Facebook announced Oculus Quest, and it's already obsolete... according to its designers!

Remember when Facebook won (and lost) a lawsuit partly waged over the way they poached John Carmack away from Zenimax/ID? I wonder if they're re-thinking that acquisition after Carmack compared their next-generation "all-in-one" Oculus device to last-generation gaming consoles?

For the record, here is how Facebook/Oculus described their new device during the actual announcement, as reported by Gizmodo yesterday:

“This is it,” Mark Zuckerberg said to a crowd of developers and press at Facebook’s annual VR developers conference, Oculus Connect. “This is the all-in-one VR experience that we have been waiting for. It’s wireless, its got hand presence, 6 degrees of freedom, and it runs Rift-quality experiences.”

And here is how Oculus' CTO described the Quest at the same conference, as reported by arstechnica:

In a wide-ranging and occasionally rambling unscripted talk at the Oculus Connect conference today, CTO John Carmack suggested the Oculus Quest headset was "in the neighborhood of power of an Xbox 360 or PS3."

That doesn't mean the Quest, which is powered by a Qualcomm Snapdragon 835 SoC, can generate VR scenes comparable to those seen in Xbox 360 or PS3 games, though. As Carmack pointed out, most games of that generation targeted a 1280x720 resolution at 30 frames per second. On Quest, the display target involves two 1280x1280 images per frame at 72fps. That's 8.5 times as many pixels per second, with additional high-end anti-aliasing effects needed for VR as well.

"It is not possible to take a game that was done at a high-quality level [on the Xbox 360 or PS3] and expect it to look good in VR," Carmack said.

So... it's wireless, but needs a four-camera room-scale setup to work, and it aims to provide a Rift-quality experience, but can't because it just doesn't pack enough processing power. Also, count on it, Quest will cost significantly more than the Go, if only because of those cameras... and Oculus Go isn't exactly flying off shelves. Why does this exist, again?

Windows 10 is "now" on 700 million devices

Compare this article, from TechRadar:

Microsoft is edging ever closer towards its target of having Windows 10 on over a billion devices across the globe, and the software giant has just confirmed that the OS is now present on over 700 million pieces of hardware.

to this article, also from HotHardware:

Microsoft just announced a new milestone for its Windows 10 operating system: it is now installed on over 700 million active devices worldwide. This is an increase of 100 million from the last time that Microsoft gave a status update back in late November 2017.

The two articles look almost identical, but there's a crucial difference: TechRadar's article was posted today, while HotHardware's article was posted back in June, about a month after TechCrunch posted this article, also citing statistics from Microsoft:

At its Build developer conference, Microsoft today announced that just under 700 million devices now run Windows 10. Almost exactly a year ago, that number stood at 500 million. In addition, the company also today noted that Office 365 now has 135 million monthly active commercial users, up from 120 million last October.

That's right, Microsoft have now "announced" exactly the same Windows 10 adoption milestone on three separate occasions, in three separate months. Which raises the question: Why does anyone still treat Microsoft's self-aggrandizing "statistics" as being even slightly accurate, reliable, or newsworthy?

More specifically, why is TechRadar posting about this? Why are they still talking about Windows 10 hitting the "magic billion" mark, when even Microsoft doesn't mention the embarrassing billion number anymore?

Of course, originally, Microsoft wanted to get Windows 10 on a billion devices in three years post-launch, or around mid-2018, so that didn’t happen (although it fairly quickly became clear that this was an unrealistic target).

Still, hitting 700 million in just over three years isn’t a world away from the magic billion, although Microsoft isn’t making much noise about this achievement, with it flying pretty much under the radar at Ignite.

That Microsoft is lying with statistics, yet again, isn't newsworthy. They do this with some regularity, massaging the definitions of different statistics to suit their needs of the moment, and releasing whatever numbers they please, as many times as they please, in an effort to keep up the appearance of momentum. They're hardly the only publicly-traded company to do this, either; most of them engage in some level of the practice (hint: the moment some company starts talking about their earnings per share, rather than their total revenue, start looking for an opportunity to sell your shares).

But why do tech bloggers, in particular, keep falling for it? In a sphere where the status quo changes constantly, and accurate, up-to-date information is essential, why are journalistic standards so damned low?

That's not merely a rhetorical question; I'd really like to know.

Hostility, demonstrated and quantified

Just in case you were thinking that my post last week about Microsoft's naked hostility towards Windows users was a touch on the hyperbolic side, today comes a great piece from the normally very MS-friendly Windows Central which illuminates the problem nicely:

Before Windows 10, a clean install of Windows only included the bare essentials a user would need to get started using their PC. That included software built by Microsoft, such as Mail, Paint, and its web browser, and it never included "bloatware" or "trialware" that one might find on hardware purchased from a third-party OEM that preloaded all kinds of crapware.

[...] With Windows 10, a clean install stays that way for about two minutes, because the second you hit the desktop, the Microsoft Store immediately starts trying to download third-party apps and games. And these apps keep trying to install themselves even after you cancel the downloads.

There are six such apps, which is six too many. These apps are often random, but right now they include things like Candy Crush, Spotify, and Disney Magic Kingdoms. You should not see any of these apps on a fresh install of Windows 10, yet they are there every single time.

[...] The only way I've found that gets rid of them permanently is to let them install initially, without canceling the download, and then uninstall the apps from the Start menu. If you cancel the initial download of the bloatware apps before they complete their first install, the Microsoft Store will just attempt to redownload them later and will keep doing so until that initial install is complete.

[...] For comparison, I recently did a clean install of macOS High Sierra on a Mac, and that experience was sublime. Once I hit the desktop, I didn't need to go in and start uninstalling extra bloatware that was automatically trying to install itself. All the apps that show up on the OS are Apple-made and not considered bloatware.

Writing as someone who recently spent most of a day doing a clean install of Windows 7, I can honestly say that this experience hold zero appeal for me; the fact that it will take less time to get to the point where the bloatware starts installing itself is simply not enough incentive.

Why not Windows 10?

Given that my recent foray into PC Gaming on Linux were... underwhelming, shall we say, some of my readers (yes, all three of you) may well be wondering why I don't just bite the bullet and switch to Windows 10 already. True, the "free upgrade" offer has ended, and Windows 10 will not cost money, but I was honestly always going to want more control over my PC than WX Home offered, which would always have meant a Professional license... in other words, Windows 10 was always going to cost me something up front.

So, why not just switch already?

It's a good question, and one which I've struggled a bit to answer myself this past week. Was I just being stubborn? Or did I still have concrete, valid reasons for sticking with Windows 7, while hoping that Valve and Steam Play would be able to solve the Linux gaming performance problem at some point during the upcoming year?

Today, though, I had a eureka moment, when I found my nebulous reasons for sticking with W7 suddenly crystallized into a single paragraph by Paul Thurrott. He was writing about Microsoft's decision to de-escalate some of WX's advertising bullshit. After downplaying the reversal as "not much of a win," he goes onto describe  Microsoft's approach to Windows consumers thusly:

Everything else that is still wrong about Windows 10 is still in the product and will move forward with version 1809.

In other words, the slippery slope I first warned about way back in 2012, when Microsoft quietly began its first sneaky advertising additions to Windows 10's predecessor, is still very much an issue. And has escalated over time. The ongoing and unnecessary compromises to Windows 10---rampant advertising, attempts at pushing users to Microsoft Edge, pre-bundled crapware, and more---continue unabated. There is no major software platform that is this hostile to its own users. [Emphasis added.]

"That's it!" I thought. "That's the reason!" It's not just WX's ongoing issues, which I could probably live with or work around, it's that Microsoft treats all but their largest Enterprise customers with thinly veiled contempt, most of the time, and has done so for years. Sure, maybe this latest half-hearted walk-back marks the beginning of a trend towards less bullshit in Windows... but I'm not planning to bet money on it.

Seriously, fuck Microsoft at this point. If Valve can get Steam Play working well enough to provide a moderately decent gaming experience on Linux (or, hell, SteamOS), I'll put up with some performance issues, rather than give Microsoft the satisfaction.

The opposite of hype is not necessarily truth

This probably sounds really weird coming from someone like me, who loudly proclaims his opposition to the cynical workings of the average corporate PR hype machine, but there's a time and a place to stop pissing on companies that are doing nothing wrong.

It seems to be fashionable lately to hate on Valve, who run the wildly successful Steam platform, and I really don't understand some of the pure manure that's passing as for journalism these days where Steam is concerned. Case in point, this piece from Kotaku about the release of Negligee: Love Stories:

Slowly but surely, Valve seems to be letting uncensored adult games onto Steam at last, starting with Negligee, which came out today. The catch is that it’s only available in some regions, and it remains banned in a globe-spanning majority of others. In a thread on Steam, developer Dharker responded to prospective players’ confusion by explaining where and why Negligee remains unavailable.

Because open marketplaces are apparently a bad thing now. Because reasons.

They're not going to make it...

Since I haven't been watching the monthly market share numbers as intently as I used to, I really didn't have the correct context in which to place this bit of news when it was first posted last week:

With the Windows 7 end-of-support clock slowly winding down to January 14, 2020, Microsoft is announcing it will offer, for a fee, continuing security updates for the product through January 2023. This isn't the first time Microsoft has done this for a version of Windows, but it may be the first time it has been so public about its plans to do so.

Windows 7 still has a large share of the overall Windows market, especially among business customers. Moving off older versions of Windows is a slow process, even with advance planning, for companies with multiple thousands of Windows desktop machines.

The paid Windows 7 Extended Security Updates (ESUs) will be sold on a per-device basis, with the price increasing each year. These ESUs will be available to any Windows 7 Professional and Windows 7 Enterprise users with volume-licensing agreements, and those with Windows Software Assurance and/or Windows 10 Enterprise or Education subscriptions will get a discount. Office 365 ProPlus will continue to work on devices with Windows 7 Extended Security Updates through January 2023.

[...]

This time around, the ESU program is being run out of Microsoft's Volume Licensing Unit and Core Windows Engineering "is producing these updates like a product," [Jared Spataro, corporate vice president of Microsoft 365] explained.

"We want to encourage people to get off Windows 7, but we want to make it more than something punitive," he said.

This was the second move that Microsoft made in a direction of accommodating the "slow process" of operating system migration at scale; they also announced that they're slowing the pace of forced Windows 10 updates, which was another pain point for Enterprise customers.

Some businesses have complained that they need more time and flexibility to update Windows 10, and IT admins are tasked with ensure apps work with the latest update. Microsoft is releasing new cloud-based tools to ease app compatibility testing, and the company is also giving IT admins more time to update. All currently supported feature updates of Windows 10 Enterprise and Education editions will be supported for 30 months from their current release. The existing policy is 18 months, so this bump brings support closer to what IT admins were used to in the Windows 7 and earlier days.

Interesting, yes? Now for the context: Windows 7 (W7) doesn't just have a large share of the market. W7 still has over forty percent of the market, according to the latest numbers from NetMarketShare.

Virtual Reality Check?

I've been writing, more or less non-stop, about VR's failure to thrive, and the reasons behind it. In the process, I've posted links to the likes of CinemaBlend, reporting on how VR's numbers show that VR is in trouble, and TechCrunch, reporting the very same thing.

But CinemaBlend is a self-described entertainment news and celebrity gossip site, and TechCrunch covers start-ups and technology exclusively. Neither one has the cachet of an a major print publication like, say, Forbes. So it's noteworthy CinemaBlend and TechCrunch have now been joined in calling out the VR Emperor's New Clothes by none other than Forbes, under the headline, "Virtual Reality: Steep Decline Is More Than A Hiccup."

There is little doubt that the market for virtual reality has entered the trough of disillusionment and it will take real progress on the limitations of the technology before it comes out again.

IDC’s latest figures (see here) make rough reading and it is likely that even the one segment that is showing growth will soon run out of steam. VR exists in three categories: screenless viewers that use a mobile phone, headsets tethered to a remote computer or console and standalone headsets where the compute is embedded in the unit.

Both the screenless viewers (-59.1% YoY Q2 18) and the tethered headsets (-37.3% YoY Q2 18) are rapidly declining with only the very new standalone segment showing any growth (+417.7% YoY). However, because the standalone segment is so small (212,000 units) it has been unable to offset the impact from falling demand elsewhere, leading to an overall market that declined 33.7% YoY.

Hence, once the initial demand for the standalone units is satisfied, it too will see a similar profile to the one being suffered by the tethered units. This is because the main issues of VR are not close to being solved.

Predictably, Forbes then go on to list price (too high), clunkiness (they're uncomfortable to wear, and can "make the user feel foolish when wearing one"), and comfort and security (lack of environmental awareness, simulation sickness). Forbes, like so many others, have continued to miss the elephant in the room: the simple fact that VR still doesn't enable any activity that can't also be done without the tech, and that people will want to do.

Failing faster with Steam Play

About a week ago, I switched from Windows 7 to Ubuntu 18.04.1 LTS Bionic Beaver, expressly to try out the new Proton technology that Valve was adding to Steam Play. In theory, Proton should allow Linux users to install and run almost any Windows game in Linux, seamlessly.

In practice, though... well, let's just say that it's a beta. Quite literally, actually. And, after a week of struggling with Linux as my daily driver OS, and trying to game through Steam Play, I'm retreating back to Windows 7 for the time being.

I'll let Jason Evangelho from Forbes explain why:

While the technical possibility of playing the thousands upon thousands of Windows-only Steam games on Linux is a revolutionary leap forward, it's far from perfect. When it works it's amazing. When it doesn't it's an exercise in frustration.

I've spent the past two weeks downloading dozens of games to test, and many of the ones Valve has whitelisted do indeed perform well without any hiccups. But some, like 2016's DOOM, simply won't launch. The game works for the majority of people who've submitted their experience to the Steam Play Compatibility Reports website, but for many others, it does not.

Valve can't whitelist a game and take Steam Play out of beta (it's now available to everyone using the Steam for Linux client) until every title they've put their seal of approval on works for everyone who launches it -- provided they meet the software and hardware requirements. I think the worst possible outcome is to see reports of gamers who've wiped out their Windows partition and parked themselves gleefully in the Linux camp, only to be frustrated that the experience they expected isn't happening.

Emphasis added, because that last sentence describes me perfectly.