Ask Paul: July 3 ⭐

Happy Friday! And Happy 4th of July if you’re celebrating. Let’s kick off the holiday weekend with some great reader questions.

“Malware found spreading through sponsored ad on X”

At least it’s true to the platform it’s on

📋 Getting Winget

Alex Strickland asks:

I’ve been catching up on past episodes of Hands On Windows and loved the episode on the Winget UI. I regularly use Winget to install one-off apps or to quickly rebuild my setup after a clean Windows installation. I’m curious: how do you personally use Winget? Do you rely on a custom PowerShell script to automate the commands, or do you have a different workflow you prefer?

First, a convoluted story demonstrating one way I use winget (the Windows Package Manager). Because, in an interesting coincidence, I started typing my response to this question, and having just switched late yesterday to the next review laptop and still being used to the keyboard on the previous laptop, I inadvertently tapped the Copilot key. Copilot didn’t open because I’d uninstalled it. But the Settings app did appear, displaying the Bluetooth & devices > Keyboard page because this key wasn’t assigned to do anything. This told me that I’d not yet installed PowerToys (because I use its Keyboard Manager utility to remap that key), had installed PowerToys but not yet configured Keyboard Manager, or that PowerToys just wasn’t running for some reason.

To find out which it was, I tapped the Windows key and typed power. And PowerToys did not come up in the results, meaning I had not yet installed it. So I opened a Terminal window and typed this from memory:

winget install Microsoft.PowerToys –silent

That installed PowerToys, of course. And so I ran the app, enabled Keyboard Manager, and configured it so that will never happen again. Is this a good/efficient way to install an app? Yeah, it is. Especially if the goal is to minimize or eliminate your use of a Microsoft account (MSA) and apps and services, like the Microsoft Store, that rely on that account. I wrote about this most recently in Switcher 2026: Minimizing the Microsoft in Windows 11 ⭐. And that represents the way I’ve been doing things for the past month or two.

This maybe isn’t the most typical way to use winget, which exists for automation purposes. To that point, I’ve also written about winget a lot, and about the ways I’ve used it. And yes, the primary use case has been to batch automate the installation of all the apps I use in a single run using a script. When I wrote Improving My Winget Bulk App Installer Script (Premium) over two years ago, I was leaning into getting as many apps as possible from the Store, as winget supports both Store and web repositories (sources) for the apps it can install. And at that time, most Store apps took advantage of the Store’s auto-update capabilities and other perks.

But things have changed a bit since then. Thanks to Microsoft’s ever-looser Store policies–aimed at getting more apps in there–more and more of the apps you install from the Store just do their own thing. You can see this by opening the Store and checking the Downloads view for updates. Some apps have to be updated manually. Also, tied to the above article, I’ve been seeing what life looks like when you ignore MSA entirely or partially–I alternate between both configurations across multiple PCs–and so I adapted my script to only download apps from the winget (web) repository.

There are signs that this will change in the future, but for now, winget can’t auto-update all the apps on your PC. So if you’re going to use winget, you will actually have to use winget, which means opening a Terminal window from time to time and checking for updates. Unlike many people, I guess, I actually have an admin Terminal window open all the time for various reasons, the key one now being that I use Git via the command line to upload book changes to Leanpub so I can preview them in PDF form. But every few days I’ll just see how many updates are pending:

winget update

And then install them all (well, most of them) using

winget update –silent –all

As with the Store, there are some apps that can only be updated through those apps. Like Microsoft Edge and Android Studio. And there are some apps you have to update individually, like Discord, for some reason.

winget update Discord.Discord –silent

To me, this has become a habit. It’s just something I do fairly regularly.

So I guess I use it much like you do, though I hope for a future update to winget that will let us automate app updates as well. I suspect you could do that right now using whatever AI, since this is all command line-based. But I kind of just want to remain fluent with the command line in general, so I like just having it there to interact with from time-to-time.

“Somebody told DeepSeek to build in-browser ransomware and it gleefully complied”

Please stop anthropomorphizing AI. It did this maliciously, not gleefully

💼 And never the twain shall meet

OldITPro2000 asks:

The Information has reporting out today on an internal Microsoft memo from new Copilot boss Jacob Andreou. His plan is to merge consumer Copilot and enterprise Copilot into one app with additional features added. This is expected in the August timeframe.

Yeah, I saw that. (But did not read it because The Information is extortion-expensive.) This does map to a rumor Microsoft confirmed during the Build 2026 keynote that it was working on a “Copilot super app,” which was a weird mention since it was clearly not ready, there was not even a single prototype image or screenshot, and Satya Nadella really stumbled over his words when he started into this bit. I have to assume they wanted to show something at the show, but it just wasn’t ready. But the key takeaways are that he said it would happen “this summer” (presumably in some pre-release form) and that it is about bringing “coding to all knowledge work within one app,” with chat, Cowork and Copilot intermingled.

Quite honestly I wasn’t aware there even was a consumer Copilot “app” per se beyond the Web application, but what I found most shocking was Andreou mentioning in the memo that Satya Nadella believes for Microsoft to really excel over the next 10 years it will have to be excellent in a consumer-focused way.

There have been two main Copilot apps for at least a few years, Copilot (for consumers) and Microsoft 365 Copilot (for businesses), and either (and sometimes both) are preinstalled in Windows 11; these days, most business-class PCs (or perhaps it’s as simple as PCs with Windows 11 Pro) come with Microsoft 365 Copilot, while consumer PCs (and/or those with Home) come with Copilot. There are also mobile clients for both.

Nadella’s belief that Microsoft needs a consumer presence is perhaps fanciful, but this has been true of Microsoft’s leadership ever since the company handed that part of the market over to Amazon, Apple, Google, and others. This is a bit tricky, though, since Microsoft has two major audiences within the consumer base: Individuals who use Windows 11 and/or Microsoft 365 on PCs and gamers in the Xbox ecosystem. The former audience is big (1+ billion people) but completely checked out and not engaged in the slightest. And the latter is enthusiast-driven, an audience that is very much engaged but also enduring a period of chaos, fear, and uncertainty about the future. As important, Xbox is the brand for gaming, and that makes sense since Microsoft is the equivalent of Oldsmobile to that audience, a dinosaur and a boring one at that.

That Microsoft could have–or did have–two sides, so to speak, two major audiences that span consumers and businesses, makes sense. There are great arguments, and many have been making them for decades, that the same people managing Microsoft environments at work during the day were still consumers out in the world and that they might choose Microsoft consumer offerings through some kind of halo effect or at least inertia. And there has always been some of that. When I was at Windows IT Pro and print magazines were still an ongoing concern, we created a publication called Connected Home to go after that audience. But it’s shrunk over time, in part because the ecosystems elsewhere are so strong and in part because Microsoft slowly killed a succession of consumer products and services that are not Windows or Office.

At one point, the split between businesses and consumers at Microsoft was 65/35, meaning that two-thirds of its revenues came from businesses (and other organizations like governments and schools). But I suspect it’s tilted much further toward businesses today, even after the Activision acquisition. And businesses provide the kind of steady revenues that Microsoft (and every company) wants, meaning subscription-based. By the time consumers became (perhaps too) comfortable juggling tons of subscriptions, Microsoft’s business was such that only Microsoft 365 and Office (Microsoft 365) made sense in that world. There’s no video, music, or smart home subscriptions, no shopping services, no anything that Microsoft could make now that anyone would want.

I would argue and I think most of us who read Thurrott.com would agree that Microsoft can’t win for losing when it comes to any of their past consumer initiatives and is generally indifferent when it comes to ordinary consumers, so reading that was quite a surprise.

This is one of those things where you can only play the hand you’re dealt (which comes up again below). It’s a bit too easy to make an argument that Microsoft made all these mistakes over time and that it dug a hole it can’t get out of now. But Microsoft isn’t a single entity moving in concert. And as time goes by, the Microsoft of today is not the same company as it was 10, 20, or 30 years ago thanks to whatever market dynamics, failed and successful product strategies, employees retiring, being laid off, or otherwise leaving, and whatever else.

Here’s an oddball example of what I mean. There is a band called Foreigner. In the late 1970s and into the late 1980s, this band seemed like a band; it consisted of some number of members, with a singer, songwriters, and all the rest. Over time, members were fired or left, and in recent years, we’ve seen versions of this band with just a single original member or member who was with the band in its active prime. Today, Foreigner is a brand owned by one of its founders, and the band that tours and plays live consists solely of new people who never recorded a song or album under the name Foreigner. It’s a cover band, but more simply, it bears no resemblance at all to the original band.

That’s what Microsoft is (as are most long-lived companies). Those of us who have been around for a long time still have this deep-seated understanding of what this thing is, but I bet it’s out-of-date in many cases. And rectifying what we think Microsoft is with what it really is can be painful. This is an ongoing issue for me, for sure.

Tied to this is what I wrote above, that even Microsoft’s leadership, its CEO, maybe can’t differentiate between the Microsoft that is and the Microsoft ideal they have. Consumers do not care about Microsoft, full stop. Nothing will change that. But the dream remains, despite institutional memory and reality. The world has moved on, and in the same way that no consumer looks to Microsoft for music, videos, or anything else, no consumer is looking to Microsoft for AI. With the small asterisk that AI is not a feature, AI makes sense for many productivity tasks, and AI does make sense in Windows and Microsoft 365. But no one is paying extra for it.

*I’m curious to know if you’ve seen or read about this memo and what your thoughts are. Happy 4th and I won’t look at the top of your head on Windows Weekly.

🙂 Thanks, but I think it’s safe to look at now.

“Steam Machine user reports GPU failure and red line of death”

You guys are copying the wrong console

🦊 Crazy, but like a fox

train_wreck asks:

I’ll preface this with “I’m just a lowly Nintendo gamer unfamiliar with the internals of Xbox”: what was the point of Microsoft buying a bunch of game studios only to close them not long afterwards? There’s talk online that much of that was “Phil’s Folly”.

That’s unfair, but it’s also an inevitable byproduct of this kind of change. If you start kicking the woodpile, you’re going to find out what’s underneath. And what’s underneath are critics and Monday morning quarterbacks.

What Phil Spencer did, generally speaking, isn’t just defensible; it was arguably the only way forward if you assume that “the way forward” means Xbox staying as part of Microsoft. He assumed leadership of Xbox in the wake of the Xbox One launch disaster, so let’s not forget those were the cards he was dealt. He didn’t cause it and didn’t play any role in it, but he had to clean up that mess.

Surveying his world, he saw three things. Competitors were being successful with exclusives. Hardware that was not profitable and never would be. And a need to reassure the base by focusing on games, not entertainment generally. (Sound familiar?) What he did about this is provably correct. He orchestrated studio acquisitions, starting with Mojang and then going bigger with Bethesda, to ensure high quality games for Xbox. He revved the Xbox One to reduce costs and lower prices for consumers, first by removing the silly Kinect requirement, and then expanded the range with lower-cost (Xbox One S) and higher-performance (Xbox One X) options. And he dropped the entertainment focus, which had resulted in ludicrous and too-late offerings like Xbox Music and Xbox Video in Windows 8.

This was the equivalent of Steve Jobs getting the iMac out the door for Apple. Not enough to turn around the company and drive explosive growth, but enough to right the ship, get it back on somewhat steady footing. But the Xbox One disaster lingered. Hardware sales kept declining, costs kept rising, and it’s not like it made any financial sense to kill cross-platform titles when its console business was a distant third. It needed those sales. And there was the problem of making Xbox make sense inside Microsoft. The Xbox One was first released in late 2013, and Satya Nadella became CEO in early 2014. And two of the more infamous stories about those early days are relevant here. Nadella required each business in Microsoft to either be profitable or have a plan to get there. And he required them to make sense within a Microsoft that was laser-focused on cloud computing.

Spencer’s solution was Game Pass, a monthly subscription that mimicked the way Microsoft earned revenues from its business customers, and a cloud streaming service, now called Cloud Gaming, that would work as a proof point for the efficacy of Microsoft’s Azure datacenters at scale. This pitch worked. Game Pass got off to a great start, quickly racked up tens of millions of customers at $10 per month, and it was one of those “no brainer” offerings I often mention, both for consumers and for developers with catalog titles not earning money anymore.

Hardware sales kept tanking. The pandemic happened as Microsoft readied the next-generation consoles, Xbox Series X and S, and was ongoing when the products launched. Game Pass subscriptions slowed, so Microsoft kept improving the offerings by adding more games and new perks, like day one access to all Microsoft Studio games. But nothing worked. The Xbox Series X and S are excellent devices, but they sold even more poorly than their predecessors. Game Pass subscriptions kept slowing and then tapered off. There was zero change it would ever be profitable in hardware and the plateau for Game Pass was lower than expected. Now what?

Microsoft could acquire Activision Blizzard and transform Xbox into a game publisher, the second biggest on earth. A business that makes far more sense than selling hardware, and only makes sense when your games are available on as many platforms as possible. This, too, was in keeping with the overall direction at Microsoft, which is often expressed as “meeting your customers where they are,” a strategy that previously turned Microsoft 365 into the company’s biggest single business. Xbox fans hated it. But they were wrong: For Xbox to survive, this shift was–and still is–necessary. Spencer did the right thing.

But he’s not a magician. Physics exists, and subsuming an organization as big and geographically as dispersed as Activision Blizzard is incredibly expensive, far beyond the $69 billion paid for the company, and an expense that continues forward in a quarterly operational costs category. With Activision Blizzard, Xbox is much bigger as a business–revenue-wise, headcount-wise, and whatever else–and much more complex. And the hardware remained a boat anchor on profits, of which there were none. If they just got rid of the hardware, this would be a nicely profitable business. But just like Nadella and his customers mistaking their perception of this company with the reality of this company, they can’t seem to let it go. So here we are.

Phil Spencer tried to launch a first-party Xbox gaming handheld but was shot down by the Microsoft board of directors and senior leadership team because there was no history of hardware profitability and no projection claiming this device would get there. So it went to a third party. Here, I have to guess a bit, but given the state of hardware at Xbox, transitioning Xbox as a platform more overtly to Windows made sense, and still makes sense, and letting third parties build all Xbox hardware should still be on the table. I believe that was his plan. But now he’s gone. Spencer decided to retire, and I suspect the toll of all these things happening to the business, issues he could never fully resolve, played the major role in that.

So, yeah. Let’s tear down the guy who kept this business going, only cared about games, gaming, and gamers, and did everything he could to make sure this thing we love continues forward. Let’s give air time to haters and those nostalgic for a past that, sorry, was not as awesome as they remember: The apex for Xbox was the Xbox 360, a product that came in third place out of three in console sales, was never profitable, and suffered from a years-long Red Ring of Death reliability problem and recall that cost the company billions. That was the high point, what these people want Xbox to return to.

Per the previous question/answer, the Microsoft of today is not the Microsoft of 2001 that launched the first Xbox, it’s not the Microsoft of 2005 that launched the Xbox 360, and it’s not the Microsoft of 2013 that launched the Xbox One, thank God. It’s the collective accumulation of everything that happened before, during, and since those milestones. Some problems were inflicted from within, some from without, and some were just happenstance. But we can’t change the past, we can just move forward with the situation as it is. Good luck to Asha Sharma and that team. They need it, because the fundamentals have not changed. They can make happy noises about consoles and Xbox as the brand all they want, but reality is knocking, and it does not have good news for this business.

We love to point fingers, assign blame, and hold onto grudges. But none of this was Phil Spencer’s fault. It was not even Don Mattrick’s fault. It’s no one person’s fault. That’s the nature of defeat. It’s rarely one thing. It’s often many things. We can understand what happened, and we can even point to individual decisions that were mistakes. Were any malicious or categorically stupid? Or were people just doing what they thought was right and/or was imposed on them from above or by circumstance?

I just don’t see malfeasance here. I see chance and some human error. And as a fan of the platform and of gaming in general, I see a business that I’d like to see survive and succeed. I hope either or both is possible.

“Gamer trades in $1,000 of physical discs at GameStop, days after Sony announces end of disc era”

Getting out while the getting’s good

📺 YouTube is also made of tubes

Bitrot asks:

Hi Paul, you wrote a while back that You Tube was a must keep subscription service and i totally agree. I was just wondering who are some of the people or channels you follow that are not political or tech related?

I don’t subscribe to any channel that could be described as political. I do subscribe to several that are tech-related, of course.

Ignoring those topics, most of my feed is algorithm-driven and based on what I’ve watched. So I will go through these repeated experiences where I watch a video on whatever topic and then see videos related to that for some period of time. If these interest me, I will watch one or more of these recommended videos, the algorithms are happy, and round and round we go.

When I bring up YouTube on a PC, iPad, or the Apple TV, I typically start with the choices in that top “Recommendations” row unless I (or my wife and I) have something specific we want to watch. Looking at this now, I see videos about software development, music, horror movies, and CDs/DVDs/physical media, plus those tied to subscriptions. Oddly enough, this came up on Windows Weekly this past week, when both Leo and Richard agreed that YouTube would be the last service standing, so to speak, and that, conversely, our viewing history/habits are perhaps the most embarrassing reflection our where our heads are at. Like if someone walked into my living room, powered up the Apple TV, and looked at my YouTube recommendations, they’d be like, what the hell is wrong with you?

And that is a thing. My wife is away this week, and I’m using this time to get caught up on the Windows 11 Field Guide updates, which is proceeding nicely. But this also means that I can do things a bit differently, especially at night. So instead of putting an audiobook on a 60-minute timer when I go to bed and wearing sleep earbuds, and I just do it through the iPad or whatever. And instead of watching something we mutually agree on at night, which can be via YouTube and other sources, I can just throw on whatever. I’m working at night in front of the TV, and sometimes the algorithm just takes over. It’s not always the type of thing you want to broadcast to others.

In my case this week, that means I’ve watched a bizarre series of Corey Feldman videos that I believe came out of a horror movie thing that turned into a Lost Boys thing that turned into a “Corey Feldman may literally be the worst human being on earth” thing. And there are weird crossovers with my normal viewing and subscriptions. For example, I subscribe to The Magnificent Others with Billy Corgan, which I like despite many episodes seeming to be a contest between the host and guest to see who can talk more, and he interviewed Corey Feldman at some point. (And was, if anything, too deferential to this guy’s nonsense.)

So yeah, Corey Feldman. Jesus.

From a volume perspective, my wife and I watch two YouTube channels regularly, and both are tied to true crime: 48 Hours, which is a TV show, and That Chapter, which is young Irish guy we really like. (He has a podcast too, but that’s not as good.) Semi-tied to this, it’s worth looking up Bill Hader’s impersonations of Dateline host Keith Morrison. Absolutely classic.

I subscribe to comedy channels like Space Ice (movie reviews, Steven Segal commentary), Rifftrax (ex-MST3Kers), Ringer Movies (video version of the podcast The Rewatchables, basically), and SNL.

For music, I subscribe to Professor of Rock and Rick Beato, both of which get far too preachy and click-baity but can be entertaining and educational, plus The Roundtable with Drew (interviews), and The Hindley Street Country Club and Missioned Souls, both of which are astonishingly good cover bands. And I subscribe to several (or more) official channels of bands and musical artists. (We listen to YouTube Music through YouTube on Apple TV on many music nights, too.)

There are a few random travel/food channels I watch less often, like Mark Wiens, Baguette Bound, Rick Bayless, Almost Retired in Mexico, Rick Steves Travel Talks, and Rick Steves’ Europe. And I really like World According to Briggs, which is basically travel/demographics. He has a great voice.

Anyway. That’s most of the non-tech stuff. But I subscribe to more tech channels than anything else.

“Meta locks essential Ray-Bans feature behind a paywall, even though it’s on the glasses not the cloud”

Meta can’t not be Meta. Meaning, terrible and pointless