Ask Paul: June 10 (Premium)

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

Fit and finish

hastin asks:

Paul: Why is updating the loading circle so hard for Microsoft? I noticed that even on a clean install of 22H2, there’s even more places where it seems they’ve only half-replaced the circling dots.

What you’re really asking here is why can’t Microsoft ever finish the job? This company—and the Windows group in particular—has a rich history of starting but not finishing things.

I’ve actually written about this extensively over the years and coincidentally came across something I had written called “In Praise of Finishing a Job” years ago while researching events for the Programming Windows series. I was going to paraphrase what I wrote, but I actually made the point well enough back in 2012:

“Microsoft culture has rewarded shipping a product or product version above all else, and if that product happens to be attached to some huge revenue stream, all the better. The problem is that no software products and services are perfect and bug-free. And the dark half of that Microsoft culture is that there’s little if any reward for those whose job it is to set things right.”

“In the old days, you could see this dichotomy most clearly in how Microsoft developed and then serviced Windows. The folks who created a new Windows version were heroes within the company and could do no wrong. But once that Windows version actually shipped—the process we all know as its release to manufacturing, or RTM—the code was unceremoniously handed off to the B-team, the folks responsible for servicing Windows. These folks weren’t heroes at all. They were almost non-existent as far as the rest of the company was concerned.”

This culture persists to this day. And you can see the evidence of that all over Windows 10 and Windows 11. The slow drip-drip of adding Control Panel functionality to Settings but never really getting rid of Control Panel, for example. Or, more recently, in Windows 11, the internal self-celebration of the new Start menu and Taskbar without any acknowledgment or caring that these new interfaces broke functionality that many millions of people were using. This is Microsoft at a high level: Favor the new over finishing the job. Every time Microsoft adds some pointless new feature to Windows 11 in a Dev build, I ask myself why it isn’t spending that amount of effort on fit and finish. But they never have. And they never will.

“There’s no real reward for tweaking an existing product, adding that missing feature, or fixing the tiny incomplete bits,” I wrote ten years ago. “If the past is any indication, work on Windows 8’s successor began months ago, and you have to think that the A-team that foisted Metro and the Start screen on us has moved right along to this next milestone and will have little to do with fixing the previous release. If that’s true, Microsoft hasn’t learned a thing. I’d like to see the team responsible for this mess spend the next three years cleaning it up. It’s time the creators at Microsoft took a step back and finished the job.”

That’s still true today. Sadly. Also true:

“It’s time for a deeper cultural shift inside the company … Microsoft needs to be a company that gets the fit and finish right. It needs to embrace and reward the finishers before it starts creating again.”

It will never happen. Sorry.

Why?

SherlockHolmes asks:

So Windows 11 22H2 is out and it is in many ways disappointing, the big question is: Why does Microsoft handle Windows 11 the way they do? They must know its not looking good for their product. Do they dont care? When the goal is to get the mainstream on Windows 11 why just give the users a minimum of what they want? Im in a lucky position that all four of the PCs in my household got a license of Windows 10 Enterprise 1809 LTSC so im out at least until 2029 on the Windows 11 train. But if you could forsee the next two or so years, will there be a reason coming to make it sane to upgrade? Thanks.

This is an interesting follow-up to the previous question. And to continue that discussion, my educated guess is that it goes like this: imagine there is some person working on Windows and they really want what we all want: for Microsoft to “fix” Windows 11, in this case, whether it’s from a basic fit and finish perspective or a more complicated project like adding back regressed functionality to the Taskbar or whatever. Without getting into how or whether any individual or group could even do such a thing, let’s just pretend: you’re doing it.

What happens is that all around you, the people who aren’t doing this are being rewarded with raises and promotions. They’re moving into exciting new groups working on exciting new projects and their careers are skyrocketing. Meanwhile, you’re sending memos about the Control Panel or right-clicking the Taskbar or whatever, and people aren’t just ignoring you, they’re ostracizing you. You’re left out of important decisions and meetings. You’re passed over for promotions. Soon, you’re the village crank, complaining to the air. No one cares. They’ve all moved on.

This is the cultural thing I discussed above. There are obviously good, caring, and smart people at Microsoft who see these same issues and want to do the right thing. But Microsoft doesn’t reward that kind of behavior. It rewards the new and interesting. It rewards form over function. There is no glory in fixing bugs or righting wrongs. And there is a ton of telemetry data that decision makers can shove in your face to prove that what you’re doing is not just pointless but is holding Windows and Microsoft back. No one right-clicks the Taskbar, idiot, we have the data. Or whatever.

This is all imaginary, I guess. But I feel like most would agree that it’s something like that. That it’s close.

As for the future, I think we can look at Windows 11 22H2 as a guide: this is a release that has dozens of what I’ll call low- and medium-level updates, nothing major, and they are about 90+ percent new things and less than 10 percent fixes for previous issues (that minor Start menu layout thing, for example) and additions that were promised previously but never delivered. Of those 90-ish percent new features, most are not particularly interesting, but some are truly great (the Snap additions, for example). This is iteration, no revolution, and that’s usually fine: Windows is a mature product, etc. But Windows 11 needs something more than iteration, in my opinion, because of the regressions. And I just don’t see that happening anytime soon. In the next two to three years, however, maybe we’ll finally get what you’re looking for: a version of Windows 11 that is close enough to what you expect/want that moving to it won’t be a complete nightmare. Maybe.

But it’s more likely that Microsoft is simply pushing Windows 11 in this semi-new direction—it’s really an extension of the touch-first/mobile-first Windows 8 push, when you think about it—and that we’re going to need to adapt to that. Or move on.

Hard numbers

helix2301 asks:

I know we have heard you and seen you write a lot words about Microsoft trying to turn Edge / Windows into a legit money maker and the pivot from charging for Windows to giving it away like Apple. My question is while Bing and there ad business might be slow to go has there ever been any hard numbers on how much Spotify pays to have the ad in Groove thats always there or Candy Crush on the start screen or the featured apps in App Store. Is this revenue bundled in with Windows revenues or is this the billion dollars that Microsoft says they have in their ad business? As someone in business for himself I would love to know how much that ad real estate is worth I could never afford it but just curious.

I’m not a financial analyst, obviously, but having reported on Microsoft’s quarterly earnings for decades now, I’ve watched as the firm has evolved how and what it reports. There have been no formal changes to the reporting laws/regulations that I’m aware of, but my non-expert/high-level assessment of this is that Microsoft, like other big companies, has simply been limiting disclosures more and more to see when or if they get any shareholder/regulatory pushback. And they’ve gotten none. And so here we are.

What Microsoft does is report hard numbers rarely, and only when they are overwhelming good. And then they stop reporting them when they’re not. The net effect is to drive up its share price and thus its market capitalization, which of course benefits shareholders and the economy, and so no one is asking questions. For people like me who want to see some number in the current quarter that was reported in the previous quarter, this is frustrating. And I think, if not illegal, then legally questionable. But again, no one is asking questions.

Looking at your specific examples, those businesses or deals are so small in the context of Microsoft’s overall business that they will never be reported. Remember, this is the second-biggest company in the world, a company that reported almost $50 billion in revenues in each of the past few quarters. And despite the diversity of its product and service offerings, it has only three main business units. And in each of those business units, there is one primary business—Azure, Microsoft 365, and Windows, respectively—that generates the vast majority of the revenues for those units. And it has never reported the revenues for those products during the time period in which the current three business units existed. Not once. So it’s never going to reveal what Spotify paid (if anything) for the link in Groove. Or that Bing has never been profitable, which I’m sure is the case. (The company probably still thinks of it as an investment.)

I believe there was one quarter where Microsoft said that Xbox was profitable in its entire 22-year run. And I think there might have been one quarter for Surface too. Were these businesses unprofitable every other quarter? Yeah, probably. But they’re subsidized by Windows and the rest of Microsoft’s profitable businesses, and Microsoft only talks up the cloud these days because that’s what excites Wall Street.

And here we are.

Access

andrew b. asks:

Paul, does anyone at Microsoft care about Access? Is it something that is likely to be killed off in the near future? If not, will it ever be brought to Mac?

Access reminds me of FileMaker, which used to be an Apple product: it has this core group of dedicated fans/users and its parent company just seems to ignore it. You can find the modest list of improvements in Access 2021 here, but the better way to measure Microsoft’s dedication to Access is to pay attention to its monthly Microsoft 365 update posts. (I think this is the most recent one.) They never mention Access, and I suspect the reason is similar to that discussed in the Windows 11 conversation above: so few people use it that Microsoft has sort of stopped worrying about it. It’s likely in some form of maintenance mode. I don’t see it ever coming to the Mac. Or the web, which is perhaps where it should end up.

Office on Linux

andrew b. also asks:

Also, a while back I recall seeing some rumblings about Microsoft considering bringing Office to Linux. Is that still (or was it ever) a thing?

In a weird coincidence, one of the first things I ever wrote for Windows NT Magazine back in 1999 was about Microsoft’s secret plan to bring Office to Linux: I happened to be in Israel on a work trip when the CEO of the company I was consulting for told me he had just interviewed two people from Mainsoft, and that they had told him that their current employer had been contracted by Microsoft to build the port just in case. He let me speak with them and I wrote about it.

Obviously, that never came to fruition, most likely because Linux never took off as a serious desktop competitor. (It did, of course, become quite successful on the server.) But I remember telling my wife a the time that if I was suddenly assassinated, that would be why: Microsoft was not happy that I had outed this relationship.

Looking at the computing landscape today, I don’t see any incentive to building native Office clients for Linux. But I do see a huge incentive for improving the quality of the Office web apps, turning them into true PWAs with offline support and so on. And such a thing would, of course, work fine on Linux (as do the apps today in their current state). But I would very much like to see Microsoft bring native OneDrive file system sync to Linux. That, to me, is job one.

The evolution of the iPad

spacecamel asks:

After WWDC, it appears to me that iPadOS is slowly being merged into MacOS to give a touch operating system for the desktop/iPad. While Apple is doing this very, very slowly, Microsoft put touch into Windows in one iteration, (Windows 8) and without any testing, input from others, etc. Do you think Apple is eventually going to end up in a better place than Windows 8 did? Is Apple doing this too slowly or doing a “bite” at a time and seeing what works with customers?

Arguably, the iPad today is already in a better place than Windows 8 was. And when you look at what Apple has done in the past few years, and at the WWDC 2022 announcements, in particular Stage Manager, you see that relentless (and, yes, slow) push to make that device make more sense as a general purpose productivity computer.

It’s interesting to me that Microsoft is making the same decisions (one might say mistakes) with Windows 11: like Windows 8, Windows 11 offers a simpler experience that removes features traditional Windows users expect. And like Windows 8, it’s mobile-focused (in the sense that the new UIs are meant to be complementary to what we use on mobile platforms), and can be used on touch-first form factors. Windows 11 isn’t as aggressively stupid as was Windows 8, but it’s still very much a response to how mobile has taken over personal computing.

But yeah, there are rumors now of a coming 14-inch iPad Pro, and my mind has certainly wandered down that path of what might this future look like when iPads are the size of laptops and are more functionally sophisticated while still retaining everything that’s great about iPads (mostly as consumption devices) today.

This also raises that debate—which also first surfaced in the original iPad timeframe, when Windows 8 was just happening—about whether it’s better to take something complicated (Windows) and dumb it down for more mobile devices or to take something simple (iOS, at the time) and improve it so it makes sense as a productivity platform on more sophisticated devices like an iPad with a keyboard dock or whatever.

It may not be worth arguing that Apple “got it right” because both companies simply started with what was successful for them. But then again, iOS came out of Mac OS X, where Apple simplified the system while adding touch capabilities. And so the genesis of the iPad evolution we’re seeing today started similarly to how Microsoft today is trying (again) to simplify Windows. So maybe Apple did get it right.

Whatever one’s opinions of Apple or the iPad, this is something to keep an eye on: Windows is still very successful on traditional form factor PCs, but Apple could very easily steal away the tablet/2-in-1 market. And probably already has.

Windows 11 Field Guide

wmurd118 asks:

Paul, I can imagine that things are rather hectic for you now and the last several months. But, can we expect a Win 11 based Field Guide any time soon. I thought I heard you mention in the last Windows Weekly that it will be based on 22H2 which in my mind indicates a release of 6 months or more. I used the Win 10 guide quite often and am looking forward to the 11 version.

Yes, sorry. There’s a lot going on here, and I can’t yet announce a big part of it. So let me try to explain what’s happening and when as best as possible.

Yes, I’ve been working on the Windows 11 Field Guide on and off since last summer. I say “on and off” because I don’t want this book to be identical to the previous book from a content/organizational perspective, and I’ve actually restarted it at least three times trying to find the right approach.

I think I have: Instead of several big changes, the Windows 11 Field Guide is organized into many smaller sections (which I guess are technically chapters). They’re more bite-sized and cover very specific topics. For example, a “chapter” on Snap instead of that being just a small part of a big chapter. There is also a focus on what’s new and what has changed since Windows 10, and in many cases, this involves a “Where did it go?” section that explains where to find things (when possible) or how to get around regressions.

Tied to this, I wanted to make the presentation more visual somehow, and so I have been experimenting with video and how to incorporate that, if possible, into the book. That was a months-long rabbit hole, but the answer finally came, and this is the part I can’t reveal yet. Basically, there will be a video companion of some kind to the book. It won’t be a “part” of the book, but rather a companion piece, and I could see people choosing one or the other (or both) based on how they like to consume information. But I’ll have more on that soon.

As for the timing, yes, I do expect Windows 11 22H2 to release formally in a September-ish timeframe, but I won’t wait until then to publish the book. It will be an e-book like its predecessors, and it will be published initially in an incomplete form and updated regularly. I expect to deliver the first version some time this month.

Sorry this is taking so long, but between the video work, me trying to wrap up the Programming Windows series (which, frankly, should be a book as well), and my unexpected Mexico City distractions, it’s been a weird time. But things are calming down, and I feel like I’m on the right path to making this book different from its predecessor and uniquely useful. It’s happening.

Cloud gaming

justme asks:

How would you normally access XBox Cloud gaming? Isnt this typcally done through an app? (I game on PC with locally installed games and dont use XBox, so I dont know the cloud mechanics).

This is going to vary by person, but I think Phil Spencer’s approach, which is to basically meet gamers wherever they are, makes sense. Today, I game exclusively on an Xbox console that is next to my PC and has its own screen. I can (but don’t) use that console to stream games via Xbox Cloud Gaming, and that would work fine. But having just that screen with a Keystone (or whatever) dongle would work just as well. As would a smart display with a built-in Xbox app. Going forward, these will all be options, with the key being seamless (and latency-free) controller connectivity.

I did experiment with doing this on the PC. But I like the separation of work and play, if that makes sense, and even when I tried game streaming off the PC to a secondary display, there were input issues that made it less than desirable. This may get easier in the future, I guess.

Why try and entice gamers to use Edge to access cloud gaming if they already have an app that does that? Also – bonus question, shouldnt this be an extension instead of built-in functionality? Arent extensions one of the benefits of being Chromium-based?

I think this is just tied to that notion of meeting gamers wherever they are. Some people live most of their computing life in a browser, and while I’m not a fan of bulking up Edge too much, this does at least bring Xbox functionality to Mac and Linux, too, where there is no Xbox app.

I’m with you on the extension thing, for sure. But Microsoft has clearly decided that adding features directly to Edge is “better” in that it furthers their corporate needs and provides a more seamless experience for users. But in keeping with the “meeting gamers wherever they are” thing, an extension would accomplish that nicely, since most people use Chrome or Safari, and not Edge. Perhaps there will be an Xbox extension as well.