Someone finally noticed...

 ...and it only took them four days. From Janet Swift, at I Programmer:

NetMarketShare Bows Out

This news item was supposed to be about the latest desktop browser statistics and the fact that during October Edge had experienced an increase in its share. However, more newsworthy is the fact the October is the final month for which NetApplications.com will collect these statistics.

Reporting the way in which Chrome has come to be the dominant the desktop browser, documenting Firefox's decline and, until recently, the failure of Edge to gain a foothold, has been a part of I Programmer's routine fare. So it came as a shock to discover that this months figures are the last, at least for the foreseeable future.

Well done, Janet Swift! As far as I can tell, you and I are the only two people to actually notice that huge banner announcement at the top of their stats page... and their home page... It makes me wonder what all the other chuckleheads are doing. 

Journalism!

NetMarketShare's closure announcement has received zero attention; their last set of released statistics, though, are a different matter

Days after posting their farewell until next time message, NetMarketShare's announcement they're shutting down after 14 years of service still hasn't generated a single tech media article or blog post (except, apparently, mine).

The evidence of NMS's influence, however, continues to trickle in,with the latest Chromium Edge browser numbers being the big story yesterday, and Wimdows 7's enduring popularity being the story today.

As reported by bleeping computer:

Windows 7 won't die, still second most popular operating system

According to NetMarketShare, Windows 7 saw a drop from 22.77% to 20.41% last month. The report shows that 20.41% of desktops still use Windows 7. Even worse, some are still using Windows XP, according to the report.

I can't help but wonder how bleeping computer will react next month, when they finally realize that frantically refreshing NetMarketShare.com isn't going to make the end-of-November numbers magically appear... because NMS aren't publishing any.

Also... in other news, Windows 7 is still the world's second-most-popular PC operating system, in spite of the fact that you can still get Windows 10 from Microsoft, for free (bleeping computer's article "helpfully" provides the steps). If you were wondering just how badly Microsoft pissed off their customers with the GWS fiasco... wonder no longer.

End of an era: NetMarketShare throw in the towel

Confirming something that I'd suspected for some time now, OS and browser tracking site NetMarketShare confirmed the obvious today:

IMPORTANT NOTICE:
After 14 years of service and being used as a primary source in tens of thousands of articles and publications, we are retiring NetMarketShare in its current form. October, 2020 is the last month of data. All billing for existing accounts has been stopped. All outstanding balances are being refunded.

Why? An upcoming change in browsers (https://github.com/WICG/ua-client-hints) will break our device detection technology and will cause inaccuracies for a long period of time.

In addition, we have focused on bot detection and removal as a key part of the quality control process. It is the most complex part of our codebase. As time has gone on, it has become increasingly difficult to manage this process. So, instead of accepting increasing levels of inaccuracy, we thought it would be a good time to call it a day.

Well... That escalated quickly.

Linux Watch: June 2020 Edition

It's been a while since I last blogged about OS user market share numbers.

In large part, the absence was due to boredom. Very little was changing month to month; with Windows 10 having finally passed Windows 7 in user share, no other tech blogs were writing about the subject, either. MacOS and Linux were stealing users from Windows for a while, but that trend slowed, then stopped, and then reversed itself. Increasingly, it looked like the personal computing paradigm that we'd lived with for twenty years was about to cement itself in place for the next twenty.

But an odd thing has been happening over the last couple of months. Linux, which had retreated allthe was to 1.36% user share just a few months ago, has been gaining ground rapidly, erasing all of its user share losses of the past year and then some.

Plus ça change...Microsoft proves, once again, that they refuse to learn from past mistakes

I had to shake my head at this report from betanews:

Windows 7 may be an aging operating system, but there are still plenty of individuals and organizations using it. With the end of support date of January 14 fast-approaching, Microsoft is getting twitchy and is eager for everyone to upgrade to Windows 10.

Having already started to notify Windows 7 hangers on that support is due to come to an end, the company is now ready to get a little more aggressive. If you haven't moved on from Windows 7, soon you will see full-screen notifications warning you that "your Windows 7 PC is out of support".

The messages are due to be displayed from the day after support ends. So when January 15 rolls around, anyone who has doggedly stuck with Windows 7 will find that they not only have no support and no security updates, but also that they are pestered by an invasive message delivered by a program called EOSnotify.exe.

[...]

The good news is that it will be possible to dismiss the message and opt to never see it again.

Where do I even start with this shit?

In case you missed it...

... Windows 10 finally managed to hit 50% market share, according to NetMarketShare.

Some takes on this development were, sadly, predictable, such as Computerworld's, "Windows 10 user share surges as loafers heed impending deadline." Others were more balanced, though, such as Mike Sanders at eTeknix:

It’s been around 4 years now since Windows 10 launched and if one thing has been made abundantly clear, it’s that people have been very slow in adopting the latest operating system platform. In fairness, there has been good reason for this. Firstly, it’s hard to ignore just how hard Microsoft shoved this down our throats. We got the option to update to it for free, but this was part of a huge campaign that saw Windows 7 and 8 users badgered for months to do it.

Since then, following more than a few problematic updates, Windows 10 doesn’t exactly have a glowing reputation with a lot of the PC community. As such, many have stubbornly refused to make the update.

Following the latest figures from Netmarketshare, however, it may have taken 4 years, but Windows 10 finally has a 50% operating system market share!

[...]

Microsoft is likely more than a little embarrassed that it has taken this long for people to adopt Windows 10. Particularly given how hard they pushed it. Like it or not, however, we fully expect that number to really start spiking towards the end of the year with that end-of-support for Windows 7 looming in the not too distant future.

If you haven’t updated to 10 yet, you’re likely going to. So, try and make the best out of a bad situation?

Oh, I plan to, Mike. I'll be switching by the end of the year... but to Linux, not Windows 10. All that remains is to pick the most AMD-gaming-friendly distro; I'm currently leaning towards Pop!_OS, but Manjaro and (naturally) Ubuntu are both in the running.

Windows 7 gained market share last month, while everybody else lost ground.Seriously, WTAF?

OK, I'll admit it... I have no idea what's going on with OS market share trends this year.

Last year seemed to have a clear pattern. With few exceptions, Windows consistently lost share each month to MacOS and Linux, with Chrome staying steady. This year, though, Windows is growing while everyone else has started to slip. That would seem to be the perfect story for Microsoft, who have been waiting for Windows 10 to finally start taking over the desktop/laptop OS market for years, and for tech punditry, who have mostly been expecting business and government deployments to finally make that happen as Windows 7 winds down to end-of-service in January of 2020.

But that's not what we see happening. Instead, Windows 10 is slowly sliding along with all over OS versions, while Windows 7 and 8.1 together posted gains that are strong enough to raise Windows' overall user market share of the desktop/laptop space. It's demented... and yet, it's also happening.

So, what the actual fuck is actually happening?

Comparing last month's numbers with this month's, by same OS version.

I could understand if some Windows 10 users were rolling back to Windows 7, or if some newly-minted Linux users had switched back to Windows 7 - after all, I've done that before, my own self.

But it's not just Windows 10 and/or Linux that are being cannibalized by Windows 7; according to NetMarketShare's numbers, MacOS users are also switching to Windows 7, apparently having already spent the Apple Tax to buy their (expensive) new Macs.

Only one possible explanation occurs to me, and I have no idea whether it's reasonable or not. I'm starting to wonder whether NetMarketShare's are being skewed by something which they themselves claimed to have dealt with before: botnets.

Windows 10 loses ground in latest NetMarketShare numbers

You didn't think I'd miss marking the 1st of March by posting the latest OS market share numbers, did you?

Windows 10 dipped from 34.29% last month to 34.06% this month, a drop of 0.23%. Windows 7 dipped slightly, too, from 42.39% to 41.61%, a drop of 0.78%. Windows 8.1 and XP both gained slightly (+0.10% and +0.65%, respectively), which was unexpected.

Windows lost ground overall, again, dipping by 0.13% to 87.66%. MacOS lost ground this month, though, dipping by 0.06% to 9.89%, which is the opposite of previous months. Linux gained overall, rising from 1.93% to 2.08%, an overall gain of 0.15%.

In short, none of this month's numbers make any sense, with Windows' most popular versions both losing ground to versions which were looking pretty much dead until right now, and none of the gains seemingly being enough to offset the losses. Which most likely means one of two things: (a) these numbers are some sort of weird anomaly, and next month's numbers will revert to the overall trend of previous months, or (b) this month's numbers contain actual errors, and NMS will be updating them later in the day with corrected values.

The third and perhaps less likely option is that some sort of shift to Linux actually is underway, that NMS's new methodology and revamped data sets were only able to obscure for so long before they reasserted themselves in the data. We'll have to see if NMS let this set of numbers stand, or what StatCounter and the Steam Survey have to say about the current state of the OS market.

Regardless, however, one thing is clear: this month's shifts in the market are still very small. Which means that we've now gone three full months since my mid-November prediction that we wouldn't see any significant shifts in the OS market. So, yes, I'm calling it: that prediction is confirmed. And since VR still isn't a thing, either, that give me a record of 3 and 1 for my prognostications, with only the Nintendo Switch performing better than I expected. Not too shabby, if I do say so myself.

Do you remember when WX was supposedly on pace to surpass W7 by November?

NMS's end-of-January numbers are out, and once again, WX has managed modest gains at the expense of Windows XP, while W7 and W8.1 remain mostly unchanged. And, no, WX still hasn't caught up to its nine-year-old rival.

WX gained, of course, from 32.93% to 34.29% (+1.36); W7 ticked down, from 43.08% to 42.39% (-0.69); W8.1 ticked down slightly, from 5.71% to 5.56% (-0.15); and XP slid the most, from 5.18% to 4.05% (-1.13). Except for W8.1's, all of these results are above the ±0.5% "noise threshold," but WX's gains are not enough to encompass the losses of W7, W8.1, and XP. Some of those former Windows users are going elsewhere.

Where are they going, you ask? By the looks of it, Apple. Windows' overall market share slid from 88.51% to 87.79% (-0.72), while MacOS grew its overall market share from 9.02% to 9.95% (+0.93), propelled by MacOS X 10.13 (from 3.53% to 4.46%, +0.93). An overall decline in Windows' user base probably isn't something that Microsoft want to see; yes, WX gained more than a percentage point to start the year, but the fact that those gains are mostly coming at the expense of the 16½ year old XP, rather than the market-leading W7, can't be good news, either.

Plus ça change, plus c'est la même chose.

Windows 10 gains market share, while Windows 7's share continues to remain the same.

There's both good news and bad news for Microsoft in NetMarketShare's latest market share statistics. On the up side, Windows 10 has gained nearly a full percentage point, moving from 31.95% to 32.93%. On the down side, those gains seem to be coming almost entirely at the expense of Windows 8.1, which dropped 5.97% to 5.71%; and Windows XP, which dropped from 5.73% to 5.18%. And Windows 7 stayed pretty much stable, dipping from 43.12% to 43.08%, a change which drops below my ±0.5% "noise threshold."

So, Windows 10 is still growing slowly, mostly at the expense of the and the least-popular Windows versions, while the most-popular Windows version, 7, continues its run as the new XP. With the new numbers being such an arguably mixed bag, it was really anyone's guess whether tech media sites would continue trumpeting Windows 10's modest gains as if they were some sort of triumph, or recognize that Windows 10 has failed to close the gap for yet another year, in spite of being freely available for the entirety of that year.

You can colour me slightly surprised, then, that MS Power User are taking the latter route, first out of the gate:

Around the middle of last year, we predicted, based on the then trend, that Windows 10 will overtake Windows 7 worldwide by the end of 2017.

Now the latest Netmarketshare numbers are in, and at least by their numbers, it appears that date may be pretty far off still.

It appears in the last quarter of 2017 the decline in Windows 7 numbers stabilized, while Windows 10 lost some momentum.

[...]

Last month Microsoft revealed there were 600 million active Windows 10 users, and we expect it would take around 800 million Windows 10 users to overtake Windows 7, something which may be achievable by the end of next year as enterprise users switch over in increasing numbers to Windows 10.

The change in tone, here, from confidently predicting that Windows 10 would overtake Windows 7 in 2017, to describing that same feat as something that "may be achievable," but isn't certain, in 2018, speaks rather eloquently to how much momentum Windows 10 lost as the year went along. Again, it's good news/bad news: Microsoft (mostly) stopped stepping on the self-laid rakes in their own yard as the year went along, but with no bad news to report, people also (mostly) stopped talking about Windows 10 entirely.

Windows 10 isn't new anymore. It isn't interesting, or exciting; consumers (and tech writers) seem to have tired of the relentless updating cycle, and now greed new feature announcements with a collective shrug, and the resigned knowledge that it means another round of problems while Microsoft tries to manage yet another updating cycle, even before the last one is complete. At least this time they'll at least be half-way there, which beats the Spring Creators Update's 33.7%, but still.

I've still got two more months to go before I can call it, but I'm still feeling pretty good about my Windows 10 prognostication.

UPDATE:

It looks like StatCounter's numbers show basically the same trend, according to Betanews:

While NetMarketShare’s monthly usage share figures show there to still be a fairly significant gap between Windows 7 and Windows 10 (in the older OS’s favor), rival analyst firm StatCounter has long reported the battle for the top spot to be much, much tighter.

So close is the race in fact, that in October it looked as if Windows 10 would easily pass Windows 7 at some point in the following month. Surprisingly, that didn’t happen, although the gap did narrow. It seemed all but guaranteed that Windows 10 would claim pole position in December, but incredibly it didn’t.

Unlike NMS, StatCounter show Windows 8.1 gaining market share over the past month, to 9.16%, which is exactly the opposite of NMS's result, leading me to suspect that something is up with StatCounter's methodology. Considering that NMS recently revamped their entire product offering to filter out bots and botnets, which were skewing their results, one has to wonder if StatCounter has done the same, or if there's some other reason for Windows 8.1's popularity in their data sets.

NetMarketShare's new online tools are more powerful than I first suspected... and also confusing...

Of all the ways in which NMS's free stat reports have annoyed me over the years, the most annoying is the way they segregate results. You can compare OS market share across desktop devices, for example, or across mobile devices, but not both at the same time... so you can't compare, say, Windows to Android, to see their growth and trends over time relative to each other.

StatCounter does offer this sort of comparison, but only as an overall picture (i.e. all Windows versions, all Android versions, all iOS and MacOS versions, &c.), which make for simple graphs... but if you do want to drill down a bit, you can only drill into one OS at a time. You can't, for example, compare Windows 10 and Windows 7 to the most popular versions of iOS, MacOS, and/or Android, all at once.

Apparently neither NMS nor StatCounter thought that people would want to do this... or, perhaps they did know that people would want to do this, and simply decided to reserve the more powerful functionality for their paid subscription services. Which is entirely fair; after all, these are businesses and not charities, and they have to make money somehow. I get that. As a dedicated online freeloader, though, I still found it ever so slightly irritating.

But the game has changed, thanks to NetMarketShare's brand new online tools, which were "completely rearchitected [...] to provide more accurate and accessible data," actually do allow you to pull from multiple datasets at once, making possible exactly the kinds of extremely nerdy comparisons that I've been idly wishing I could do, without having to pay to do them.

The results are, to say the least, surprising:

The surprise? According to NMS, Android is still only 33.48% of the market, well behind Windows' 45.41%. This is surprising because, to quote Wikipedia:

According to a January 2015 Gartner report, "Android surpassed a billion shipments of devices in 2014, and will continue to grow at a double-digit pace in 2015, with a 26 percent increase year over year." This made it the first time that any general-purpose operating system has reached more than one billion end users within a year: by reaching close to 1.16 billion end users in 2014, Android shipped over four times more than iOS and OS X combined, and over three times more than Microsoft Windows. Gartner expected the whole mobile phone market to "reach two billion units in 2016", including Android.^[307] Describing the statistics, Farhad Manjoo wrote in The New York Times that "About one of every two computers sold today is running Android. [It] has become Earth’s dominant computing platform."^[18]

StatCounter disagrees with both assessments, clocking both Android and Windows at about 38% of the market, but with Windows' share of the combined OS market growing in recent months... which makes no sense, since desktop/laptop PC sales have been declining.




I know that NMS revamped their product to remove botnet traffic, but why would that affect Android's numbers? Are Android botnets a widespread thing that nobody knows about, or can detect, except NetMarketShare for some fucking reason? It makes even less sense than the effect on Linux' numbers. And how have Gartner, StatCounter, and NetMarketShare all ended up with such apparently divergent results? Seriously, WTF is going on?

I think I want to see some serious critique of NMS's new product/methodology. Their new online tools look very slick and very powerful, but if their data collection and analytical methology is badly flawed, then their online tools may be useless, no matter how slick or powerful they seem.

NMS make major changes, and finally explain themselves.

I'm quite late posting today's OS market share numbers, in part because I honestly wasn't expecting much to change from last month. Well, was I ever wrong on that, because NetMarketShare have changed just about everything... and, more importantly, they've finally explained why:

Welcome to the new NetMarketShare.com

---

We completely rearchitected the product to provide more accurate and accessible data. Here's what's new:

Elimination of botnet data

---

As bot traffic across the web has risen dramatically, it has been a challenge to detect and remove it from our dataset. This is a critical issue since bots can cause significant skewing of data. In particular, we have seen situations where traffic from certain large countries is almost completely bot traffic. In other countries, ad fraudsters generate traffic that spoofs certain technologies in order to generate high-value clicks. Or, they heavily favor a particular browser or platform.

The primary focus of this release was to build detection methods to eliminate this traffic. We rewrote the entire collection and aggregation infrastucture to address this issue.

Please note: This dataset is separate from and replaces the legacy data. Subscribers that need legacy data please contact us for access.

The "bot traffic" explanation is... plausible, actually. Bots are a huge online problem, and for a business whose business is measuring online traffic, it makes sense that large-volume bot activity would seriously mess with their data, and their analyses.

For the tracking of OS usage numbers, of course, this basically renders every previously posted result (i.e. the "legacy data") completely irrelevant. Therefore, I will let it go, and start over fresh, much as NMS have themselves. It's important to keep some sense of the history here, through, so... let's see what NMS have on tap for history.

Annoyingly, the new table which accompanies this graph doesn't break out each month's numbers in a nice grid, but rather shows only the average for the period. Each month's numbers are only available as mouse-over tooltips, which is worse for at-a-glance comparisons but probably better for NMS's business model (which is, after all, to sell the numbers to subscribers). Also, the new NMS data doesn't go all the way back to Windows 10's launch, so we can only look at the range from May of 2016 through the end of November, 2017.

Even so, the botnet-less trend lines look rather different than the botnet-filled versions. Mousing over the data points on that graph, we can learn that Windows 7 currently holds 43.12% of the market, basically unchanged from last month's 43.05%, while Windows 10 holds 31.95% of the market, an uptick from last month's 28.86%. Interestingly, Windows 10's growth has come mostly at the expense of Windows XP, which declined from 7.64% to 5.73%.

Generic "Linux" only shows up as 1.65% of the market, down slightly from last month's 1.85%, although that doesn't include Ubuntu (which is also Linux) or Chrome OS (which is also Linux), etc. That kind of fragmentation is part of the problem that Linux has in attracting new users, though, so that seems fair enough, and at least NMS is now listing Chrome OS separately as a desktop OS, which will allow us to track its progress over time - potentially very interesting. Combining all the various versions of each OS together gives a better overall picture of how Linux-based OSes compare to Windows:

Exactly why botnets would be inflating Linux' OS market share numbers, rather than Windows', is something that I'd like to see explained in more detail at some point. Linux is famously more secure than Windows, generally, so it would make far more sense for botnet traffic to be inflating Windows' market share results, instead. I'll be keeping an eye out for a better explanation there.

In the meantime, though, we have a new, and hopefully more reliable, baseline for watching the progress of the OS market from this point on. It'll be interesting to see what happens. I'm still not expecting to see much movement for the next few months, for the reasons previously stated, but I wasn't expecting NMS to revamp their entire business in a month, either, so who the fuck knows? If nothing else, it should be interesting to see what happens.

Boldly predicting more of the same... whatever that means.

We're just a few days away from the latest OS market share numbers's release, so let's take a minute to discuss expecations.

For the last two months, NetMarketShare's initially posted results have shown a significant shift away from Windows, and towards Linux. Both times, NetMarketShare have blinked rather than standing by their own data collection and analysis, and re-weighted those numbers to hew closer to tech reporters' expectations. In both cases, though, their carefully targeted re-weighting has failed to make the shift go away entirely; they've masked the shift, but not eliminated it.

Looking at both the first-reported numbers and the later revised numbers together shows both trends:

The truth lies somewhere between these trend lines... but where?

So, what should we expect this month? More of the same, would be my guess... as you can probably tell from the fact that I'm tracking both 1st-posted and revised numbers, and plotting them against each other. We'll have to see just how big the reported shifts are, both as first reported, and then after NetMarketShare cave to the pressure and massage as much of the shift as possible out of their reported statistics.

Once could be an accident, twice could be a coincidence...

One of the more troubling recent trends in journalism is that of the pre-existing narrative frame. This occurs when reporters, who are expecting a thing to happen, and waiting to report on the thing as it happens, ignore evidence that contradicts that pre-formed narrative. They don't report what's actually happening, as it's happening; instead, they cherry-pick from the available stats and factoids, telling only the stories that they were already planning to tell.

We see this a lot in political reporting; false equivalencies, false dichotomies, and pre-conceived narratives that exclude important-but-contradictory facts are all rampant "in the beltline." It's one of the reasons why public confidence in mainstream news reporting is at historic lows; it's one of the reasons why political reporters and pundits have spent the last year constantly being blindsided by actual developments in politics. People were more comfortable getting their news from The Daily Show than from their daily newspaper because Jon Stewart didn't do this; Last Week Tonight wins Peabody awards because John Oliver doesn't do this.

The phenomenon isn't limited to political reporting, either. It happens in tech reporting, too. When MSPowerUser was reporting on this month's OS market share trends, they did it explicitly:

This month Netmarketshare’s data appears to have an error, claiming a 5% increase in Linux market share in one month, so I think it can be safely ignored.

Never mind that this was the 2nd month in which this shift was visible; never mind that Satya Nadella is already trying to get out ahead of the Shift, downplaying the importance of Windows to Microsoft. MSPowerUser's writer, like many tech media writers, had spent months waiting to tell the story in which Windows 10 finally surged in the marketplace to overtake Windows 7, and so he cherry-picked the stats which seemed to finally be telling that story, while explicitly ignoring stats that didn't fit.

ZDNet is doing it, too. Unable to completely ignore the undeniable trend in the data, they've instead taken issue with the source, NetMarketShare, in this recent piece:

There have been numerous stories that the Linux desktop has more than doubled from its usual 1.5 to 3 percent marketshare to 5 percent. These reports have been based on NetMarketShare's desktop operating system analysis, which showed Linux leaping from 2.5 percent in July, to almost 5 percent in September. But unfortunately for Linux fans, it's not true.

Neither does it appear to be Google's Chrome OS, which tends to be under-represented in NetMarketShare and StatCounter desktop operating system numbers, being counted as Linux. Mind you, that would be fair, since Chrome OS is based on Linux.

The real explanation is far more mundane. It seems to be merely a mistake. Vince Vizzaccaro, NetMarketShare's executive marketing share of marketing told me, "The Linux share being reported is not correct. We are aware of the issue and are currently looking into it."

Again: ZDNet, confronted with with data that didn't fit their existing narrative, went looking for a rationale that would let them ignore it, rather than reporting on it as it was happening... you know, the way reporters used to.

And the insidious thing is that NetMarketShare have been co-opted by this. Faced with skeptical reporters who refused to use actual facts as the jumping-off point for their reporting, NMS are hard at work, tweaking their statistical models so that their reports can more closely match reporters' expectations.

Déjà vu, all over again

So, just like last month, when NetMarketShare balked at reporting what their data was clearly telling them, they've once again massaged their stats and reposted them. And, as expected, the result is a slightly smaller swing than their earlier posted stats indicated.

The old version put Windows' overall OS market share at 82.2%, and "Other" (a.k.a. Linux) at 13.04%. The new version looks like this:

This version of the numbers puts Windows at 86.21% overall, with Windows 7 declining to 46.22% (-2.21%), and Windows 10 increasing to 28.65% (+0.66%). Linux, meanwhile, gained 2.15% to finish at 9.99% overall. That 86.21% number is especially interesting; it's almost exactly what NetMarketShare had originally reported as Windows' overall market position in their first version of last month's stats.

While this new analysis of their data shows a smaller swing from Windows to Linux, it still does show a swing from Windows to Linux, and a big one. Especially in light of Satya Nadella's recent remarks, downplaying the importance of Windows to Microsoft, it seems that the Linux Shift is actually a thing. We're not left wondering whether or not it's happening, only how quickly... and, of course, why now?

I fully expect that next month's stats will show a continuation of this same trend. Windows 10's growth is still sluggish; breathless reports that it would overtake Windows 7 in a matter of months are missing the point. Yes, Windows 7's user base appears to be changing operating systems, but they're not migrating to Windows 10 -- they're migrating away from Windows altogether, and trend which Microsoft are already trying to downplay the importance of. We could be months away from a market in which Windows 7, Windows 10, and Linux split the market more-or-less evenly between them.

What impact that might have on the world of desktop computing is impossible to say. Microsoft's Windows has been so dominant on desktops for so long that it's difficult to imagine what the world of desktop PCs will look like when that dominance comes to an end. One thing is becoming increasingly clear, though; Windows' dominance on the desktop is coming to an end, and sooner than anyone ever thought possible.

Once again, I will preserve both of my posts on this topic, in order to preserve NetMarketShare's originally-posted evidence of the nascent Linux Shift as it progresses. I have the feeling that this same sequence of events will play out again in November, and in December, and in January, and at some point, the limitations of regression analysis will simply not be able to mask the evidence of the Shift anymore. How Microsoft's supporters react on that day will be interesting to see.