Ask Paul: June 24 (Premium)

Happy Friday! And welcome to an epic edition of Ask Paul with some great reader questions to kick off the weekend a bit early.

Google Fee Fi Fo Fum

Lots of Google Fi questions today. Let’s dig in.

drewtx asks:

Google Fi. Why do you only use it when you are traveling, rather than just staying on it permanently (even when you’re home)?

I actually only use Google Fi again, after two years of using Mint Mobile. But the reasoning for switching from Fi to Mint was simple enough: Mint Mobile is considerably less expensive than Google Fi. But coincidental to this change in early 2022, the pandemic hit and so I wasn’t traveling, internationally or otherwise. So Mint Mobile made sense.

When travel opened up again in 2021, I do move to a system where I used Google Fi only when traveling internationally. The idea here is simple enough: you can “sleep” a Google Fi account for months at a time and not pay for it, so I’d only use (and pay for) the service when I needed it.

So why did I switch back to Google Fi earlier this year? Two reasons: convenience (it’s nice dealing with one phone number and one SIM/eSIM, and I got into weird issues with text messages while traveling) and I knew I’d be traveling internationally a lot more this year and going forward, and so it seemed worth the additional cost.

Not looking for a detailed cost benefit analysis of Fi – just curious why you don’t use it permanently

Check that second link above if you do want a detailed cost-benefit analysis. I looked at the cost of Google Fi over several months and compared it to the cost of Mint Mobile. It wasn’t even close: I cut my bill in half.

I am using iPhone (an aging but still viable XR) and Fi has some limitations on iOS. In the US it will not roam networks; it stays on TMO – but that was my previous provider so no significant problem there. Also, when traveling overseas The Fi+iOS combo does not support tethered or WiFi hotspot.

Hm. I definitely used a Wi-Fi hotspot in Mexico off of Google Fi before our cable Internet was installed, but as I think about this now, I’m wondering if I didn’t do so via my Google Pixel 6 Pro, as it has a Google Fi data SIM. These are free, but as I’m looking at that page I linked to, it says that tethering from a device with a data-only SIM isn’t supported. And now I’m not sure what to say.

So I looked it up. I did tether for a day, and wrote about that in What I Use: Mexico City, May 2022 (Premium): “I only used the Pixel for a few moon/distant photos, and to tether my laptop at the apartment the day before our Internet was connected.” So it definitely worked via the data-only SIM, despite Google’s support note. The screenshot in that article shows all the data I used that day. (I don’t recall if I tried this on the iPhone first, but I bet I did.)

jt5 asks:

I am thinking of switching to Google Fi. I noticed that Google says it is still in beta for the iPhone. From what I can tell there is no 5g, no auto switching between networks, voicemail only comes in the Google Fi app, and there is some additional setup for the SMS messaging. It seems to have been in “beta” for the iPhone for quite awhile(Which seems to be the norm for Google?).

How well does it work with these features missing?

With regard to the iPhone being in (perpetual) beta, this hasn’t been a huge issue for me. As you note, it only supports 4G/LTE data and not 5G, but whatever. You have to use the Google Fi app for voicemail (or you can just call voice mail from the iPhone Phone app like it’s 2005 again), but Google also transcribes voicemails and texts them to you. And I don’t get a lot of voicemail. And there’s no networking switching, but like you, I was on T-Mobile (via Mint, an MVNO) for the previous two years, so that’s fine too.

Have you seen anything indicating when the features come to the iPhone?

No. And it may never happen, thanks to the strained relationship between these two companies. Interestingly, Samsung’s latest flagships support the full Fi experience (and are even sold via Google Fi and can be added to the monthly bill). They even support networking switching! (Apple could easily support this too, but I bet they won’t.)

Would you recommend it over a carrier like T-Mobile that does have international data?

This is kind of subjective, but there are pros and cons going in either direction. To me, the big and unique benefit of Google Fi is that it works like a streaming video or music service (Netflix, Spotify) in that it’s month to month. You can cancel it at any time and go elsewhere, and you can pause the service, cost-free, for months at a time. That’s cool if you need it. But T-Mobile customers could benefit from priority traffic and probably other perks.

I do sort of look at traditional carriers from time to time, but the costs are not as transparent as with Google Fi and the last time I looked a T-Mobile, it amounted to me basically having to talk to someone to figure out what the real monthly cost would be. It is absolutely worth doing this, of course, and if you have an international need, I would just pay close attention to that and try to figure out what the monthly bills would look like on each service. Google Fi’s costs are transparent, as noted, so that bit is easy. The trick is figuring out what you’ll really pay on T-Mobile (and what the speeds/data limits will be like).

This all varies by carrier. My wife is on Verizon and, as with other carriers, international access has gotten better over the years, but it’s still mind-boggling how it works. She is on an unlimited family plan with five people (her, our two kids, and her parents) and it costs about $240 per month. When she goes to Europe, she can pay $10 per day to get her normal access there, or she can buy a monthly pass at whatever cost. For Mexico (and Canada), she doesn’t have to pay extra, but this assumes she doesn’t surpass her average monthly data usage; if she does, her data is throttled.

This mess sort of explains why I like Google Fi. In addition to transparent pricing, Fi also has a cap on pricing: once you hit 6 GB of data (which is $60 in a month, plus the $20 for talk/text) you’re done paying, and you can just keep using more data. (Yes, there is a throttle at some point, but I’ve never come close to it.) When we were in Mexico in May and ended up staying for four or five more days than expected, I hit that 6 GB limit for the first time in years. My total bill that month was $86.11, and I used 9.24 GB of data. I just love that.

Mexico

drewtx asks:

Can we look forward to more writings from on you and/or Stephanie regarding your transition to Mexico?

Yes, but I will probably confine most of what I write here on Thurrott.com to tech-related topics related to things like the More Mobile setup, smart home products and services, and so on.

But I will still be creating content related to whatever it is we’re doing in Mexico. As someone else noted, my wife Stephanie and I have created a YouTube channel called Eternal Spring and while I haven’t promoted this yet because it is very much not ready, we also have a related Eternal Spring blog with written content. Yes, it’s bare-boned right now. I have a lot of other responsibilities that take precedent over that, including my job and the book. So, it will come together slowly over time.

To be clear, we don’t see this as a big revenue thing or whatever. But as I have written elsewhere, there is a real lack of quality information about this topic, and not just on YouTube. And we both feel that we can help others to do what we’re doing and, hopefully, not make some of the mistakes we made. Where this goes is as unclear as what our future is in Mexico, but in both cases, we will just take it as it comes and see what happens.

Steam Deck

OldITPro2000 asks:

Do you plan on reviewing the Steam Deck? Or perhaps you did and I missed it? I’ve been checking out reviews about it recently and it looks like a cool device, even in the first generation.

No, I don’t believe so. I don’t often review non-Microsoft gaming hardware. I’ve not ever reviewed a Nintendo console, for example, and I think the last Sony console I reviewed was the PlayStation 4.

Follow-up to this: seems like this is something Microsoft should have considered doing as a hybrid Surface/Xbox device. Do you know if this is something they ever considered?

I know that they have considered a portable Xbox console. But with Xbox Cloud Gaming, in particular, we’re entering a world in which any computer, even a Surface Go, is a viable gaming machine, assuming you have a decent Internet connection. If we are one day able to stream our Xbox-based digital game collection to any Cloud Gaming-based device, that would really put this service over the top.

That said, a gaming PC-based Surface PC, especially a laptop, would be very interesting. That would be the best of both worlds, with the ability to play high-quality PC games locally and stream other games. I feel like such a PC makes more sense than the low-end Surface-branded PCs they’ve been making in recent years. Gaming PCs are premium PCs and would fit in nicely with the Surface (and Xbox) brands.

Windows on ARM: chosen one?

will asks:

Windows ARM. I caught what you and Mary-Jo were talking about with regards to Windows ARM and I am curious if this something Microsoft is truly invested in or if this is just a side project, aka Windows Mobile? The reason Apple is successful with ARm is they put all their weight behind any change they implement and ARM is their future.

I was thinking about this after that discussion on this week’s Windows Weekly and it occurred to me that the basic reason why Apple is successful with Arm on the Mac is that Apple doesn’t partner. (And when it does partner, it’s out of necessity, and the company is always looking to the future where they can kill off that partner.) That is, when Apple makes a strategic decision for the Mac, there is no other company out there that can hold it back. They decide on Arm, their own chipsets, whatever. It’s all them.

(Additionally, Apple realizes synergistic benefits to having all of its platforms on the same hardware—and software—bases. This isn’t the case in the PC world. HP, Dell, and most other PC makers do not make smartphones, though some make tablets. Regardless, making the underlying platforms similar or identical does not benefit them as it does Apple.)

It’s not like that in the PC space. Microsoft can’t just “declare” that Windows is going Arm. Intel and AMD, its two biggest hardware partners by far, would revolt and/or simply ignore this. Many of its PC makers would also balk, as they’ve spent decades investing in the x86 ecosystem and have a massive list of suppliers and other partners that they need to create their PCs. Developers would simply ignore this, too, as they pretty much have. There is no one entity, Microsoft or otherwise, that can unilaterally declare that the PC is going to Arm. It can’t happen.

So is Microsoft fully invested in Arm? I think the way to phrase this is that Microsoft is as invested in Windows as it is willing to be and that it will do what it can to make Windows successful going forward. That includes investing in Arm, which it has clearly done, and working within the confines of its partner-based system to push this platform forward with Qualcomm and with PC makers. And you see the results. They’re lackluster. Qualcomm does not understand the PC space. And the PC makers do not understand Arm. (This is vague and general, but let’s just run with it.) They’re all trying.

How do we know these companies are serious? Well, Qualcomm spent $1.4 billion on NUVIA, a company that made PC-class Arm chipsets, so they’re serious. Microsoft has (somewhat suspiciously) not expanded the availability of Windows on Arm past Qualcomm despite public calls for this from Samsung and at least one other hardware maker, so they must feel that this NUVIA acquisition will pay off. And PC makers, inexplicably, keep releasing new WOA PCs. Lenovo just launched a WOA-based ThinkPad, of all things.

This issue here is a classic business case: these companies all have successful existing product lines that need to be supported and evolved, but they also agree that Arm, in this case, is either “the future” or at least will impact the future. And so how do you do both? Support what works now and keep working on this new thing? You do so conservatively. Intel, which is enormous, has slowly evolved its x86 chipsets to have Arm-like hybrid architectures with many big and little cores, for example. If that work continues to be successful, maybe literal Arm on Windows won’t matter.

Microsoft seems to sort of dabble in ARM but as far as I know, and you would know better, there is no plans to move away from x86 anytime in the next 3-5 years. Yes there are new ARM chips that are supposed to close the performance gap with Intel and Apple, but those are a year away and Apple will then be moving on to M3. So, is this Microsoft hoping ARM will catch up and this in turn puts pressure on Intel to do better or is ARM just a side project?

Terry Myerson told me explicitly that the exact outcome didn’t matter. Either Arm would win, lessening the market power and influence of Intel and x86, or Intel would finally get off its ass and make more efficient (and Arm-like) chipsets, and they would continue forward as the platform of choice for the PC. Either way, Microsoft would get what it wanted, and, either way, PCs would become more efficient and offer that improved “power per watt” that Apple always talks about.

The dig against Intel has always been that they’re really good at piling on the performance at the expense of efficiency. But the reality of the PC market is that most people haven’t cared much. Our PCs are thicker, heavier, and louder (from fans), but they get decent battery life, and the performance and compatibility are excellent. In this month’s WWDC, Apple showed a chart proving that Intel’s high-end chipsets are faster than the M2, but the M2 won on “power per watt.” Someday, that will matter to the mainstream, I guess, but that’s the same argument for solar panels or electric cars. Yes, each has seen some level of success, but most consumers go with the less efficient but more common and well-understood thing from the past. These things don’t happen overnight.

Put another way, Intel is to hardware as Microsoft is to software: the old way of doing things. Microsoft adapted to the cloud-focused world nicely, and it is trying to make Windows move forward with Arm. But Intel took longer. Pet Gelsinger saw the issues and is evolving Intel, and the 12th-Gen Core chipsets are perhaps the first solid example of its mainstream chipsets finally taking this step. That they are available in multiple efficiency levels speaks to the issues with bridging the past and the future. But they are at least doing both now. I bet it’s going to work out well for them.

Related to this, crunchyfrog asks:

With the near total adoption of the ‘M’ series processor for Mac computers, it is no longer possible to run a Windows VM with x86 and end users must run the ARM version instead. Do you believe with the popularity of Parallels for running Windows on a Mac that this could benefit a more widespread adoption and use of ARM Windows?

Mary Jo and I were just discussing a major potential change to Windows on Arm that could result in public ISO availability and license purchasing. And that would, of course, make running Windows 11 on Arm on an M1/M2-based Mac legitimate. And would be a major step toward making WOA a mainstream computing solution like its x86/x64 siblings. So we’ll see what happens there.

But I don’t think that Parallels on M1/M2-based Macs is popular enough on its own to make WOA successful. This is sort of a niche solution at this point. When virtualization came to the Mac via software-based solutions like Virtual PC over 20 years ago, there was a real need: the Mac didn’t have a big library of native software, and the ability to run Windows apps, if slowly, helped make the platform viable. And Apple adding Boot Camp to Mac OS X achieved the same goal.

But these days, most Mac users don’t really need Windows software anymore. There are ample native apps, and web apps, and now mobile (iPad/iPhone) apps too. And so Parallels is there for those that do need this. But I feel like that market is shrinking. And that as Mac usage grows, the number of people who want Windows apps will not grow with it.

EdgeOS

martinusv2 asks:

Did you notice that Microsoft Edge started showing some animated backgrounds when you open a new tab?

No, but that might be because I’m using Momentum as a replacement for Edge’s new tab page. But I can’t stand superfluous nonsense like that. A web browser should get out of the way.

Loved your tweet about Edge is not an OS. Would that not be the idea? That eventually Windows would finish to be like ChromeOS? And everything is running inside the browser?

This notion of an EdgeOS, a sort of Microsoft Edge-based Chrome OS, will not go away. And not coincidentally, Brad and I sort of talked through that a bit this morning on FRD. I joked that it could be like Windows RT, but instead of Store apps, it would run web apps.

That doesn’t actually make sense, and I don’t know that Microsoft is considering such a thing. But I do know the history of Microsoft platforms that look like Windows and aren’t (full) Windows, and that such things—Windows RT, Windows on Arm, Windows in S mode, Windows 10X—have always failed, and for the same reason: if they can’t run all Windows apps, they are not Windows, and no one is interested. And an EdgeOS would be the most extreme example of such a thing.

In the spirit of being that wrench in the bicycle spokes, I will share another thought I had recently, and it’s borderline evil. What if Microsoft was working on an EdgeOS, but instead of offering basic support for Windows apps, it instead runs Android like Surface Duo? That would make it a lot more like Chrome OS, of course, though there would be issues with app selection unless they can get the Google Play Store on there. But of course they can. It’s on the Surface Duo. Why not on a tablet, 2-in-1, or laptop?

I could really see that happen: Android arguably makes more sense on a low-end PC-like device than does Windows or some subset of Windows. I know it’s not what a lot of Microsoft fans want to hear, but it could make sense. What if Microsoft is part of Google’s new Android tablet push? It would be weird.

Microsoft, the security company

martinusv2 asks:

After watching this week’s “Security Now!” where Steeve talks about how Microsoft handle (badly) vulnerabilities, fixes, time to get those fixes and how they do test. How can Microsoft say to us that any of their products are secure? And still say that we are using the most secure Windows to date? See each month the high numbers of 0-days and vulnerabilities they must fix? How can Microsoft be so inapt to create secure software?

I wonder the same thing all the time. It is unbelievable to me how often there is a story about some security researcher who found a vulnerability, properly alerted Microsoft, and then got so frustrated by its inability to ever ship a fix that they went public with it, exposing Microsoft’s ineptitude. On the flipside, Microsoft’s security solutions are generally well-regarded, and Defender has probably surpassed Outlook as the brand that’s used on the most Microsoft offerings. None of this makes any sense.

RTFM, RTM

madthinus asks:

So my question is more of an observation. Windows RTM now means the Driver model and Kernel work is complete. The UI and everything else now is subject to patching. UI in settings can change from Patch Tuesday to the next. Explorer tabs is now a patch Tuesday introduction.

Yeah. The biggest change in Windows since the release of Windows 10, arguably, has been the expansion of ways in which Microsoft can update any component of the platform without requiring a version upgrade. And that makes Microsoft’s promise to go from two feature updates per year to just one with Windows 11 a lie: it can, will, and has updated Windows at its discretion at any time.

Is this good or bad? It’s both, I guess. But the bad news will be seen most clearly when the inevitable happens and Microsoft updates some little corner of Windows without alerting anyone via a cumulative update and PCs stop booting. This sort of just happened with the OG Surface Go, where an update bricked the  PCs, and Microsoft is just replacing them because they can’t be fixed.

How are you going to handle this for the book?

Poorly.

Kidding. I’m going to have to be nimble enough to deal with it. With the Windows 10 Field Guide, it was tricky—OK, impossible—to stay up to date on every single app update that occurred. With the Windows 11 Field Guide, that pace of updating could occur with non-app system components. This only happened a handful of times with 22H1. But we’ll see what it looks like with 22H2. It’s definitely something I’m worried about.

C-level emotions

Harrymyhre asks:

Why does nobody ever talk about this? When corporate buyers sit down to buy apple products, are they happy dealing with single vendor sourcing?

Today, Apple is taking baby steps into the corporate world with programs like Apple Business Manager and Apple Business Essentials, so the support is there for companies that want to go Apple. But there is probably no example of even a single large corporation that is all Apple, so Apple is really just another supplier, alongside PC makers like Dell, HP, and Lenovo, Android device makers like Samsung, and whatever else. (And I bet most corporations do standardize on a single PC vendor and/or product line when they’re all-PC.)

And let’s be honest here: Apple has a mesmerizing effect on some people, and sometimes those people are in decision-making capacities at big companies. Executives demanding Apple products—whether they’re iPhones, iPads, or Macs—drove adoption in corporations years ago. And companies like Microsoft that offer device management solutions have long supported them as well.