Ask Paul: June 27 (Premium)

Happy Friday! Let’s kick off the final weekend of June a bit early with another massive Ask Paul and some great reader questions.

? Tariffs, price hikes, and tech

wright_is asks:

We are looking at replacing a lot of our technology this year. In the US, we rushed through purchases for new infrastructure equipment, which we won’t actually configure and install for months, because the tariffpocalypse. Are you noticing price rises, or is it still pretty much status quo at the moment, when it comes to tech?

Coincidentally, I just happened on a Wirecutter article on this topic and looked specifically to see if there were any tech products in there. There are only a few: Bose QuietComfort Ultra Headphones are up 4.66 percent, for example. Granted, it’s a sample of products, not a complete list. But overall, 25 percent of the products tracked were pricier and roughly two-thirds were the same price as before.

Without getting into the stupidity and flip-flopping nature of our government, it seems like we went through a period of uncertainty before figuring out that they will always back off when it comes to the biggest fights. And so we’ve sort of collectively calmed down. Apple isn’t going to charge anyone $2000 for an iPhone that had cost $1000 a year earlier. Or whatever.

Where the tariffs seem to hit hardest here in the U.S. are on the constituents that did vote in these bozos, the poor and uneducated middle and lower-class Americans who suddenly find their grocery costs have gone up dramatically. (Oddly, given all the craziness in the world, gas prices are low, at least.) But so far, it seems like tech–with a few exceptions–has sailed through this. So far.

Uncertainty is the new normal, of course. So this could change any day. And then change again the next.

? Hot or not?

iantrem asks:

While AI isn’t exactly maturing (far too early for that), there are definite winners and losers emerging. Do you think different AI products might become “best in class” for particular areas.

This is what we see right now, and given the fast-moving nature of AI, it’s perhaps not surprising that the best AI for whatever workflow seems to change by the week. My expectation is that these things will eventually be roughly on par with each other, at least from the major players, and that the smaller AIs will either be good enough for many, broadly, or will be so specific to some task–like grammar/spellchecking and general writing help–that they will likewise be very good. This is sort of a “if everything is AI then nothing is AI” view, I guess, but it also speaks to the competitive issues Microsoft faces going forward.

I don’t think I’ve written about this at length per se, but it’s come up on Windows Weekly a few times. With Microsoft, Google, and others racing to establish their respective AI empires, we’re seeing an interesting dynamic that isn’t very typical in such a competitive market: They’re all agreeing on interoperability standards. My belief is that this is because these companies realize that time is of the essence and, that to move quickly, they can’t get bogged down in their usual anticompetitive behaviors. But this means that Copilot will not have many inherent technical or functional benefits over ChatGPT or whatever else if you want to use it with Microsoft 365. (There will always be some advantages, policy management of course, but licensing and deployment simplicity will likely be the biggest ones.) And that’s true across the board. With the possible exception of Apple and its walled garden on iPhone and other devices, customers will be able to freely mix and match AIs with whatever productivity (or other) solutions they use, and do that at the system level with deep integration.

To me, this all points to a single destination: Rough parity for all major AIs. And so the fight will switch to convenience, cost, and trust. And the problem for Google and Microsoft, in particular, is that the big brand today–meaning popularity, mind share, and whatever else–is ChatGPT. Not Gemini or Copilot. Whether it is (or remains) best in class is almost beside the point, though it will be something that we track and try to understand. For most people, it’s just what AI is and it doesn’t matter.

For instance, With it’s links into GitHub (so that’s why they bought it), Co-Pilot might be the go-to AI for developers and Microsoft also seems to be pushing into education given announcements this week.

GitHub Copilot does seem like the most secure of Microsoft’s AIs from a usage/mind share perspective, thanks mostly to the success of Visual Studio Code. That’s a relatively small market but it’s an important market, too. And it’s also fair to point out that the AI editor Cursor has seen surprisingly strong uptake, so I guess anything could happen. But this may be one of those convenience/cost things, too: If you use Visual Studio or Visual Studio Code, it may be best (or easiest) to just use GitHub Copilot, just as those using Android Studio may lean towards Gemini and those using Xcode will use whatever that is.

This is the hope for Copilot generally, that Microsoft 365 is so ubiquitous that it will drive Copilot usage too. But here, too, we’re seeing that the “better” brand appears to be winning out and overcoming the power of the default, as Chrome did in the web browser market.

Open AI seems to be leading the race to be the general “office use”/”client use” AI assistant despite Co-Pilot being everywhere in Office.

Yes. ChatGPT is that unique and rare thing where it has incredible mind share penetration with mainstream users. It’s why I referred to it as the Kleenex of AI, it’s the one everyone knows about and maybe even thinks they’re using, even when they’re using something else. When you look at OpenAI, its stiff leadership, and the weird, minimalist communications on its website, there is nothing there to suggest that anything it makes would ever resonate with consumers and businesses alike. But that is exactly what happened. I cannot explain this. All I can do is acknowledge it.

(It is likewise rather amazing how poor the perception of Google’s AI work continues to be with the public. This is an astonishing turnaround, given the billions of people who use and rely on multiple Google services every single day. I’m not sure if I ever saw anything like this, certainly not at this scale.)

And, given the Apple ad I saw this week (surely a parody), Apple seem to be after the office staff member that everyone hates who manages to keep their job despite all they do. Maybe they just want the “mobile AI” crowd?

Apple’s market is pretty solid: It’s huge, it’s loyal, and it’s affluent (at least from an aspirational perspective), and I think what they’re doing will work out in the long run. If you look back at I Will Not Pay for AI (Premium), my point was that AI is not a product, a thing to pay for, or a standalone service to get, it’s a feature of the things I do pay for and use. And that’s Apple’s approach. You use an iPhone, and AI will be there. You use Messages, and AI will be there. And you use Apple Music, or whatever, and AI will be there too. And in each, it will make sense for that solution. It’s an ingredient. It’s a value add for the product and the entire ecosystem, part of the many advantages of being there.

Or, if you’re Open AI, it’s a product. One with 500 million monthly active users that many pay for on an ongoing monthly basis. It really is astonishing.

? The Xbox future’s too bright to cast shade

iantrem also asks:

I really enjoyed the extended FRD episodes last week, it’s obviously a format that works when there’s more to talk about but I still think “daily” (when available) is better so you can react to the news as it happens.

Yeah, this was not planned, and it really can’t be. It just sort of happened. Oddly, the best conversation we had, perhaps, was one that wasn’t recorded properly and so we weren’t able to publish it. But if you think of FRD as two friends with many shared interests just chatting, it can be serious sometimes and frivolous other times. Sometimes these things take on a life of their own.

Anyway, the future of Xbox. You seem to be suggesting that the next generation of Xbox will see 3rd party companies allowed to produce Xbox PCs (or other devices) alongside an official Xbox console.

That’s my supposition based on the evidence, which includes leaks, rumors, and public statements. Here are the facts: Microsoft has added a new type of Xbox hardware device, called a gaming handheld, that sits iPad-like between the PC and the console. There will be first- and third-party Xbox gaming handhelds. This platform isn’t just based on the PC, it is a PC, but it’s running a stripped-down and optimized Windows with customized Xbox app and Game Bar experiences. AMD is handling the silicon, and Microsoft is working with AMD on future Xbox silicon across its devices. Microsoft loses money on Xbox consoles, and if it could just get rid of that without losing any customers, Xbox/Microsoft Gaming would be a financial juggernaut. Microsoft licenses Windows to PC makers.

Here is the educated guess: Microsoft could do the same for next-gen Xbox consoles, which are really just PCs, like the gaming handhelds. And in doing so, it could make Xbox not just make sense but be a winner, financially and from an audience size perspective. Xbox gaming handhelds are not going to turn Xbox around. Xbox consoles that are really just PCs and can have third-party hardware makers bringing improved scalability, pricing, quality, diversity, and whatever else could.

While this might work with devices like the Ally handheld, will other companies want to produce a PC/Console? The Xbox is a loss-leader and Microsoft makes it’s money from the software licences? How will 3rd parties make their money unless they sell move expensive hardware (that won’t sell because people will buy the cheaper Xbox). As for the Xbox PCs, will this work in a similar way to Windows Media Centre back in the day, an Xbox layer over regular Windows? It worked for the few people who liked it, but will it fail in the same way where people didn’t buy a PC to plug into their TV?

Yes. We have established players like Lenovo, HP, Dell, Acer, and ASUS that have found a way to make the low-margin PC business profitable. And one of the key elements of that is gaming PCs. Each makes a range of those devices, and a coming generation of “Xbox PCs” could be the entry level part of that market, with the more premium gaming PC lines they already make being the upsell. Why not? They will be fully compatible with all the games. This is a so-called “aikido” move, where Microsoft can turn the console market it has lost on its head by making it an interoperable, open market with multiple choices like the PC market, and gain a unique advantage. It’s also a win-win because it benefits Microsoft, hardware makers, developers, and gamers. It even benefits companies like Steam, Epic Games, and other game developers that ostensibly compete with Microsoft/Xbox.

The “Xbox PC” device type isn’t a layer on top of Windows, it’s a stripped down and optimized version of Windows aimed at making the platform more device-like. This is why I used the NUC as an example of what could be a future Xbox “console,” these are simpler devices that can work only with a controller. The upsell isn’t just a more premium (read: expensive) gaming PC, it’s full Windows. And that upsell doesn’t have any downsides, because those PCs will be powerful enough that allowing Windows to work normally won’t impact games anyway. We have Game Mode. It’s a black box.

The more I think about this, the better I feel about it. This will likely save Xbox, but I think it’s going to be much bigger than just that.

Tied to the above, spacecamel asks:

Thinking through what Microsoft has done with XBOX in Windows. Where do you think Microsoft is going with this? Are they heading toward letting OEMs sell general XBOX computers similar to what OEMs did for Media Center Computers in the 2000s or will they keep the “XBOX OS” for specialized hardware such as the handheld they showed off? Will my next XBOX be a NUC with a Steelcase controller?

I think so, yes.

I noted in my write-up about Hellblade II: Senua’s Saga Enhanced that Microsoft has changed the way it refers to Windows PCs when it lists the platforms with which a first-party game is compatible. Now, this platform is called “Xbox on Windows.” Previously, as with, say, Call of Duty: Black Ops 6, it was “Windows PC.” (It also listed Steam separately, when available, and still does.) As per above, this is a supposition, or an educated guess, but given the evidence, it appears that future first- and/or third-party Xbox hardware will include what we think of today as consoles, gaming handhelds, and then full-on gaming PCs. And that each can/will be made by the same companies. I could see Microsoft wanting/needing to make its own first-party hardware for the coming generation of “consoles,” and then perhaps not bothering after that if the market responds as I think it will, since other hardware makers will be more successful at that.

From a software perspective, current and future gaming handhelds could run “normal”/full Windows, but they are using this optimized (and as-yet unnamed) Windows version to make up for the hardware restrictions and to make them more device-like. It’s reasonable to expect the next Xbox “consoles,” whether they are first-party, third-party, or both, to do the same, as the goals are simplicity (with a device-like experience), performance, reliability, and so on. But they, too, could of course run full Windows as they will just be PCs. Those that want more can get real PCs and just play games. This works fine today on mainstream laptops. It’s only going to get better as we move forward.

? And never the twain shall meet

rtillie asks:

Hi Paul, in a recent article you suggested a good password manager, and a seperate good 2 FA solution. Couple of questions I have, curious on your line of thinking. Why do you want these to be 2 seperate apps? As most password managers also do 2FA? Is it a separation of concerns?

I’m not sure how others think or do things, but my brain works in a sort of “set it and forget it” fashion where I figure out the best way to do something, make my choice, do the thing that way, and then can’t remember why I do that. This usually works OK, but I sometimes have to defend this choice or, worse, I will revisit it and I end up reading what I wrote about it (if that exists) or have to go through the whole process again. I can’t tell you how many times I’ve done that only to go right back to the exact same way of doing things. One good example is the way I consolidate email from multiple accounts into a single account. There are all kinds of ways to manage email, what I do works well for me, and when I do revisit this every several years, I stick with what I was already doing.

With that in mind, as I wrote the bit about using separate authenticator and password manager apps, I experienced a moment of self-doubt. Why do I do this? Why am I suggesting to others that they do this as well? I had just written about a nice upgrade to Proton Pass, the password manager I use, and how it can now be used to store anything securely, and I know it can be used as a 2FA authenticator too. Surely there is some reason. In that moment, I decided to move on and just finish the article. I wanted to, and did, present it as more of a conversation than a dictate from on high, it’s not like I’m a security expert, and I was curious what others thought and whether there might be disagreements or nitpicking. And so I figured it would be OK.

But there is a reason. More than one, really. But the big one is that you don’t store the key to a safe in that safe. Your password manager is protected by 2FA and it is a basic security best practice that you do not use a thing to authenticate the same thing. If that makes sense. Some will position this as a single point of failure issue, which is a sort of red herring or excuse for those that don’t trust or use password managers. But I do not see it that way. Because that is not how I use Proton Pass.

I have a sign-in account for Proton Pass that includes a username and password. Neither are saved/stored anywhere: When I sign in to Proton, on the web, in an app, or anywhere else, I type those things in from memory. But this account is also protected with 2FA, and that’s done through Google Authenticator, not Proton Pass. Yes, I could store that in Proton Pass, and I could then access it from another device. But I don’t want that there. I want it somewhere else. Somewhere else that automatically syncs to a cloud and is itself protected with the full extent of security possibilities. Something I can access from a different device than the one I am using to sign in to Proton. (It could also be the same device, of course.)

I didn’t “invent” this line of thinking, obviously. But I did get it from an expert. As passkeys were becoming more viable, I spoke with a friend from a security firm who warned me about separating 2FA/authenticator from password manager at length. I took it to heart, and now I just do it. And I don’t really think about it.

You suggested Google Authenticator for non-microsoft accounts. Doesn’t the microsoft authenticator also backup your codes? I believe it did. And how do you balance using Google Authenticator with getting away from big tech? When you would be locked out of your account, would it also lock you out of this google authenticator app?

Microsoft Authenticator supports backup and restore, but not live sync. Google Authenticator does live sync. If I add an account to Microsoft Authenticator on one device and opened the app on another, that account is not available on the second device. It’s only on the first device. With Google Authenticator, this is automatic. All accounts are everywhere, instantly.

One of the big issues with security is that it’s inconvenient. This is why many still leave their phones unlocked or whatever. It’s not so much being lazy as it is security being a hassle, something that gets in the way. I try to balance security and convenience. That’s why passkeys are so crucial, especially now that they’re portable through some password managers (another reason I love Proton Pass). And it’s why I use Google Authenticator. It’s secure enough. It syncs everywhere, instantly. And so it’s convenient, too.

Tied to this, I also use multiple devices. That’s true for PCs, and it’s true for phones. I have four devices right now that have Microsoft Authenticator and Google Authenticator on them–my Pixel 9 Pro XL, iPhone 16 Pro Max, Galaxy S25+, and my iPad–and each is using Proton Pass too. But I also have Proton Pass on every single PC I use. And there are probably 20+ here in my home now. So the security is good and the convenience is excellent. This helps as a fallback of sorts if something goes wrong. For example, if Google suddenly kills my account for whatever reason, I can get into Proton Pass on whatever device and set up a new authenticator app. I also have a recovery key from Proton for this, and it’s not stored in Proton or any authenticator app, it’s in my OneDrive Personal Vault. There are all these redundancy fail-safes to help with various circumstances.

Given my age, it’s perhaps not surprising that I have all these pat sayings to summarize complex ideas. One that applies here is “optimize for the everyday.” Yes, it is possible that Google could lock me out of my account and that, if that happens, I would lose access to Google Authenticator. And yes, this sort of did happen to me, though it was “only” my YouTube channel access and not the full account; whatever, I know what that feels like now. But is this likely? Is this really a concern? It’s a minor concern. But I don’t actually believe it will happen. So I optimize my security posture, so to speak, for security and convenience. And it works well. I use these tools every day. Every single day. And if the worst case scenario should unfold, I guess we’ll see what happens. But I feel like I’m prepared.

Thanks a lot for your work and articles. Really convenient for me that your investigating all these topics that are also on my mind now, always good to see other peoples approaches, and saves me a lot of research.

You bet, thanks. This is one of those areas where we’re all sorting through it and trying to figure out the right things to do. I hope I’ve emphasized that I am not the be-all/end-all of this topic, that it needs to be a conversation and that we all need to be ready to defend what we’re doing and then change if we’re proven wrong. But it has to work and then we have to get on with life. It’s easy to obsess over this. For myself, I want to just get it right. And then stop thinking about it, at least day-to-day.

? It’s almost literally 25H2 now

hastin asks:

Microsoft and the Windows team have been awfully quiet about “25H2” for the next version of Windows, meanwhile – it feels like more and more stuff is getting tested in Insiders that allows Windows 11 to match Windows 10 (such as a better start menu, customizable location of the volume slider, widgets location when taskbar is set to left-adjusted, small taskbar icons, clock with seconds in flyout, full PC file backup and restore, etc).

I think in a couple of months, it will roll-up these changes and basically create a situation where you can essentially migrate from W10 without feeling it – and they’ll promote it that way. Maybe as an enablement update that gets installed during OOBE?

What do you think the plan for 25H2 is?

I think you’re correct. I think it’s obvious that the next release is 25H2, and that Microsoft will not continue using 24H2 past the end of 2025. But I also think it’s odd that it hasn’t simply said this. Why be so silent and weird about something so banal and unthreatening? This makes me wonder if more is going on. It’s literally the type of thing I’m wired to obsess over. And the type of thing a “normal” person, like my wife, would never think about for even one second.

This is not a defense of Microsoft at all, especially given that it invented these stupid circumstances, but there are a couple of interesting changes that have occurred over the past year or so that I think play into this. In fact, I guess I would argue that this dates back to September 2022 and Microsoft’s decision to short-circuit Windows 11 23H2 by making all the features that would have been in that update available as a normal cumulative update for 22H2 a month earlier. At the time, it did so to drive Copilot usage; Copilot was going to be the marquee new feature in 23H2, but it didn’t want companies skipping that release. So it became mandatory.

So, two things changed.

Since then, schedules haven’t mattered as much. The move to H2 naming for feature releases (version upgrades) was designed to give the Windows team wiggle room on the schedule. But they’ve extrapolated that out to all updates. The most recent example just happened: Week D preview updates, like Week B “Patch Tuesday” updates, are supposed to come out that Tuesday. But for much of the past year, the 24H2 Week D update has always arrived later in the week. This week, for the first time, all three Windows 11 Week D updates shipped late, a first.

Also since then, whatever supported Windows 11 versions have been kept identical, functionally, regardless of versions (with just small exceptions). This means it doesn’t matter which version of Windows 11 you’re on, as long as it’s supported, as you will have access to the same features. Today, that’s 22H2, 23H2, and 24H2. Once 22H2 goes out of support in October, there will likely be a 25H2 to take its place. But it kind of doesn’t matter. These are different versions, which matters from a support lifecycle perspective. But they are (mostly) functionally identical. So it doesn’t matter to most users.

There are, of course, changes and differences that do matter. 24H2 is actually quite different under the hood, so to speak, from its predecessors. And so that old, pre-24H2 architecture will eventually be phased out, and it’s possible we’ll see new features that require the 24H2+ base, features that will never come to 22H2 or 23H2. Actually, that’s already happened, right? But whatever. The point here is only that we think about this stuff as fans or users, I think about this stuff from a book support perspective, perhaps, but most people don’t. And while I would very much prefer that Microsoft simply discuss 25H2–it’s too late now for a Windows 12 this year, I think–I guess it kind of doesn’t matter. I mean, it does matter. But nothing seems to matter in Redmond these days when it comes to Windows. There’s no logic to any of this anymore.

? Biting your nose to spite your face

jgraebner asks:

Recently, you’ve talked quite a bit about moving away from the “Big Tech” companies in favor of smaller, more ethical services. I definitely understand the motivation for doing that, but I’m curious about your thoughts on the risks.

I often reference some article I started and may or may not publish, but in this case, I see something I should have probably explained better but didn’t have plans for.

When I think about the tech products and services I use, it’s more pragmatic than anything. Meaning, I’m not on some moral crusade here. I will just do what’s right for me. And right for me is all kinds of different things, I guess, but in this case what we’re talking about is portability, convenience, utility, cost, and maybe some other aspects I’m not landing on at the moment. Sometimes, I will just make some change organically, maybe I see an announcement or a blog post or video about some new thing and I try it and it clicks. Other times, something bad happens, like with the Google/YouTube thing from January, and it triggers a rethinking. That latter event did just that, and the result is unfolding in the ongoing Online Accounts (2025) and includes the NAS I moved to recently. That’s something I knew I would use, but I was surprised by how well it works for day-to-day file sync and so I’ve switched to it completely for that.

Big Tech is unavoidable. We can pretend otherwise. We may have issues with privacy, data governance, or whatever. But at the end of the day, I use Apple and Google smartphones, I use Windows PCs, I use Apple TV, and I subscribe to what feels like 117 different online services. Whatever balance we strike there is personal in nature and specific to one’s needs. But I guess I view this as a lessening of exposure, if that makes sense. If I do lose my Google account, can I recover from that?

I don’t want to find out what that’s like. But the answer is yes. For example, I own the domain Thurrott.com and, yes, it’s tied to a Google Workspace account. But if Google shuts me down, I can move it. So I will lose some time. But life will go on. I will lose my YouTube content, which is mostly backed up now, and hopefully will be automated soon. Subscribed services and online sign-ins are a problem, of course. I wrote about the “sign on with Google” issue a while back, and the single sign-in thing, but if I were truly fastidious, I would only use aliases for everything I do online, but this isn’t a full-time job, I have a life to live too, so I may never fully make that change. Security vs. convenience. Or maybe obsessive vs. evenly keeled. I don’t know.

Anyway, I guess my point is that I try to be pragmatic about this stuff. Switching from, say, Microsoft Word to a Markdown editor like Typora may or may not seem like a big deal to some, but I’m a writer and I want to keep my writing, and it’s important to me. And that took a while. I only moved from OneDrive to Google Drive because of enshittification. And I only moved from Google Drive to Synology Drive because–holy s#$t!–it works so well and now I can control that, and it’s not overly expensive or onerous to do so. But these aren’t hard stops. I still have things in, say, OneDrive Personal Vault. I still have backups of data in OneDrive and Google Drive and elsewhere. I’m not a monk living in the woods. It’s OK that these things are there.

It’s also OK to think about change, evaluate choices, and then stick with what you were already using. It being Big Tech doesn’t inherently make it bad. There are conveniences to it. These things can cost less because of scale benefits. Etc.

Recently, I decided to start migrating to an email address on a custom domain. I seriously considered some of the smaller providers, particularly Proton Mail, Zoho, and Fastmail. After a lot of consideration, I ended up narrowing it down to Microsoft or Google and finally going with Microsoft. The main reason for this is that I felt very confident that either of those two companies are going to be around, and in the email business, for the foreseeable future. I also feel a lot more confidence about quick resolution to outages and that, Microsoft in particular, would be very responsive to questions or problems. In short, I felt like signing up for Microsoft meant that I could mostly just do the initial setup, activate auto-payment, and then largely not have to give it another thought.

People sometimes ask me why I use Google for email. My answer is two-fold. First, I had moved Thurrott.com between various services over time because that’s what I do, I test different things, retest them, figure out what’s best for me. Coincidentally, when I linked up with George and we started what became Thurrott.com, I was testing the domain on Google, and he was using Google at Petri, so it made sense to keep it there. Pragmatic.

But the second reason is that it’s excellent. I love the interface, I know the keyboard shortcuts, and the spam filters always work well. Moving to something else now is possible, but it would be work, and I don’t see the benefits or why I’d even bother. So this too feels pragmatic. It works.

That’s me. We all have our own criteria, that “matrix of options” that can go into any decision, with some weighted more heavily than others. And if Microsoft makes sense to you, there is no reason to second guess it. Having something that works and makes sense to you is just peace of mind. It’s what we all want.

There definitely are other areas where I feel comfortable with smaller providers. I’ve already migrated to different password managers twice (Roboform to Lastpass and then to 1Password) and feel pretty confident that any future migrations would be pretty low pain. Similarly, I’m also using tools from smaller companies for notetaking (Notion), budgeting (YNAB), RSS (Inoreader), music (Qobuz and Roon) and other areas where migrating elsewhere probably wouldn’t be that big a pain.

Right. There are two issues you raise here which really resonate with me.

One is portability. You want to make sure that you can move to something else if the thing you are using disappears or is replaced by something better. This is also pragmatic, and common sense. And just good thinking.

The second is this fear of the unknown, and how just doing something once can help you get over this. This was a big deal for me when I wanted to switch Internet providers: The first time was terrible and I was worried it would be difficult or expensive, but it was fine in the end. Switching wireless carriers is the same thing. Now it’s easy and not something I’d ever worry about. (Coincidentally, I recently switched to Inoreader for RSS too, and I was thinking about writing about that.)

It’s just for mission critical areas like email, cloud storage, the PC and phone operating systems, etc. that I find myself nervous about relying too much on companies that seem more precarious.

Totally fair. There is perhaps some combination of the two issues noted above that could change that, for you or anyone else reading this. But again, peace of mind is important. Whatever works.

I’ll also add that I can’t help being a bit skeptical about the theories about smaller companies being more ethical and trustworthy than the big tech companies either. To be frank, it wasn’t that long ago that Google was viewed as the scrappy, more-ethical alternative to bigger companies like Microsoft, AOL, or Yahoo. We all know how that turned out.

Sure. With the understanding that there are some exceptions–(re)read what Proton did structurally to ensure that it will continue, in perpetuity, to put people ahead of profits and stay true to its original ideals as an admittedly unique example–it’s important to remember that companies, like people, should be selfish. That is, they should be doing what’s best for them. For publicly held companies, this is complicated by them having to do what’s right for shareholders. And for Big Tech, it’s complicated by scale and monopoly and protectionism. But in yet another example of the pat phrases I can’t seem to stop repeating, good relationships are a win-win. That means they benefit users, customers, too. And that is possible with Big Tech. It’s why we all use those products and services. If it were just one-sided, no one would bother.

I guess what I’m trying to say is, don’t doubt yourself. You know what’s right for you. You have clearly done research to examine alternatives. And you have made decisions based on that work that are right for you. Others can only aspire to do the same thing, and hopefully do. Myself included.