-7.
sed -e 's/Chuck Norris/Rajnikant/g' joke > fact
If you're looking for polish and consistency (well, those don't seem to coexist happily very often; think about it), look no further than Ubuntu, Kubuntu, Mandriva, Linspire (which will be based on Kubuntu if it isn't already), and Fedora. I look forward to KDE 4 for a polish that should surpass any polish or consistency you come to expect in Windows (yeah right; polish, maybe, but consistency on Windows? laughable) or Mac OS X (polish definitely, and consistency is still weird when it comes to the GUI; I hear that will change in 10.5), but that's just me. For polish now, like I said, look at K/Ubuntu, Mandriva, Linspire, etc.
'Yes, firefox is indeed greater than women. Can women block pops up for you? No. Can Firefox show you naked women? Yes.'
FUD or not, not everything is is easy for Joe Schmoe to install on Linux as it is on Windows. Besides apt-get or yum, or even cpan, it's a pain in the balls for the home user to install something and get it to work. Hell, even using yum or apt-get, a lot of times you still have to play around with conf files to get something to work. Too many years of being given that, and the average person won't spend the time playing around to get something to work.
http://www.mhall119.com
..create MORE 'Windows Legacy?' This is one of the major, abyssmal design choices in current Windows versions responsible for truck loads of the issues every Windows user and their grandma have come to hate said OS
This is at least somewhat correct. Legacy support is responsible for a lot of Windows's problems.
However, it's also perhaps the biggest single reason for its success.
They tried "Microsoft Gates" at one stage, but people got confused.
"I've got more toys than Teruhisa Kitahara."
The reason you're not seeing the business side jump all over this OS isn't because of just compatibility issues. It's the Genuine advantage.
For example. here where I work, we had Vista running everything most office workers need; Office, IE, SCT, Even wintegrate, which is an ancient terminal program from 96. There was three reasons we didn't go to vista. One was the System requirements we were not ready to meet, another was that F-secure did not have an official Vista version at the time, but the real reason we decided to stay with XP was simple. The Genuine Advantage is for lack of a better word a total pain in the ass.
In vista there are two ways of handling corporate keys. One with a Key Management server and the other with a Multiple Activation Key. Under KMS. You are required to have a KMS server on your network, tie it to DNS and give it your VLK (which can be changed if your old key is disabled and propagated to networked PC's). once you do that it will activate any Business version of vista automatically every 3-6 months without entering any keys, but if the computer is no longer on the network (say a Laptop) after 3 months, the system locks you out in a reduced functionality mode which can be described as useless.
The Second method; MAK isn't much better. basically MS handles the KMS for you. this means that you don't have to worry about traveling users not being disconnected from your network for too long since it works over the internet, but now MS is handling your activations, and you have to contact them every time you hit your quota in order to activate more windows. (which isn't as bad as it sounds. According to MS activation isn't counted against your licence count, and you can request indefinitely) However, if MS sees a huge activation spike. (say your activation rate average goes from 100 a day to 10000000 a day) they disable your key (which brings us to reduced functionality mode for all MAK'ed PC's) and then you must go to each and every MAK managed PC and change the key to a new one supplied by MS.
So basically, to use Vista you either have a server on your network and pray no one's laptop cripples while their on a business trip, or you contact MS until the break of dawn and pray that no one pirates your key so you don't have to touch 1000 Crippled PC's with the Dreaded "YOU ARE A PIRATE!" message
Office 2007, however, doesn't have the "YOU ARE A PIRATE!" system built in it and still has the old VLK licencing system like XP. I can guarantee that it's adoption in business is much higher than Vista. I know we're using it here, but Vista is sitting on the shelf.
In Soviet Russia, Trojan exploits YOU!
Interesting that MS users tend to be short-term orientated (won't put up with apps not working from one version of windows to the next), while OSX users tend to be long-term orientated (put up with the short term problems because they realize in the end, the better design is the right thing to do.)
I would go that much credit for the Mac Comunity. There is just a smaller portion of them overall and many have been useing Macs for a long time because there is a smaller comunity base they tend to feel more personally bonded to Apple. Apple could sell iPoopOnAStick and probably 1/3 of the Apple Faithful will buy it. There are probably about the same number of Apple Zealots and Windows Zealots who will stick to their product no matter what... But because Winodws has far more users the number of Zealots is a lot less percentage wise then Mac Zealots. If you look at my guestimate windows would still be the dominate OS in market share, but there would be a large amount of people going different ways.
If something is so important that you feel the need to post it on the internet... It probably isn't that important.
Just implement a process that involves actually throwing out 2/3+ of all the features before coding, then cycle that about three or four times to let the really needed items rise to the top. Then simplify, simplify, simplify.
"Just cause it fits in there - doesn't necessarily mean you should put it in there." ~ Words of wisdom from my sophomore roommate after 3 months of debauchery.
Get your tagline off my lawn.
on that note check out Altiris: http://www.altiris.com/
I'm ambivalent, at best, about the 31 Flavors of Windows. But it raises an interesting point...
Old computers that still kick A55 and would be a shame to throw out. I have a PIII-400 that stills does what it needs to, and a Mac 8500 that still does nearly everything I ask, except that IE Mac doesn't work on most sites. What we need to do, sometime <BLINK>REALLY SOON</BLINK> is to freeze a subset of computers and OS forever into an R2-Unit standard.
Recall that the R2 unit loaded into Luke's X-wing was the SAME unit that Obi-Wan used. How likely is it that ANYTHING we have on a computer now will even physically plug in, let alone work in 40 years. Some computers can do 90% of what we need from now until at least 20 years from now. Can we PLEASE pick a set of standards and let that class of computer be supported? For example, ATA-100, USB 2.0 (or Firewire800, I don't care), DVI, RJ-45... I have peripherals in my garage with no computer capable of connecting them. I still have a copy of X-Wing vs. Tie Fighter... as if. Something. Anything.
And Windows Whatever. XP, XT, 2KSP17. I don't care. That way, as we retire, the Geek Squad can say, "Do you want us to replace your computer? This one's seen a lot of wear." "Not on your life. That G4 Titanum and me have been through a lot together."
PS: The blink tag was fake.
Actually, they already do this:
For Microsoft to be talking about the next Windows release so soon after its first main release in 6 years, Vista, and potentially putting customers off buying into Vista now, speaks legions about just how bad Vista adoption is going (notwithstanding fanboi propaganda FUD like the /. article comparing Vista to OSX marketshare). It says to me that Microsoft must really be panicking, badly.
Thom Holwerda puts forward a convincing argument that Windows needs two operating systems, a backward compatible operating system and one on which future application development can be done. He is far less convincing in his contention that the Windows NT kernel is a good design for a long range committment as the basis for future software inovations at Microsoft.
Windows development got into trouble through poor design. In order to bundle application with Windows Microsoft consistently designed applications to be non-modular. Pieces of each application were scattered throughout Microsoft code, including the kernel. This meant that the total bundled software package became more and more unwieldy as development progressed. Adding a new application entailed rewriting all Microsoft software instead of simply adding a new module containing the new functions on top of the existing stack. As Microsoft's software became more and more unwieldy the development effort slowed until in Longhorn it failed.
Now Thom Holwerda is proposing that Microsoft start over by taking the NT kernel and throw all of the entertwined legacy code out of the kernel. This will make development on NT a lot easier. But what about the current set of applications? DRM will still be intertwined through the NT kernel and probably the other current applications will be also. This lack of modular design will still hamper development on NT even though by getting rid of the legacy code the development effort now becomes doable. And what of future applications? Are they going to be intertwined into the NT kernel just like the existing applications? If so, then the new NT development tree will eventually suffer a Longhorn like crisis serveral years down the road.
Microsoft might be able to slough off a lot of legacy cruft by switching to a NT kernel with the legacy code removed but the basic design flaws remain to bedevil future development of Windows NT. Microsoft would be best off to design a new operating sytem from scratch and get rid of the lack of modularity once and for all.
-----------
Steve Stites
Why comment on "Windows 7" at all?
It's obvious that MS is pushing the PR now in order to draw attention from Vista. Vista is a trainwreck, so they're playing the "look, shiney!" game.
Assorted stuff I do sometimes: Lemuria.org
More choices is absolutely the WORST idea. We already have what? 5-6 variants of Windows and that's just English. Two completely different versions of windows each with X variants would just complicate the problem.
What we need is Three things.
A. Make features, not bullet points. This means give us features to help us. Not a newly designed interface that just looks pretty. Make stability and bug free a FEATURE. Look for features we can't get elsewhere, and ways for us to extend it. That means don't worry about firewalls (ship with Zonealarm) don't give us a weak anti-virus and pretend that's a major feature. Don't pretend "integrated music player" is a feature. Microsoft's current beliefs are bullet points are better than other goals. Games that run at 60 fps are more important than games that are "fun". Office suite that integrate perfectly are better than bug free. Get over it and get us actual innovation. And if you offer Backwards compatibility with old windows code make sure it's 100 percent Backwards compatibility before you ship.
B. Ignore the side projects. Windows 7 is about WINDOWS not Media player, outlook, office, and the rest. Want to include those? Great make them bug free, and allow us to uninstall all of them, otherwise focus on Windows. Giving us 30 programs along with windows doesn't make you my friend, when I have to work around 29 of them to get MY functionality back.
C. Cut the price, cut the fat. Two versions of Windows. Upgrade for 100 dollars, Full for 200 dollars. don't try to nickle and dime us saying "well ultimate has..." Ultimate has shit. Either an upgrade or full and make them AFFORDABLE. When Windows costs more than any of 4 tvs I own. (Including a 52 inch CRT) that's a problem.
Vista died because no one needed it and no one wanted it, but Windows is slowly forcing it's bloated corpse on us. That's what caused the Vista Like release, an unwanted unneeded product who's only benefit is making Microsoft more money and looking pretty.
Exactly. New Coke was actually better. Blind testing proved that people loved it. It failed because people were terrified and resentful of change. They didn't hate the soda. They hated the change. The very idea of change made a delicious beverage taste bad.
There might be a lesson for those who long for large scale Linux adoption in that story.
Actually "New Coke" did fail because of taste. When New Coke was "enginered" it was tested for marketability by with "sip tests" (i.e. people taking a sip of the drink and comparing it to other versions). "New Coke" was sweeter than "Original Coke", therefore more people prefered it to the other candidates when taking sip tests. But when "New Coke" was relesed and people drank the stuff for real the taste was too sweet to bare.