Ask Paul: January 12 (Premium)

Happy Friday! Let’s kick off the weekend a bit early with another blockbuster edition of Ask Paul with some terrific reader questions.

What reviews?

Leo_W asks:

Typically your computer reviews revolve around laptops. Do vendors offer you the opportunity to review desktops?

No, not typically. And I’m sure this is tied to the sales data they see, that laptops outsell desktops by a wide margin. But there have been a few exceptions over the past few months. I reviewed an Acer SFF PC in October, and I two HP desktop PCs in now—an Envy Move AIO and a SFF mini-PC I’ll write about soon—so that’s at least something. I am interested in what ASUS is doing with the NUC, though the gaming-class NUC they just announced is a bit north of what I’m looking for, capability-wise. And for whatever it’s worth, I actually use a desktop form factor workstation here in my home office now. (There are challenges to this, but whatever.)

I’m open to all kinds of PCs for review. But I am somewhat constrained by what the hardware makers offer me. I don’t typically approach them to ask for specific devices or whatever.

As Windows 10 gets pushed further to the retirement home I will need to consider a new computer purchase. I prefer desktops since they’re easier to repair/upgrade and historically a comparably spec’d desktop is less expensive than a laptop, especially if one can repurpose existing monitors, keyboards, and mice.

Yep. This is understandable: Even with the gains we’ve seen in recent years around laptop repairability and serviceability, it’s still kind of a crap shoot. The most common configuration I see now is soldered-on, non-upgradeable RAM but with other components (SSD, Wi-Fi, battery, etc.) that can be replaced or upgraded. I’m hoping that Framework and the right to repair collectively “inspire” PC makers to do better. But in the meantime, yes, a desktop PC does have certain advantages, among them the upgradeability and serviceability stuff.

I like the thoroughness of your reviews and would be interested in seeing your write up of a mid-level desktop. Something with an i5 or i7 equivalent processor and the option for 4K video.

Thanks … I mean, I would be open to that. I guess I could at least write about my experience with the workstation and perhaps that will trigger some conversations with PC makers. But again, they’re pushing what sells and does best for them.

Be brave

Brumfondl asks:

Based on your recommendation I have finally transitioned to Brave from Edge (which I was also on due to your recommendation :P) and I was wondering if you were considering doing a 2024 guide to setting up and using it as there have been a number of features added in the years since you transitioned to it, including vertical tabs and LEO.

Yes, I would have published that by now, actually, but I’ve been distracted over the past month, largely by the security topics I’m finally putting behind me. But that article is mostly written and just this past week I took a ton of screenshots of me configuring Brave for that post. So it should be soon.

Related to this …

Edge (which I was also on due to your recommendation :P)

I know you’re joking. My own joking response to that was, “Hey, hey!”

But in my defense, Microsoft Edge was exactly what I wanted (and what I think the ecosystem needs) when it was first released, a lightweight browser that was 100 percent compatible with Google Chrome but stripped off all the Google integrations, especially the tracking. But Microsoft has enshittified the user experience over time by piling on too many superfluous features. And while this wasn’t clear in the beginning, it also quietly includes its own tracking functionality in there that’s just as bad as what Google does. (That it did so while promoting the giant Tracking Prevention UI in Edge settings isn’t just ironic, it’s hypocritical and, I think, unethical.) The problems with Edge’s underpinnings only became obvious over time. And I cannot recommend this browser to anyone anymore. Which sucks because there are some good features in there.

This isn’t really the time or place for this, but I feel very strongly that browsers should follow the model Microsoft uses with Visual Studio Code, where the base product is small and lightweight and users can add the functionality they need/want via extensions and then those things sync to their accounts so each time they bring up the browser for the first time on a new PC, they get “their” browser. It’s so obviously the right approach (to me), and Microsoft/others could even offer extension bundles during first setup based on things like security, tab management, AI, or whatever else that would make it easy for mainstream users to get up to speed.

No browser maker does that, as it turns out, not really. But Brave comes the closest, I think, and you can remove the 2-3 Brave-specific features you don’t want very quickly and easily. Plus the whole privacy/security by default thing. I’ll be looking more closely at alternative browsers like Arc and DuckDuckGo this year as they evolve. You never know.

I should also thank you for the Duck Duck Go search engine article as it works well and I can pretty reliably find that which I seek 🙂

I love Brave and was surprised by how well DuckDuckGo met my needs given that I’ve tried and failed, repeatedly, with every alternative search engine imaginable. But I did a similar experiment with Brave Search over the past 10 days-ish, and it failed where DuckDuckGo succeeded, meaning I wasn’t getting the results I needed, and so I would try the same search in Google and get that immediately. And so I went back to DuckDuckGo and everything is fine again. I can’t explain why some things work and others don’t in this case, just that they do or don’t. And for whatever reason, DuckDuckGo works great (for me). As always, our mileage varies.

The enduring problem of email

Christian-Gaeng asks:

Hi Paul, I wanted to ask you how you use and process emails. Do you still use email programs like Outlook or Mailbird (Gmail), or do you just use the corresponding websites? I always find it a shame that Gmail cannot be used properly in Outlook with a private account. I find Gmail to be more user-friendly than Outlook, especially the calendar.

Like most, I spent years using and experimenting with different email clients. And in the end, I decided that using a single web client on PCs (and a single native app on mobile) was the right approach for me. I very much prefer Gmail to any version of Outlook (web or otherwise), and so I consolidate all of my email accounts into a single account (which happens to be my Google Workspace account). And the short version of this configuration is that I set up each secondary account to forward all emails to Workspace and then delete the originals, and I configured my Workspace Gmail account so that when I respond to emails, I do so using the original account, and I can send new emails as all of those accounts too (which I rarely do). Having a single place for this just makes sense to me.

My issues with Outlook, especially the bloated legacy desktop client, are many, and I have PTSD from using it. But key among them is that it has never turned into a good Gmail client, even though Microsoft never had issues making that work in other email apps like Windows Mail and the new Outlook. I don’t get this. But I also don’t care: I would use Outlook if they paid me to do so. I hate it. I know others don’t. We all have our own preferences.

Anyway. Yes, I very much prefer Gmail and don’t see any reason to ever change.

You’ve been SKUed

helix2301 asks:

What was the purpose of all the SKUs with Vista? I remember Starter, Vista Home, Home Premium, Business, Enterprise, and Ultimate. I was and still am a PC gamer and I remember getting video games that said on the box requires Vista Ultimate. I know 10 and 11 have Sku as well but a video game is not going to work if you have home on pc instead of pro. I was wondering what the point of this was and how it affected the industry.

If you think back to this time—Vista came up out of Longhorn, which went public in 2003 but didn’t ship until late 2006—there are some key bits of history that most people probably forget. Microsoft had transitioned from being a one-product company (Windows) to being a three-product (Windows, Office, Server) company, which made it more resilient to outside threats. But even less well known or remembered, Office became its biggest product by both revenue and user base, surpassing Windows. So it was only natural that the Windows and Server teams looked at Office to understand why it was so successful.

I mean, Office had been successful to some degree since its inception. But it exploded in popularity largely because that team employed a strategy that was common in other markets: It offered multiple versions of the product, each with different capabilities and price points, got customers in the door thanks to the low cost of the entry-level version, and then upsold them to a more expensive tier. You see this most obviously with automobiles, for example.

But with Office, Microsoft had expanded the number of SKUs (product editions) pretty dramatically, and it had started adding new apps that were at first only available in the higher-end versions, and new business/enterprise editions that integrated with or required a growing family of Office servers. Windows NT had started with Workstation and Server variants, and that continued through Windows 2000, which had a single desktop SKU but a handful of Server SKUs. And then Windows XP started with Home and Pro (in part to attract 9x users to upgrade), it greatly expanded over time with multiple Media Center, Tablet PC, and x64 editions, each of which carried a premium and was more profitable. (And then Starter Edition, which was about killing Linux on netbooks.)

By the time Microsoft shipped Windows Vista, it had gone a bit nuts with the SKUs. There weren’t just too many of them, there was a confusing mix of features in each and few were true supersets of each other and the line between each seemed arbitrary. I’m sure PC makers and customers complained, and so Windows 7 had fewer SKUs. And by the time we got to the modern era, that had calmed down a lot at least from a retail/consumer perspective: No one is buying Windows in stores anymore and so having multiple SKUs for that audience doesn’t make any sense. But really, there are multiple SKUs out there, especially in the commercial space. Where this same premium fragmentation model is probably still working out well for them.

I don’t like Apple or the Mac for all kinds of reasons, but they always offered just a single client version of OS X/macOS, and they were quick to lower the price on upgrades and then eliminate those fees. Today, Windows is pretty close to that model, where we will have two SKUs for consumers (Home, Pro) but upgrades are free. I think this is all we can expect.

Shorter version: Vista arrived at an awkward time in which Windows had been super-successful and no one knew that it was about to be supplanted by mobile, and so that product represented a world that was about to end. This reminds me of a sports team that wins a bunch of championships and then builds a new stadium thanks to that success, but then does poorly because all the players who made that success happen retired.

Agility is easier when you don’t rely on others

iAlrakis asks:

It’s not like I’m keeping track of this every day but this is the impression I get. Apple with Mac OS, going super strong on ARM processors. Recently starting playing with my Raspberry pi400 and learned that there are plenty of linux ARM distros. Why aren’t we hearing more about Windows on ARM or even ARM based Microsoft desktop solutions. They’re a huge company, why isn’t this going much faster?

This is tied to an interesting business case study that’s really about corporate culture. Which endures over decades, and past the point where the people who started it are long gone. This explains why Google still hates Microsoft today, even though these companies would be best served by partnering to counter the Apple monoculture. Google came up in a world in which Microsoft dominated the market illegally and its co-founders feared and hated them.

Microsoft’s corporate culture is changing a bit under Satya Nadella, finally. But it was long based on the personality of Bill Gates, its most aggressive and public co-founder, and its first CEO. Microsoft came to power in the 1990s in large part because it partnered with other companies, first the PC makers that distributed MS-DOS, Windows, and then Office, and later the partner ecosystem that rose up to support its complex server and infrastructure products (among many other things). There is a real partner mindset there, though the cloud computing push has finally chewed away at that in the most lucrative parts of the company.

But Apple’s culture was (and still is) tied to Steve Jobs and his prickly reactions to the companies that pragmatically abandoned him and his companies when they were doing poorly. This explains his lashing out at Adobe, repeatedly (Flash, Photoshop for OS X, etc.) when he was CEO: In his mind, Adobe only existed because the Mac made desktop publishing possible, or at the very least it was the partnership of the two firms that led to a generation of success. And he saw what Adobe did as a betrayal. He believed that the only way that Apple could thrive was to call its own shots, and so he instituted the Apple we see today: Partner only when necessary and then only temporarily, and then abandon those partners as soon as possible.

There are so many examples of that, but the first, perhaps, was Microsoft, which it partnered with in 1997 to survive. Short-term, it got the money it needed to keep the business operating, and longer-term, it got a commitment from Microsoft to keep producing Office for the Mac. Oddly, or coincidentally, right when Adobe was ignoring Mac OS X, Microsoft was supporting it with Office. But Apple made its own browser and dumped IE, and then Apple made its own office productivity suite in a bid to sever that outside necessity. Apple’s partnerships since then have all been one-way, dead-end streets. And that all came from Jobs: Don’t rely on anyone else.

So here we have two companies with vastly different strategies. They are literally the two biggest companies in the world, but they couldn’t be more different in this regard. Apple’s strategy shouldn’t work, frankly, they are a bizarre outlier in the personal technology world. But whatever: Because of this corporate culture, because both of its previous two microprocessor suppliers, Motorola/IBM/Power PC and Intel, both let it down by not delivering exactly what they needed, it went down the path of creating its own microprocessors and chipsets, and in doing so, it eliminated its reliance on outsiders over time. (Its current partnership with Qualcomm is a sore spot for Apple, which wants to drop that company like a bad habit but still needs its Wi-Fi/communications chipsets for now.)

Think about Microsoft’s experiences with microprocessors, especially on the client side: Whatever they’ve done they’ve done in partnership with hardware companies like AMD, Intel, and Qualcomm. This is the Microsoft way, though its jealousy at Apple’s success and its own slow progression in eliminating partners is starting to make it more like Apple. But that’s for the future. For today, Apple has a rich history of doing its own thing, and that has helped it be agile in evolving its microprocessors. Microsoft has to partner, and it’s thus subject to the whims of, and in this case, the slowness of, its partners. Intel is just now delivering on the low-watt, efficient PC chipsets that Microsoft had been begging them to make for years. And Qualcomm? You have to think Microsoft cannot wait to be rid of that boat anchor.

Anyway, that explains it: Apple’s go-it-alone mentality has had this positive outcome in the chipset space, while Microsoft’s partnering mentality (necessity), which once served it quite well, has had this negative outcome: Microsoft has done everything it can do to make Windows on Arm successful, but it still relies on an outside partner, Qualcomm, for the hardware. And it just hasn’t happened, and there’s nothing Microsoft can do to speed that up. This experience will surely color what happens next, either through more Arm chipset suppliers and/or by Microsoft finally (really) making its own chipsets. It should be an interesting year or two either way on this front.

This is the year that Qualcomm puts up or shuts up

a93nckd2nakrhjw3 asks:

How optimistic are you that Qualcomm’s forthcoming Snapdragon X CPU will compare favorably to Apple Silicon while running Windows on Arm? Assuming that Qualcomm is able to solve the historical hardware shortcomings, what software hurdles remain for Microsoft to bring Windows on Arm to parity with Windows on x86?

I am not at all optimistic. I guess the way I’d put this is that I hope their performance and efficiency claims are proven with real-world hardware, but that years of experience force me to be pragmatic and I expect to be disappointed. I read all the rumors, and I’ve seen some positive signs. But we’ve been burned by this company so many times. Too many times. I just can’t trust them.

But let’s say it happens. The first Snapdragon X Elite-based PCs appear in April, or June, or whatever it is, and … they just work. Which I define as them offering comparable performance to Core i5/Core Ultra 5-based PCs but 25 to 50 percent better battery life. When Microsoft first revealed Windows 10 on Arm, my takeaway was that success for this platform meant that it would be boring. It would just work, and anyone who bought one of these PCs wouldn’t notice any differences or shortcomings. And that’s still true today: These things can’t trigger a gotcha moment.

It could happen. With one exception—software driver/utilities, which is becoming less important anyway—everything is there on the software side. What’s missing, as noted above, is compelling chipset hardware. And if they pull it off, the battery life advantage, if present, is what will help this platform gain acceptance and, perhaps less importantly, look less foolish when compared to Apple Silicon-based Macs.

If it doesn’t happen, and that’s a fair bet, we just enter that same old cycle as before, wondering if maybe some change in the coming year will make a difference. But I’m not sure Windows on Arm can sustain another defeat. This is getting silly.

Tiny Windows

Pakeha asks:

Do you have any thoughts on the “Tiny Windows OS” being an option for new users, such as seniors and such? Seniors usually find the complexity of Windows 11 Home too daunting for their confidence.

I’ve continued using Tiny11 on a near-daily basis since I wrote about it in mid-December, and I’ve had very few issues with it. There’s no photo-viewing app (Photos) installed by default, and some apps require WebView 2, a component that’s missing because there’s no Microsoft Edge (it can be installed separately). But it works normally for the most part, and as I keep going with it, I see fewer and fewer reasons to be concerned.

What it doesn’t do is address the need you mention here. Tiny11 isn’t simpler/easier than full Windows 11, it’s just smaller. And this lack of simplicity has been a concern for Windows for years. As people move to mobile platforms, they expect the same simplicity everywhere. But the only desktop platform that supplies that is Chrome OS. So one might argue that what Microsoft really needs is a Chrome OS competitor. And we all know they’ve tried and failed multiple times on that front—Windows RT, Windows 10 S/S mode, Windows 10X, etc.—and there are indications that some of the bloating we see in Edge today is tied to a new effort to create that “Edge OS/Edgebook” thing we’ve long speculated about.

Windows is absolutely too complex for people like my parents who are now in their late 70s and have no interest in deciphering that mess. (The Mac has the same issue, in my opinion, but that’s why iPads are such a great choice for this audience too.) The questions here are simple enough, I guess. Can Windows be made simpler while retaining its “Windows-ness,” meaning its full app compatibility? And is there any call for a Chrome OS product based on Edge that can run only web apps and Android apps (and then only from a third-rate store)?

The most likely outcome is that the answer to both questions is no, and that if Microsoft attempts either it will fail because this part of the market is already well-served by iPads and Chromebooks. For better or worse.

Monetizing Copilot

will asks;

With the launch of everything Copilot from Microsoft, outside of Windows Copilot almost all of the other Copilots have a monthly cost to use them. Do you think that Microsoft will look to monetize Windows Copilot at some point with enhanced features or functions? The basic stuff will be free, but anything further would need some level of a subscription.

This feels inevitable to me. But there are so many open questions here, and how/whether they do so will depend on variables we don’t yet understand.

But we can speculate.

Microsoft Copilot, the foundational layer behind Bing Chat, Bing Image Creator, Copilot in Microsoft 365, and the Copilot feature in Windows, is free. I don’t see that changing, in the sense that this thing, which is a web service, will almost certainly always have a free tier that is ad-supported. But I can see them charging a subscription fee that gives paying users precedence to accessing cloud resources, making their queries faster. And that it could add features to Copilot that are only available to paying users (Copilot Pro or whatever). Perhaps a hybrid functionality that combines local NPU with cloud-based AI for better performance and latency.

One thing I looked for when reviewing that NPU-powered HP Spectre x360 14 was which, if any, Windows AI features actually hit the NPU. And the answer was one: Only Windows Studio Effects (which requires an NPU) even used the NPU at all. So I could see updates to existing AI features (background removal in Photos and Paint, text recognition in Snipping Tool, etc.) that add NPU support when that is present, or new AI features that require and/or work better with an NPU (and/or GPU). And I could see these things not being free. That we all get some set of software or cloud-based AI features for free, but have to pay for priority access and/or hardware AI acceleration locally. Perhaps through a Microsoft 365 subscription, perhaps through a new subscription (Copilot Pro or whatever).

Short term, everything is up in the air. There’s a new Windows version coming that may or may not be called Windows 12. There’s no clear killer app for AI, no reason for anyone to bother buying a new PC, and how they fix that issue is a big question: Just waiting for something to show up didn’t work out well for HoloLens, for example, so it’s on Microsoft to seek out compelling solutions that require AI and justify the extra cost to customers. So this year will be very interesting, not just for AI generally, but for AI in Windows specifically. The Copilot story today is weak, horribly weak, and it’s just one of many features on the list of things that are public but only partially implemented (Windows Backup, Dev Home, so much more) in Windows 11 now. They need to finish the job in so many places.

Since Microsoft is betting so heavily on AI, something will happen in this space. It’s just not clear what.

Phone reviews

gartenspartan asks:

You used to review other cell phone brands like one plus. Do you think you’ll ever return to that or are you pretty much planning on moving back and forth between pixel and iPhone going forward?

I don’t what happened to OnePlus: I received most of their phones for review over several years and they just dropped off the face of the earth one day. My guess is that my primary contact there moved on and/or they were tired of my constant refrain of the phones being nearly perfect except for the cameras. I never asked.

But as with PC hardware, I’m open to whatever. The issue is tied to relationships and cost. I don’t have a formal reviewer relationship with any phone maker—Apple, Google, Samsung, whatever—and so I have to buy devices to review them. I tried for years to get that going with Samsung, and I was positive that was going to come together, but …then it disappeared. I don’t know why.

Anyway, when it comes to spending my own money, there are only two choices, Apple and Google. And so that’s how that happened. But Samsung is obviously a much bigger deal than Pixel, and I would love to review those devices. I’m not going to pay for them. (Though I have several times in the past.)

Woke up, fell out of bed, dragged a comb across my head

LordMartarius asks:

Have you ever thought about doing a “Day in the Life” video? I would love to see the various things you do during the day to create the content for the site.

No, but that is vaguely interesting, once I get past my qualms about appearing self-important. Maybe.

My workdays days are pretty standard. I work at my desk in the home office in the mornings and early afternoons. But I spend some time on laptops before that, in the mid-to-late afternoons, and between dinner and 8pm, which is basically TV time. I work on weekends and holidays too, but split the time up a bit differently. I tend to allocate time for specific projects, the photo consolidation/digital decluttering being the big one right now: I bet I spend two or more hours a day on that right now, every day. (This is what I do in the early mornings right now, plus evenings.) But I have spent an inordinate amount of time on the security stuff in the past month. That’s mostly in the afternoons and some evenings.

I also try to find time to walk and go to the gym. I have meetings and podcasts that interrupt days. I spend a lot of time resetting Windows on various PCs, testing different configurations. I have a Mac and a few Chromebooks that I work on from time-to-time, in part to keep them updated and in part to just refresh that experience and see what if anything has changed. I’m behind on the Linux stuff, but that is starting to wind up again. It’s hard finding the right combination

My big goal right now is to finish the photo stuff before we go to Mexico on February 3rd. I want to get back into the Tech Nostalgia series, which got pushed aside because of that. And I always have several or dozens of articles in various stages of completeness in my To-do folder. Some have been there for a long time (the Brave configuration one), and some are newer (an editorial about the need for Microsoft to adopt casting standards in Windows, for example). I dump things in there that are just ideas sometimes so I don’t forget them, etc. And I scan that folder every day.

Plus the book. Let’s not forget the book. And the site maintenance, which involves figuring out features to do or fix, dealing regularly throughout the day with comments, making sure new forum posts are live, and removing spammers as needed. My wife and I set aside time each month to go over the bills associated with the business, and to meet with the newsletter folks. And there are all these real-life things that pop up from time-to-time, as is the case with anyone. It’s difficult to even think about it all, lol. Maybe I need better/more formal time management.

As I write this, it’s almost noon, and I’ve been working on this all morning (which is typical for this post), aside from an hour or so on the photos and the podcast with Brad. After lunch, I am going to try and finish up a new chapter for the Windows 11 Field Guide, right a bit more of my Pixel 8 Pro review, and perhaps finish that Brave article I keep mentioning. If not, I bet I finish that over the weekend. But I will wind down around 5 pm, after which time my wife I will go out to eat and then probably have a music night, so that will be the end of work aside from stray emails.

I always see advice from writers to aspiring writers that the goal is to write 1,000 words per day. This article is over 5,000 words, and I write well over 40,000 words each week, not counting the book. (Last week, it was somehow 117,307 words, Grammarly tells me. Crazy.) So I guess a big chunk of my time is spent just doing that.

I’m not sure if any of this is interesting per se. It’s just what I do, I guess.