Ask Paul: April 10 ⭐️

Happy Friday! This is possibly the longest Ask Paul I’ve ever worked on, thanks to some great reader questions. So settle in, this is one for the ages.

? Going dark

jt5 asks:

Concerning Dark Mode- I recently started using dark mode more and more- on both Windows and Mac and am really frustrated with both for different reasons. It seems wildly inconsistent – some areas it works and others it does not. My bigger concern- and question for you is this: Have you heard of any plans to make it customizable? For example- instead of black use a dark grey? If you take the grey option in office, it turns the areas you would type in to white- with no option I can find to change it. It is not just office- but that is what I use the most these days.

Sadly, this is a baby steps thing: The inconsistency in Windows you correctly note is a big problem because you can work in Dark mode and some dialog opens in Light mode and it’s like a Klieg light blasting into your eyes suddenly at night. (Mac is at least better in that regard.) Microsoft is working on this and maybe the 2026 “pain points” focus will help. But that’s the first step.

The second step is automatic Dark/Light mode switching, which the Mac does support. You can at least add this with a PowerToys utility called Light Switch, but that should be part of the OS.

But then we get to the third step, which is what you’re asking for: A more granular Dark mode configuration capability. I see this sort of thing in individual apps–Affinity, for example–but the ability to configure this type of thing at the system level would be ideal. And no, I’m not aware of any way to do that, especially in Windows. I think the problem is that the base capability is so poorly done and isn’t used everywhere anyway.

I do know from my .NETpad/WinUIpad apps that you can control the theme/color of any individual component, including the text box you write in, and so it’s possible for this to work on an app-by-app basis. One of the features I had in .NETpad before the 2024/2025 rewrites was a theme feature so that the text box could be whatever combination of text color/background; I liked using a light gray background instead of white or whatever. I am curious apps as powerful as those in Office don’t offer that, but I guess they are complex as Windows in some ways and it may just be a legacy issue.

I haven’t done this, so I’m not sure what this would even do, but it may be worth looking at the accessibility features in Windows and/or Office to see if there are any workarounds in there. The few times I’ve looked at the high contrast features haven’t been good experiences though.

?️ Retro Windows

NickTech asks:

Paul, I’ve seen you explore the development of Windows and other platforms in your Switcher and Programming Windows series. Have you taken a retro look back at Windows in day to day use or for use with software of the time (like retro gaming with the hits of the day on XP)?

One of my focus months this year will be retro computing and I was just looking at this, sort of. (I dug up some old Windows 2000 and XP screenshots for an episode of Hands-On Windows, which you can see below, and made the point that Windows 11 is to Windows 10 from a UX perspective as was Windows XP to Windows 2000.)

I have a couple of ideas here. One is based around playing old MS-DOS games, where you can use an environment like DOSBox. One is virtualizing old Windows versions, which I did while writing that Programming Windows series. But I am also curious about bringing up an old Windows version–probably 2000 or XP–on physical hardware; I’m not sure this would work well on modern laptops or PCs, but I do wonder about some older hardware I may have lying around. I will check on that when I get back to Pennsylvania in May.

Interestingly, Linux with the Steam/Proton stuff might be a way to play older games as well. I will look at that too.

And then there are all the hardware emulators and new versions of old computers like the Commodore 64 Ultimate. I mentioned that I had preordered a “new” Amiga 1200, and I am curiously excited about that. I feel like software emulators are generally fine, and as I write more in the Tech Nostalgia series this year I will try to point to those where appropriate.

? Unpublished

helix2301 asks:

What do you think about Microsoft discontinuing Publisher after all these years? I know Canvas and all those are available and good. What do you think Microsoft reasons that because there is not enough usage, or are there better alternatives out there?

My guess is that desktop publishing is no longer a mainstream activity with a large enough customer base. If you think back to the late 1990s when the web was exploding, there was a chance that apps like Word and Publisher could have become how we published content to the web, and there were efforts to that effect. But both of these apps are inherently paper-based, meaning that the goal was always to get a document to the point where you could print it. And that’s just not a big deal for most people anymore.

So there are professional desktop publishing apps that serve that market. And then things like Canva or whatever that are more oriented to digital publishing but can be used for light paper-based work, too, I guess. Sort of like PrintShop back in the day maybe.

Anyway, I have to assume that Publisher just doesn’t have enough active users to accommodate the expense of keeping it up-to-date from a security perspective.

? Finish the job

ryguy asks:

I know this isn’t a new thought, but I’m just so gobsmacked by how badly Microsoft is fumbling things.

Two thoughts up front.

Brad and I recorded First Ring Daily just before I saw this and the idea he put in my head, not for the first time, is the sheer number of stories of times when Microsoft was onto something, fumbled it, and then it went on to become a big deal elsewhere. He mentioned OneNote and how it could have been Notion, for example. (And now, with Markdown powering all these AI chatbots and agents, how that is driving yet another change in that space.)

Secondly, Microsoft has a rich history of not finishing the job. This is basically a fit and finish issue, if you think about it, similar to the Dark mode issues noted up top. There’s a lot of half assing in the Microsoft ecosystem and it’s likely a cultural issue tied to the company’s founders and the way it does things and rewards employees. There’s a lot of excitement about something new happening but once that product or service ships, the good people move on to the next big thing and nothing ever gets completed. No company is perfect in this regard, I suppose, but Apple typically does better at this than Microsoft, for example.

The main reason to have M365 Copilot is access to the MS Graph, which has all my info. Right? Except when I ask it to query my calendar, it can only see my primary calendar. So when I say, please look at my Travel (secondary) calendar, or my Family (secondary) calendar, it just says “nope you’ve got nothing going on this week”. Super helpful! Meanwhile, Claude now has integration with Fantastical (my preferred calendar app on Mac), and it can flawlessly read and write to all of my calendars, even those that aren’t hosted in M365.

Yes. This is a classic, and there are probably many examples like this if you were to compare Claude’s capabilities within Microsoft 365 to those offered by Microsoft itself (Copilot). Richard Campbell speaks to people at Microsoft all the time for his Run As Radio podcast and he was astonished by how quickly Anthropic was adding features to Microsoft 365 compared to Microsoft.

To turn that one on its head, one of the big complaints about Microsoft, Copilot, and AI is that it’s all happening too fast, but the bigger issue is the chaotic, non-strategic way it’s happening. To bring up the cultural issue again, Microsoft often thinks too big when it comes to solving problems, so it looks at adding AI to 365 at a high level and of course this requires a platform, which is big and messy and time consuming. And these two things—the speed/chaos of AI, the need to make big platforms–don’t mesh well. Meanwhile, Anthropic is small, moves quickly, and it decisively moves to add features it knows people will want and use. It’s kind of Apple-like, in a way, but without years of deliberation. Just solve the problem.

I have a Surface Pro (Arm version) and a Macbook Air. I like them both. I have a top-tier Windows 365 Cloud instance, that I can use on the Mac, and can sync to my Surface too. Its a good service! However, there are features on my “Copilot Plus PC” Surface, that are NOT available on my Windows 365 Cloud instance. Why? Apparently because a two year old Snapdragon with NPU is more capable than a Virtual Machine running in Azure with dedicated memory/gpus/etc? Huh?

This is a big one for me.

I addressed this to some degree last summer in Microsoft, It’s Time to Set Copilot+ PC Free (Premium), but now we’re closing in on the second anniversary of the Copilot+ PC announcement, and I was just looking into this again: What was announced (as unique Copilot+ PC functionality) at the time, what’s been added since, and why the hell none of this (well, most of this) has not come to other PCs, including desktops and laptops with dedicated GPUs that could easily handle these activities (and should). It makes no sense.

But I had not considered the Windows 365 Cloud instance issue, which is basically the same thing: How is it possible that Microsoft doesn’t offer NPU/GPU-based Copilot+ PC capabilities in Windows 365? And the answer, sadly, is that inconsistency is Microsoft’s middle name. Or M.O. Or however you want to say it.

I don’t have an answer to how this could/should work, just some ideas that are probably based more on fairness to customers than on Microsoft’s underlying business model. But these are both symptoms of the same problem, for sure. And it’s not a new problem.

Will they ever figure this out? Will they ever take even baby steps to fix any of this? I hope so. But history suggests otherwise, sorry.

? Rekindling our unrealistic expecations

justme asks:

The BBC recently reported that Amazon will stop supporting some older Kindles. I understand that some of the devices affected are old in electronics terms, but I have always seen them as relatively simple devices needing next to zero maintenance. Apart from Amazon perhaps not being able to collect as much telemetry or push some new AI to/from older devices, is there anything to be made of this decision or is it just the natural age-out of electronics? My own Kindle experience is the app on an IPad Mini, which does the trick for me.

It’s impossible to see this story and not think about the drama that unfolded when Sonos tried to stop supporting its oldest smart speakers a few years back and the user base exploded, triggering a change to the support policy and, more importantly, a major change in the way it how creates its devices to ensure they will never be bricked down the road. One of the similarities is the age of the devices: The Kindles that Amazon is dropping support for are 13 years old or older (dating back to the original 2007 model). One of the differences is pricing: Most Sonos products are expensive whereas most Kindles are not. And that’s the thing.

You have a Kindle from 2007 to 2012, or whatever. It still works, you’ve bought whatever e-books from Amazon over the years and maybe you get books for free from the library. And now Amazon says that it is ending support for this device. To me, the value of this thing is the content and of course the first thing I do is look at Amazon.com and see that you can get a new Kindle for as little as $110 before any sale. And Amazon is offering these customers 20 percent off a new Kindle, plus a $20 coupon for content. I don’t understand the complaint.

But let’s say you want to keep that old Kindle and keep using it. You can: You just have to download e-books on your PC or Mac and copy them over a USB cable to the device. You just can’t access the store anymore on the device itself. Here again, I don’t understand the complaint. Chances are that anyone still using that old of a Kindle was doing things that way anyway.

I have seen headlines claiming that people are looking at Kindle alternatives because of this change. But I look at this and think, wow, I can’t believe how long they supported those devices and the situation moving forward seems acceptable. Spend $110 over a decade later and just keep going. (Really, $70 to $90, but whatever, maybe you splurge and get a more expensive model.)

We’re going to run into this kind of thing a lot in personal computing electronics. Maybe there is a way forward that makes sense, like the way Sonos makes speakers now. Maybe this is just inevitable with certain products. No one is using a smartphone from 2007 to 2012 these days, or a PC or Mac from that era.

I feel weird defending anything Amazon does. But in this one case, I don’t see any major issues. This feels reasonable to me, but perhaps I’m missing something.

?️ Stripping away the wallpaper

anderb asks:

If full backwards app compatibility was no longer going to be a requirement in a future Windows version, what frameworks/technologies/APIs would you like to see removed from Windows?

Hm.

This is a tough one. I don’t have a particular problem with specific technologies in Windows, legacy or otherwise, and there is a case to be made that the success of Windows in some ways is tied to its backward compatibility promises. This is especially important to businesses, but of course that focus on Microsoft’s part has its problems, too (for individuals). You could argue that it was the first step towards enshittifying Windows.

The other approach can be seen at Apple, which has traditionally been more aggressive about killing off legacy technology. Apple performed two major platform shifts with the Mac in the modern era, the move from PowerPC to Intel and then the move to Apple Silicon, and in both cases it offered a compatibility layer, Rosetta, but only temporarily. My understanding is that in both cases there wasn’t so much a hard stop on the time frame for offering that. But as more and more apps moved forward to the new architecture, there was less need for Rosetta, and it was eventually dropped. (Apple will drop Rosetta 2, which lets Apple Silicon Macs run Intel software, in late 2027, it seems.)

Maybe there’s a happy middle ground between these two approaches. Or, maybe Apple’s approach is the better of the two. Even if you appreciate the backwards compatibility stuff–which people do, based on the complaints about Windows 11 on Arm that make no sense to me–there’s security to consider. All this legacy code sitting in Windows is like debt. It’s security vulnerabilities waiting to be exploited.

I do feel like Windows 11 on Arm is in a good place with regard to this. But more generally, we still ship things like the classic .NET Framework in Windows. Is the Visual Basic 3 runtime still in Windows? I’m almost afraid to look. (Plus, Windows Search is terrible.)

This isn’t about bloat to me. But security/technology debt is real, and Windows is full of it. That this is tied to two or three of the above conversations is an interesting coincidence. I don’t know that I can point to a thing—.NET or whatever–and say that this should not be in Windows. But I have to assume that a big part of the Windows Resiliency Initiative is a code review that includes seeing which bits should be left behind.

I’m curious if you or anyone else has ideas about what should be removed. Most of my issues are UI/UX-related, I think.

? DOOMed to repeat history

madthinus asks:

At the time of release, you did seem to enjoy Doom 2016, you had some harsh words for it. Then you mentioned it in your Linux article as using it to test and I was curious how you feel about it a decade on.

When DOOM 2016 first came out, I wasn’t impressed: It didn’t feel anything like the classic DOOM/DOOM II, and the movement to me was like skating on ice, not natural or normal or whatever. But I’ve since played through quite a bit of DOOM 2016 and DOOM Eternal, mostly on PC, and though I haven’t finished either yet, I will. DOOM Eternal is an improvement, to my mind. DOOM: The Dark Ages, the latest game, is a strange outlier that doesn’t seem to have much to do with the previous two games. I am not sure why it’s the way it is, but I will pick it up again eventually as well.

The one thing that does tie DOOM 2016 and DOOM Eternal to the OG games is that they are, at heart, single player campaigns that focus mostly on fast action and shooting and not on the plot, as with DOOM 3 (which I really liked; similarly, I liked Quake 2 and 4, which were also plot-driven single player experiences). This is maybe the problem, it’s just the same thing over and over again, and you’re only playing against game-controlled enemies. The newer games do have some kind of multiplayer support as well, but I’m not super interested in that for whatever reasons.

In any event, I like that these games keep happening. It’s like any franchise, there will be good entries and not so good entries. But the more the merrier.

? It’s not about us

OldITPro2000 asks:

Tom Warren had a piece out today on how Microsoft is making the reported improvements in Windows 11 as a competitive response to the recent MacBook Neo from Apple. I was really thinking it was all the general negativity around Windows that seemed to be reaching critical mass that made them respond; it reminded me of the Xbox 360 red ring issue that did hit critical mass right around the time Microsoft said they would take action to make things right.

The improvements happening in Windows were triggered before the MacBook Neo was ever a money grab in Tim Cook’s eye, and they weren’t driven by any worries about consumers and whatever computers they might or might not buy. They were driven by the enterprise, its biggest corporate customers. And those customers are not going to buy MacBook Neos. They’re going to complain about yet another Windows Update borking something.

Cory Doctorow of enshittification fame has said something that really resonates with me, that there is nothing we can do as individuals to convince any Big Tech company or other abuser to change their ways: We can’t vote with our wallet. This is true, and it’s true with Windows especially. But what we can do as individuals is make decisions that are right for us. And if using a MacBook Neo is that decision, God love you, enjoy that. But enterprises can vote with their wallets, and while none of them are switching platforms, almost literally, they spend enough money on Microsoft every month that they can institute change. That’s what’s happening here. Windows coasted for too long and now the bill is due. So Microsoft is taking action. And we will all benefit to some degree, which is great.

Beyond that, it made me think deeper about how, generally speaking, Microsoft over its existence seems to spend far more time listening to its competitors and not its customers. One can argue that they do listen to enterprise customers, but even in that case they always seem to be secondary to competitors. Do you agree that the planned Windows improvements were only from perceived competitive pressure? I remember you writing a few weeks ago that the Windows improvements likely started because enough enterprise customers complained.

When I was writing Programming Windows, which became the book Windows Everywhere, it occurred to me in reviewing the history that the changes Microsoft made to Windows over time were driven by one of two things: Some internal desire to move the needle forward, so a sort of proactive decision to make (hopefully) positive change, or in reaction to things happening externally, be they competitors or just more general industry-based changes. Either way, evolution occurs.

Today, Windows is in a weird place. PCs are not the future of computing, we all understand that, there’s no big change coming that will reverse all the trends in mobile, cloud, and AI. But it’s still a big market, especially in businesses, and Windows still dominates there. If history shows us anything when it comes to Microsoft and dominance, it’s that complacency occurs. Internet Explorer was so dominant in the late 1990s and early 2000s that Microsoft basically pivoted that team to focus on what became .NET–a proprietary “embrace and extend” moment for web technologies–and dropped the ball on its browser for five years. (IE 6 came out in 2001, but IE 7 didn’t occur until 2006) This is what happened with Windows: Microsoft was distracted by its focus on cloud computing for several years and then again by AI more recently.

The silver lining to Microsoft’s AI focus, sort of, was that you could see it was paying attention to Windows again. This makes sense because Windows as a platform is the right place (or, a right place, so to speak) to implement AI. But there were negative impacts there, too, obviously: Copilot icons and AI everywhere, whether it made sense or not. The chaotic and constant feature updates every month. The CFR nonsense, where no two PCs, otherwise updated identically, even have the same features. And so on. This is what Microsoft is finally addressing. And, yes, it’s because of enterprise customers. Not because we as enthusiasts, yelling into the wind for the past decade, want the Taskbar on the side of the screen or don’t like the inconsistencies in Dark mode or whatever.

But again, it’s all good. We’ll all benefit, no matter the impetus.

As an aside, I’m inclined to agree with you. I don’t view the MacBook Neo as enough of a competitive threat to Windows to make Microsoft pay attention. If anything maybe it would make them complain to some hardware OEMs about price-to-value and hardware quality but Windows is Windows.

I feel like there’s a future where the MacBook Neo makes more sense. Adding a 16 GB RAM upgrade feels obvious. Keyboard backlighting? Duh. But the thing is, this is not the volume device market in personal computing. Apple already controls half the market for smartphones and it owns tablets. Improving share in computers is fine. But it’s the overall ecosystem that matters to these companies, the monthly spend on subscription services. Microsoft has that with businesses, and Apple has that with consumers. I don’t see this changing. And businesses, almost unilaterally, use Windows. I don’t see this changing either, no matter how good the Mac is. And the Mac is excellent with or without the Neo.

? Wake up the captain, we just clipped an iceberg

OldITPro2000 asks:

Bloomberg has a piece out today on Amy Hood. I’m wondering if you’ve read it and if so what your thoughts were. Back to my Xbox 360 example above the story was Peter Moore going to Steve Ballmer and explaining the problem to him, Ballmer asked how much, got the estimate from Moore and Ballmer says “do it.”

Yes.

So, for the past year, maybe year and a half, I’ve come around to the notion that it’s Amy Hood, and not Satya Nadella, that really runs Microsoft. This is based on a growing list of evidence. It’s fascinating how they both speak on the quarterly earnings calls, and Hood is clearly steering that, but it makes some sense as she’s the CFO. You can see it in Satya giving up more of its CEO duties to others so he can focus on what was, in the Bill Gates era, a Chief Software Architect role. And then there’s this notion about what Microsoft really is in this AI era, more of a real estate/physical infrastructure business, if that makes sense, than a maker of software. In that world, it’s money that matters, not products that make money.

I did read the Bloomberg article you mention, and it was more complimentary to Hood than I expected and for a reason I never saw coming: She’s being credited with seeing the spending binge I and others have complained about for many months if not years, and she apparently put the brakes on it. This is interesting timing because everyone is questioning the AI spending across the industry, finally, and Amazon CEO Andy Jassy just made an impassioned defense of his company’s insane spending levels. But here’s Microsoft, which kicked off both the AI wave and the AI spending wave, and she’s being credited for being … conversative?

Whatever anyone’s opinions on Microsoft, its executives, its AI push, or whatever else, one thing is clear: This era we’re in now is more about Amy Hood and financials than it is about Satya Nadella and engineering. And the company’s success hinges on her being successful more than it does on him being successful. We’re in the Amy Hood era.

I feel like today Satya Nadella would say, “Let’s check with Amy first.” Actually, no, I feel like she’d be in the room to start with. Actually, no, wait, I feel like Nadella wouldn’t be in the room at all.

There it is. Same. 🙂

? To the stars

christianwilson asks:

Starfield got a nice new update this week and was also launched on PS5. Have you played it? If not, now might be the best time to grav jump. Starfrield seems like it might be your kind of game and curious of your thoughts. I’ve been playing it since day one and love it. It’s like comfort food for me.

I’ve never played Starfield, sorry. I’m not even sure why, but I’ve never really gotten into any open world role-playing games. I will take a look, thanks.

?‍? Thanks, I’m just browsing

richardbottiglieri asks:

You had mentioned somewhere that you were using Firefox more. Curious how you are finding it as your daily driver and how it compares to using Brave or other Chromium-based browsers.

Mozilla’s sharpening focus on Firefox over the past several months has rekindled my love of the organization and this product and everything they stand for. And so, yes, I’ve been using Firefox a lot lately, and everywhere, across Windows, Mac, Linux, iPhone, iPad, and my Pixel phones. Mostly it’s been a good experience overall, and though I have to move from browser to browser just to keep up on all the changes, Firefox works well.

Mostly. The one issue I’ve had, repeatedly, is that it gobbles RAM on the Mac especially, for reasons I can’t explain. And this has triggered performance problems I’ve never seen before on my MacBook Air, to the point where I’ve had to force shutdown the system twice so far because it couldn’t recover. I’m hoping that its Anthropic bug-finding partnership will get ahead of this, but I have not had this problem on Windows or Linux at least.

There are all kinds of thoughts around web browsers. But a few that are perhaps relevant here:

Almost any web browser can be protected against tracking and bad behaviors if you use the right extensions.

You don’t have to use the same browser everywhere. You could use Firefox on Windows and Safari on an iPhone, or whatever. It’s fine.

Moving between browsers, or using multiple browsers, is likewise easy enough. I always have at least three different web browsers on every computer and device.

It’s healthy to try new things. A certain browser may speak to certain people’s sensibilities, and it may turn off others. But you have to experiment to find that out. I keep checking out different browsers because I have to, work-wise. But I feel like I would do that regardless.

Anyway. I do like Firefox. I’ll keep using it. And I’m very curious to see how they improve it in the coming months. I hope they can win back some usage share: We need Mozilla.

⚛️ Going all-in on Little Tech

anoldamigauser asks:

Since you have two NAS devices, geographically isolated, have you considered simply dropping OneDrive, Google Drive, and iCloud?

Yes. But I’m in a weird spot for all kinds of reasons, divided between work and personal.

On the work side, I sort of have to keep up with what Microsoft and Google are doing, and so I have whatever work and consumer subscriptions from both. And those have OneDrive and Google Drive cloud storage. It makes sense to use that, if only in a secondary backup-type way. But I am using OneDrive now, Windows and Mac (and Linux) for testing purposes. I write books, etc.

On the personal side, my wife still uses OneDrive as part of a Microsoft 365 Family subscription that’s been paid out for several years. And our kids and her father are on there too. So even though I don’t use Office per se, she does, and others do.

Synology Drive is my primary data storage, though. My source of truth or whatever. And it’s not difficult to imagine a future in which it is all I use. For now, I have to keep a toe in this other stuff. Which also works fine, honestly.

Have you tested Proton Drive yet? This seems to be another of their products that just works as one would expect (except for Linux where it is cloud and web only.)

To date, I haven’t used it that much because of a weirdism that’s unique to the Proton consumer plans: They just haven’t offered that much storage. Where Microsoft has been on 1 TB for longer than I can remember, Google was on 2 TB for many years and it just changed to an incredible 5 TB. But Proton Drive gives you only 512 GB of storage through the Unlimited plan I’m on. And that is ridiculous.

This is starting to change. Proton Duo, which is essentially Unlimited but for two people, now offers 2 TB of storage, which is acceptable. And Proton just announced Workspace and Meet for businesses; Workspace Standard provides 1 TB of storage per user while Premium gives you 3 TB. And so I am thinking about this. I mentioned last week, I think, that I have brought a custom domain into Proton and I’m testing Mail now more than I’ve ever done. I could see a combination of Synology and Proton in my future as well.

With the exception of Linux, which has its own issues, switching to another platform seems to be exchanging one crummy corporation for another. We may each consider one or another to be less bad, but they are all bad. Other than personal preference, is there anything in your workflow that would really prevent a switch? It seems you have already moved on from most big tech software solutions.

All these things vary by person, I guess. I don’t have a concrete plan to eliminate Big Tech from my life per se, but I guess I do think of it as a general plan to lessen my exposure to, and reliance on, Big Tech. And to do that over time. As I look at various Linux distributions, and at other platforms like Chrome OS, Chrome OS Flex, the Mac, the iPad, and so on, I am looking to see which of the apps, services, and workflows I use are available on each, and when they’re not, what it looks like to find alternatives.

Here’s a weird (or maybe just very specific) example. I was working on the Chromebook I recently bought yesterday and wanted to see what a basic workflow I do every day–write an article, download an image, edit that image (for aspect ratio and size, usually), and then post the results online–looked like. Chrome OS is interesting for all kinds of reasons, but there is the full Chrome web browser for web apps, which can be good, there are Linux apps, and there are Android apps. But there are also just apps in Chrome OS, things that are bundled with the system and can be interesting. One, which is very simple, is called Gallery and, among other things, it can edit images. And … that worked great for my needs even though it is nothing like the app (Affinity) that I use on Windows and Mac. I downloaded an image, resized it, cropped it, and exported it as a JPEG, and that workflow worked really well. I found myself wanting that app elsewhere. Oof.

Gallery can’t replace Affinity. I also use that app for more complex work that occurs less frequently, and I use advanced features like layers and whatever else. But from a day-to-day perspective, Gallery is nice. It does work for that job. And then I guess I could use Photopea or Photoshop Web, or whatever, for more complex jobs. It goes on and on.

But that’s the thing. We all have our specific needs across apps, services, and workflows, and whatever preferences. And whatever ability to change or even desire to change. So it will vary for all of us. And things change: If this career comes to close and I retire or whatever, which never really happens for me, but let’s dream for a moment, then my needs will change. And maybe things that don’t work for me now will be fine then. Or even better.

When I think about what it is that I do everyday and the various platforms I could do that all on, there are a few conclusions. I have all kinds of choices. I have a lot of hardware. I can mix and match, and I can experiment. I think I could do the basics of my job, so to speak, on almost any of them. But I also know I don’t have to choose just one. There is no emergency forcing me to decide. This is somewhat freeing. But I also have preferences, of course. Familiarity matters, etc.

Could I eliminate Big Tech from my life? Not really: Smartphones are 100 percent Big Tech right now, though I find Jolla interesting. I could move to Linux on PCs, I guess, and there are good choices there. I don’t see a world in which I’m not using at least some Big Tech services, entertainment or whatever, but then I could exist on the content I have on the NASes for quite a while too. I think it will be a mix. The shift is toward Little Tech, for me. But Big Tech will likely always be in there somewhere.

? Home audio

jrzoomer asks:

Paul can you give a rundown of your home audio setup? Do you play music from your PC? Which speakers and/or sub and/or amplifier do you use? Are you using Sonos? Or using some of the home devices (google home/Echo) for some of this?

In Pennsylvania, we have numerous Sonos speakers, including two big Play:5 speakers and two smaller IKEA Symfonisk Wi-Fi Bookshelf Speakers (Sonos compatible, basically Sonos Ones) in the living room for music. But we also have two Apple HomePods attached to the Apple TV, so we go back and forth. We use the Sonos speakers for music. We usually use the HomePods for TV. But we also mix and match: The HomePods and Sonos speakers can all be used together through AirPlay for music or TV/movies. It’s not Dolby Atmos, but it does sound fantastic. We have several other Sonos speakers throughout the home, too, in the bedroom, bathroom, and my home office. When we have people over for New Year’s Eve or whatever, I move a Sonos Move into the dining room and use that together with the speakers in the living room.

In Mexico City, we have a much smaller place, so there’s less stuff. We also use Apple TV here, so we have two smaller HomePod Mini speakers for TV and music in the living room. They’re fine, but I kind of wish they were bigger/louder, so I may upgrade in time. We also have a JBL Charge 5 Bluetooth speaker here that I use in the bathroom for podcasts, audiobooks, YouTube videos/Music, or whatever when shaving and showering, and we use it out on the balcony too. (We originally had two JBLs here in a stereo pair, but it was non-optimal, so I brought one back to PA so we can take it with us out in the world.)

Because of all the Sonos stuff and HomePods, using Apple devices for music works better than Android: AirPlay is amazing. I wish this worked the same way on both sides, but Google and Sonos have their ongoing legal issues. I’m not sure about the future: I like where Sonos is now, after years of terribleness, but I also like the idea of something more open that may or may not involve some modern version of one or more set-top boxes and then non-powered “dumb” speakers. But … what we have is paid for and it works, and you don’t jump out of a perfectly working airplane. So we’ll see.

?️ Going native

lindhartsen

Between the news of a team at Microsoft looking to make “native apps” for Windows 11 and some of the color commentary on Windows Weekly I’m curious what you think prevents Microsoft from making all of these various frameworks feel faster on Windows. Is it truly a technical problem they can’t overcome, or is there a problem internally at Microsoft that can’t get the teams working on these frameworks to somehow work closer with Windows to reduce the performance gap?

It’s fair to say I’m borderline obsessed with this topic right now. Microsoft announces it’s going to fix the pain points in Windows. But it also doesn’t address most of the actual enshittification in Windows, which is troubling. And then it (foolishly, to my mind) heads further into fan service territory by claiming it’s going to make “100 percent native” apps for Windows. And for the love of God, Microsoft.

I think most would agree that, at any point in time, Microsoft releases whatever version of Windows and then whatever accompanying developer tools and experiences, and that what it should always do is use those technologies itself. That the apps in Windows should inspire developers about what’s possible. And it should eat its own dog food, in other words. But it has rarely done this. And in some cases, it has completely ignored the technologies it was giving to developers, as it did tragically with the Windows Presentation Foundation (WPF).

But here’s the thing. In a world in which Windows is not the center of personal computing, where asking developers to create Windows apps or even to focus for some reason on Windows apps, that old approach doesn’t make sense. With some random exceptions, any developer would be crazy to create native Windows apps, apps that only run on this one platform. Many apps, maybe even most apps, should be written to cross-platform frameworks, ideally web-based, so that they can run anywhere. And while the Windows app platform situation is beyond f’ed up these days, with too many choices to count, Windows is a melting pot of sorts where everything just works. No matter which technologies a developer wants to use, they can make apps for Windows.

To me, this is a strength. So when Microsoft talks about native apps, and certain enthusiasts cheer this on like it makes any sense at all, I kind of die a little inside. This doesn’t make any sense, not even for Microsoft internally. Just ask the Office team: It adopted web technologies broadly for the same reason others do, because it works everywhere. And those apps need to be everywhere their customers are.

What Microsoft should be doing is making sure that web apps, and other kinds of apps, work as well on Windows as is technically possible, and that they look and feel native. It should evaluate every surface, every experience, in Windows and then just use the thing that is best for that scenario. It should be optimizing, not picking a single way forward. We have many ways forward, and Windows should use them all where they make the most sense.

Consider these two examples of things Microsoft has promised to fix this year: The Start menu and File Explorer. Both are hybrid apps/experiences that consist of legacy and modern technologies (Win32 and WinUI for File Explorer, and WinUI and JavaScript/React Native for Start). Both have performance issues, I guess, though I only see that in File Explorer personally. And by performance, I think I really mean latency. But whatever.

So Microsoft is going to fix Start by removing web technologies. Presumably, this means that the code will all be modern and (sort of) native, meaning WinUI 3/Windows App SDK. Great. But I attribute the performance problems in File Explorer, which I experience daily, to Win UI 3/Windows App SDK. And I do that because I know you can go back to earlier (non-WinU 3) versions of File Explorer using ExplorerPatcher and that when you do so, those performance issues disappear.

Here’s what else I know: Last year, Microsoft promised to fix the Windows App SDK, and I want to be super-clear here, as the Windows App SDK is a complete clusterf$%k. That is not happening. The person leading that push quietly left Microsoft months ago, I assume because that job is impossible to pull off successfully. And the Windows App SDK that this new team inside Windows is going to use to create “100 percent native apps” is just twisting in the wind.

Basically, I don’t know what the answer is here. I don’t see Microsoft creating a new native Windows SDK/API. I don’t see further modernization for WPF, which is .NET-based and superior to the Windows App SDK. I don’t see it finishing the job on the Windows App SDK, which still lacks so much functionality that its predecessors had. And I don’t see it fixing the Windows App SDK in any meaningful way. I’m curious why it didn’t just embrace React Native and web technologies fully. This is a mistake it’s been making since the late 1990s, really. And in an era where Windows really doesn’t matter as a top-tier target for app developers, not embracing cross-platform technologies, web technologies, feels like a fatal mistake. This could exacerbate the decline, not help stop it.

The amount of work that needs to occur here, no matter which direction they choose, is just too daunting to not be problematic. I’m curious to see what comes of this, in Windows 11 version 26H2 later this year and beyond. My guess? This silly team will produce a handful of non-essential PowerToys-level apps that will be modern but not truly native and it won’t matter in the slightest. And then it will disappear quietly. The platform-level stuff, the low-level improvements that will float all boats, is so much more important. No matter which tech stacks they use and where.