What I did wasn't near that invasive or complicated. I just right-clicked the taskbar-->Cortana and selected "hidden" and removed the search bar from the taskbar itself too. You can still search for local apps by opening the Start menu and just typing, just like Win7. It doesn't search online it just offers an option among the results to do a web search for whatever you type. I can't stand things taking taskbar space. I like tiny icons and nothing else.

Cortana is deeply embedded in a few places. The AllowCortana=0 registry setting (may be available via GP on some versions) can fully disable her.

If I don't notice her existence, does it matter?

Depends on how paranoid you are. 😀

The thing about paranoia, I've learned, is we lack enough imagination on how invasive these kinds of things actually can be.

I guess that depends on how you define invasive. If it's to the point of accessing and retrieving personal information, I understand that.

puts on tinfoil hat

Microsoft will eventually remove the option. They can do so with impunity; the general public is too entrenched in Windows to make a switch to anything else. Apple has gone off the deep end. Unless a third company rises to fill the gap that will be left by Microsoft's inevitable butchering of Windows, they will be able to do whatever they want. And they will butcher Windows at some point. They have billed it now as a SaaS product, but it will not be what was promised. I think a lot of people were thinking "great, now I'll never have to upgrade to a new version of Windows! MS will just keep pushing bugfixes and security updates and converge on the perfect OS as time goes on!" but that's not what will happen at all. MS will, of course, not be able to resist the temptation to completely revamp/change the OS. So many changes in software get pushed through for no other reason than because the prevailing theory - among developers as well as among the general public - is "if it's old, it's bad." So MS will keep pushing large, bloated updates which are nothing more than "new versions" of Windows in the traditional sense, except branded as Windows 10 still. They will break compatibility, require better hardware, and change entire parts of the user interface, forcing people to change things, upgrade their hardware, and relearn the OS just the same as always.

TL;DR it would be great if MS would have tried to create the "perfect UI", only ever push bugfixes, security updates, and new drivers, and asymptotically converge on a "perfect" Windows. But they will not do this. As a result the "Software as a Service" Windows 10 system will fail to provide anything better than the old system of buying a new computer with OEM Windows when a big upgrade comes out.

keenmaster486 wrote:

Microsoft will eventually remove the option.

It will be hard for them to do so completely, because of enterprise client backlash (the only clients that actually make MS money on Windows).

There are plenty of things I would not have thought corporate clients would accept, but they have [at least where I work, which is heavily security-sconcious]. So eventual acceptance of some form of cortana may happen, it will depend on how much corporate IT can control or divert it, which is not about the principles but about implementation details and which Microsoft will work sensibly with.

Things like Cortana are like the fabled frog in a pot of boiling water. As long as you change slowly enough, people adapt. Someone above mentioned XP activation. I remember that debacle. And now, nobody cares at all. It's just a given. Or how about the Pentium 3 unique serial number thing? People are quick to anger, and just as quick to forget. You just have to survive the initial outrage and let it blow over and become normalized. In ten years, our insurance company will be well aware of how many often we go out to eat, or drink beer with buddies, and it will just be the way it is. Given human nature, and especially capitalist motivation, that's kind of terrifying.

As far as the future of software, there are only really a handful of models, currently, that are known to work:

1) Free, as in open source. Linux, BSD, etc. Doesn't cost anything but time and effort, but it's not necessarily a well-organized project with clear goals and aims. Will probably never be the primary model - at least not without a secondary revenue stream.

2) Advertising-driven, as in Google. Leads to the inevitable outcome of "private" information being bought, sold, and shared between various organizations. Things change on a whim, and you've got no say. (Remember Bump? Or Google+?) We're at a point now where the public is just starting to understand the real cost of this model. I doubt it will matter in the long run. Free (or, no $$ anyway) is king.

3) The Apple model -- i.e., integration. I think this is what Microsoft was trying to do, but they're too naive to understand that OS X works as a series of point-releases, for free (or in the past, at a trivial cost), because of the mandatory hardware sales. From an accounting and user point of view, the OS is just the software what makes the computer work. Not a product unto itself. It doesn't have to be profitable. (But of course it is -- it's the reason to buy a MBP or an iPhone, for most people.) Apple has a case of the dumb right now, but they've been down many times before. Don't count them out yet.

4) Retail (to include volume licensing as a CapEx purchase.) This implies the product stays a discrete product. You're buying something tangible. Updates require some incentive, otherwise the customer base just keeps using the one they have perpetually. (Or, you get the silly "OEM" licensing stipulation... "You can only use this on this one computer, ever, then you have to stop and buy a new license. Promise?")

5) Subscription. The "as a service" model. You pay for the product on a schedule. Like anti-virus, or web hosting, or O365. Rolling updates, no guarantee of entitlement, and typically some automated form of enforcement that requires regular or constant connectivity. The product exists as long as there's a way to activate it. For those of us that are collectors and preservationists, it's the worst possible option.

I don't know how Microsoft intends to turn Windows into Mac OS. The industry, in general, has given the Surface line a resounding... "meh." All the IT geeks (a group to which I belong) and pointy-haired managers love them, because it's so futuristic to scroll and move windows by dragging your meaty paws all over a nice glossy screen, but touch is not going to be the way forward for productivity UI. Just ask Windows 8. And without touch, the Surface is just an expensive, buggy, poorly-designed laptop. So without the hardware lust of the Apple fan club, and without big, important-sounding updates to the OS as a tangible product (where, in the past, they mostly just re-organize the control panel, hide all your settings behind three layers of dialogs, and add a new skin), how is this going to work in the retail channel again?

In the enterprise sector, VL is often already consumed as a kind of a subscription model. Many (most?) just go the Software Assurance route and toss it on the OpEx pile. Many (most?) consumers just get a new version of Windows when they buy a new computer every few years. So what do you do with everybody else? If I buy Windows 10 Retail, do I get a lifetime of updates for free? THAT's clearly not going to work. So which way is it going to go? Ad-supported, or subscription?

Here's the thing with the Software as a Service model.

Most people were already effectively using this model - because every time MS would come out with a new version, they would buy it. It just ran on maybe a 3 or 4-year basis instead of a yearly basis.

The difference with SaaS is that, by entering into the agreement, you have forfeited your choice in the matter! If you choose not to "buy the new version" by renewing your subscription, MS will simply prevent you from using their software altogether.

The solution here is, obviously, to simply not use Microsoft. I'm already doing this, and it's great. There is a lot of excellent competition in the free software world that I don't even remember being there 4 or 8 years ago. Extrapolate forward and I think I see where this is going. People may use MS's SaaS - but MS will be on thin ice! People will have places to go if MS pisses them off, and already do.

I also think that eventually, MS, as a company that must make money in order to survive, will not be able to compete with a veritable army of volunteer programmers working on free software projects and releasing them at no charge to the end user. What we have here is essentially a group of people (developers) paying for the software and giving it away to the end users. How is this? The developers pay in terms of their time and work, which otherwise would be spent at their day jobs making money for themselves. Ultimately the software is NOT FREE. The developers pay for it! We should be eternally grateful to them.

Now of course, the above does not apply to developers working for companies who have a business interest in making open source software projects better. Or companies like Canonical who make their money by providing support services, and pay programmers to develop their code. I do think that this sort of thing will increase a great deal over the coming years. We could see a thriving "open source economy" with companies paying developers to develop free open source software, companies collaborating with each other, and independent developers also working on code. The reason this could happen is the same reason any transaction happens in the free market: everyone benefits. The developers benefit because they are getting paid. The companies benefit because they get to use the software for whatever business they conduct. The end user benefits because the software is free to use.

It's possible the open source system may supersede proprietary software altogether, because it may prove to actually be more economically efficient and financially viable in the free market.

But we'll see about that. It's just a possible outcome.

OSS has done really well in certain markets. For example, appliances -- that is, hardware in a datacenter that isn't a server. Everyone who is anyone runs a variant of BSD or Linux under the hood. Juniper's JunOS is BSD. PANOS is Linux. I think ArubaOS is too. Supervisor modules that require an IP stack are almost always Linux. It all comes down to royalties, being that there aren't any. I've seen a grand total of ONE device that was Windows CE under the hood -- an EMC SAN controller.

Certain other projects have done extraordinarily well for themselves, too. Apache, Firefox, Wireshark, Gimp, OpenSSL... A whole desktop OS though.. I dunno. I can't even remember how many "this will be the year of the Linux desktop" claims I've seen, and there are SO MANY reasons why that never came to pass.

Ironically, the model is changing in a way that could completely shuffle the deck. Microsoft are adapting to it somewhat, and stand to do well for themselves if they're agile enough. OTOH, it could also break them. Namely -- the move to the cloud. The desktop OS is becoming less and less relevant now, as user interaction moves from PCs to a myriad of devices. Android and IOS own the market there, and it's just expected to be able to consume services from those platforms. Windows' relevance is thus linked to a paradigm that is still solid now, but will the workforce be bound to a desk in a decade? Some, yes, absolutely. But probably not so many as today. MS's response to this has been a strong push to services -- Azure, O365, business intelligence, collaboration, unified communications. All of this fits the subscription model fairly well.

Again, that begs the question.... what happens to Windows? It's not really their flagship anymore. It's still important, but less and less all the time. The OS will be a commodity soon, as will the browser (which I think MS already learned.) Maybe in 2030, what's left of the PC market will be running Android. Or maybe Intel will step up and decide it needs to provide a reference software platform for its integrators.

Hm, this has gone a wee bit off topic. Apologies if this isn't interesting to anyone else.