Ask Paul: January the 13th (Premium)

Happy Friday the 13th! Here’s another monster installment of Ask Paul with some great questions to kick off the weekend a bit early.

Microsoft’s datacenter moves

wright_is asks:

Microsoft has bought DPU builder Fungible and HFC manufacturer Lumenisity in the last couple of weeks. Are they trying to restrict other cloud providers possibilities for expanding or are they moving into the datacentre and long distance fibre optics device manufacturing?

Lumenisity looks good on paper, it could increase the performance on transoceanic cable runs, HFC (Hollow core fibre) is much faster than traditional fibre, because light travels faster through air than through the fibre, on a 5KM transatlantic run, it could save about 16ms in latency, for example.

We can only speculate. Given that Azure is Microsoft’s future, it makes sense that it would make ongoing investments in the form of acquisitions and partnerships. I’m no expert in this area, but I would say that these two investments are both for in-house needs and that what Microsoft sells to customers on the other side will continue to be cloud services that run in datacenters that use these technologies. And that’s how simple I am when it comes to Azure. 🙂 More seriously, apologies.

Microsoft’s AI flanking maneuver

poppizel asks:

In what looks like the most significant step by Microsoft in years may be bearing fruit. Its investment, both financial and technical ,in a leading AI “startup” has yielded some early and very compelling results, summarized well in a recent NY Times article. 

Yes, I was going to point people at that New York Times article, which I read with great interest this morning. There are a couple of key new bits of information there:

• We all know about Microsoft’s initial $1 billion investment in OpenAI and that Microsoft is rumored to be looking at investing another $10 billion with various conditions around commercial availability and ownership. But the article cites multiple sources who say that the company had quietly invested another $2 billion in OpenAI since that initial investment. (This other investment was also reported elsewhere.) So it clearly saw the value here and raced to secure as much of its work for itself as possible
• The article confirms that Microsoft is in talks to invest another $10 billion.
• Microsoft has been quietly integrating AI technologies across its stack in recent years and at least some of that came out of OpenAI. I would love to see a breakdown of that, and it almost makes me want to go back and reexamine every time that Microsoft touted anything as being AI- or ML-based. Almost. But that Tay chatbot experiment may very well have been an early example of Microsoft going public with this work. (“We’re pushing the envelope in A.I., and we’re going to do that across our products,” Microsoft’s Eric Boyd told the publication. “Microsoft incorporated GPT-3, DALL-E, and similar technologies into its own products,” the report notes.)
• The capabilities of what became GitHub Copilot, which can write code, was a “jaw-dropping moment,” Mr. Boyd also said.
• Nadella said last week in India that as much as 10 percent of all data could be A.I.-generated in just three years, which could lead to as much as $7 billion in revenue for Azure.
• The article’s conclusion is the most compelling bit, in soundbite form: “There is an argument to be made that [Microsoft, Google, Meta and other companies pushing AI] all end up smelling the same,” Index Partners’ Mike Volpi said. “There is another argument that what OpenAI is doing is truly special and that all the money goes to [Microsoft].” If that latter comes true, it could set up another golden age for Microsoft, akin to the 1990s. It’s an incredible thing to contemplate.

Given your 20 years of industry experience, what is your opinion not only on what areas Microsoft will focus but given their history of missteps with managing, do they have the ability?

The way I view Microsoft right now is that it’s in a transitionary period, and this was true before we found out anything about the OpenAI stuff, and it’s still true. It has all these legacy products that have been paying the bills for years, and it has this new(ish) direction—the cloud—that we all sort of cite as “the future” of Microsoft. But in recent years, the present and the future have been converging, and if we look at Microsoft from a high level, it’s not impossible, even given Microsoft’s non-transparency efforts, to make the determination that revenues from the cloud are finally starting to overtake those from legacy products. That’s the transition. That’s where we are right now.

There’s no version of the future in which legacy (or what we might call “on-prem”) products—things companies and people use locally—go away. But the mix shifts so that the cloud, overall, is dominant. I’ve written many times that this new Microsoft is not particularly interesting to me. The things I care about are software, services, and devices that actual people use. Personal computing, to give it a simple if not totally accurate name. (I care deeply about productivity, and this is more of a commercial computing topic, so to speak, but it still very much involves people using things, hopefully efficiently.) And so things like Azure are of little interest to me personally. It’s infrastructure, the back-end. This has been a struggle for me as Microsoft has deemphasized Windows, especially.

But AI is thrilling and interesting to me in a way that much of what Microsoft has done over the past decade was not. And it’s probably because of how it impacts people, right? There are a million examples, but something that can help someone who is not a writer communicate effectively with Microsoft Word is fascinating. Something that helps someone who is not good at making, let alone giving, presentations, present information visually in an effective manner using PowerPoint is too. And so on.

So the question here is whether Microsoft will bungle this. And my answer is, no, I don’t believe they will.

Two reasons why.

First, look back at that quote I highlighted above: Microsoft is either going to be one of a handful of firms that have the capacity, intelligence, and market power to control AI or it will be the only company that does so. Either work out fine. Microsoft has become a much bigger and much more successful company in a world in which it does not define or control personal computing than it was when it was dominant in the late 1990s. That’s an important fact, and it’s why Windows being the smallest of the big three personal computing platforms (after Android and iOS) doesn’t matter (to the company and its shareholders) in the slightest. Microsoft today is bigger, earns far more revenues, and is far more profitable than ever before. So that future is good either way.

But the bigger point, to my mind, is related to a concern I’ve voiced many, many times. Microsoft shifted its emphasis from Windows/Office/whatever to the cloud over the past decade for very good reasons, and that’s why it’s the second biggest company in the world: this speculation, now turning into reality, has reinvigorated its stock price and catapulted its market capitalization. And this shift triggered a similar shift inside Microsoft, where the brightest and best employees left Windows, Server, and Office (and other groups) to shift to cloud work, where they could find both fame and fortune, start new businesses, and make a difference. The best and the brightest, which we absolutely do not see in Windows today, is in this very part of the company. These are the talented adult leaders who can succeed. And, I think, will.

I’ve never been particularly happy with Satya Nadella. He’s an engineer and technologist, and so he was a reasonable reaction to Steve Ballmer, who was a businessman and a salesman. But he seems flat and robotic, and I don’t see much in the way of vision. Even Gates was more interesting as a person, or a presence. Nadella does bring a sense of pragmatism. But he’s not inspirational.

But here’s the thing: my ambivalence about this man is probably rooted in the concerns that I noted above, that he’s pushing Microsoft into this cloud future, and I don’t care for that. But now we learn how hard he’s pushed on AI, and I can see the first glimmer of hope that this guy may have been onto something all along. Not just pragmatically—the cloud is the future—but inspirationally. If that makes sense.

I think Microsoft’s future is bright.

HoloLens

john_m asks:

What do you think is the future of Microsoft hololens? Both Apple and Meta are putting effort into AR/VR but it still seems to be a solution looking for the right problem.

I see no future for HoloLens. Period.

That said, there is a platform there, and technologies, that are very interesting. And it’s possible that HoloLens as a platform used by other hardware makers could have a future in commercial computing. This U.S. Army thing was so predictable that it’s painful to watch it unfold, but some of the commercial use cases that Microsoft previously highlighted are still very interesting. But that’s true only in theory when the headset costs $3000 and is never updated anymore. This needs to come to a broader market akin to what we see in consumer electronics with lower prices and broader availability. And that’s just not going to happen with Microsoft. It could happen with partners.

Mexico City

john_m asks:

On a very different topic, how reliable has your Mexico City internet and cell service been?

Excellent, though we haven’t been to Mexico since October. When we arrived on that trip, our Internet was down, but something had happened out at the street. The speeds there are higher than they are here at home.

As for the cell service, that’s worked wonderfully in the United States, though I don’t really need it here. When we go back, probably in February, we will get a second line for my wife.

Glucose monitoring and getting help

karlinhigh asks:

Quote from last week:

“In fact, I’d really like to figure out a glucose device so I could see how different food types spike my blood sugar.”

I’ve been using Zoe (joinzoe.com) since last August or so, and it seems very good. It just DOES something to me to wear a glucose monitor and watch what happens when I eat something. They also do a finger-prick blood test to measure blood fat. (Or can use lab test results from within last 18 months.) Also a stool sample to study gut health, what bacteria someone has going on, or doesn’t. They ship special-recipe muffins to eat at specified times in those tests, to get readings from known-quantity food items.

After test results, they assign scores to all different foods, from 0 to 100. The scores are not the same for everyone, based on how the person showed responses to sugar vs fat. Then enter meals into the app, and it says how things rate. Most people did not consider me in need of dieting. But following this thing’s advice, I still lost 20 lbs without really doing anything else. I can post a $35-off referral link if anyone wants to try it. (I feel too much like a spammer to post it in advance of request.)

I’m overdue in going to the doctor for my annual physical and a glucose monitor is among the things I’ll be asking about. I did step through the Zoe wizard just for the heck of it. I’m not sure that I’m personally interested in a plan like this right now, but I’m OK with it if you want to post the referral link for others.

Macs, touchscreens, and more

helix2301 asks:

Have you listened to Leo’s new show ask the tech guy?

No, but I spend about three hours a week with the guy. 🙂 And the show is a sort of replacement for his radio show, which is aimed at non-technical people.

There are rumors Apple is doing a touchscreen mac. Do you think these rumors are true? After listening to you talk about Apple and their thoughts on touch and iPhone and iPad I can’t picture it.

I have never quite understood Apple’s reticence in bringing touchscreen displays to the Mac. Such a thing would benefit developers creating iPhone and iPad apps because they could interact more naturally with them in the emulator during development. And it would benefit customers who wish to use iPhone and iPad apps on their Macs.

That said, I’ve also come around to the notion in recent years that I don’t like or need touchscreen capabilities on my laptops. I had argued in the past that having this capability didn’t hurt anything and that some might find themselves interacting with the PC in a new way. But touch on laptops is awkward, and too many times I’ve reached up to brush away a hair or other detritus on the screen and have inadvertently interacted with something onscreen. So I’ve given up on it.

As to whether I think Apple ever will add touchscreen capabilities to the Mac, I’ve learned to never say never with Apple and to regard every one of their strong denials as a holding pattern until they do that exact thing. They were never going to bring mouse support to the iPad, for example. And now it has full keyboard and mouse support, and Apple makes expensive peripherals to enable it. Also, the source of this rumor, Mark Gurman, has excellent sources and is quite accurate.

Why do you think they are not doing a mac pro? I have a lot of customers that are waiting for this and just wondering what your thoughts on that are.

I believe they are making a new Mac Pro, it’s just that it won’t be an all-new design and won’t have some of the higher-end CPU configurations that were originally planned. The issue there, as I understand it, is two-fold: it was supposed to support high-end ‌M2‌ Ultra and ‌M2‌ Extreme chipsets that Apple is allegedly having problems producing, and it suffers from the same non-expandable issues as existing M1/M2 chipsets: you configure the RAM and GPU at purchase time and they can never be upgraded. So I view the current plan as a holding pattern: use the existing box instead of a new design and at least ship something, something that is more expandable than a Mac Mini.

Surprisingly enough we have been very busy getting users off of windows 8 and now getting customers off server 2012. I was surprised not at the number of customers with 2012 server but the number of windows 8 users we are upgrading. I think we will see more people having 8 then we thought as the eol comes up soon.

Statistically, Windows 8.x usage is not high: StatCounter says that only 0.66 percent of Windows users are on the system, so under one million people. But that’s still one million people. I can’t imagine many of them are emotional about Windows 8.1, but at least some must be confused why they can’t just keep using a PC that otherwise continues to work well. I assume most can at least be updated to Windows 10, ride that out for another 2.5 years, and then they can consider moving to a new PC.

Bad travel

ggolcher asks:

I’m in the middle of a hellish travel experience where I’ve doubled the amount of time I’ve spent in transit and still haven’t gotten home. All this got me thinking: you’ve traveled quite a bit. What’s the worst experience you’ve had traveling? (Misery loves company!)

First of all, sorry to hear this. Traveling is stressful in the best of circumstances, but it’s awful when things go wrong.

I’ve been mostly lucky, and I’ve traveled a lot. My worst travel experiences, in some ways, were not things I experienced, but rather those times when my kids were having issues.

For example, several years ago, when the kids were in middle/high school and we were still in Dedham, they flew together out to Colorado to visit my Dad in the winter and go skiing. They arrived in Denver to switch planes and my daughter called because they couldn’t find the gate for the connecting flight. I literally had to tell her, sorry, we can’t help you from here: you need to find someone who works for the airline there and have them help you. She did, and they did, and they made the flight.

Then, on the way home, their flight from Denver was delayed repeatedly because of snow and they were stuck there. We were supposed to pick them up in Boston in the late afternoon, but because of the delays, the flight finally left really late and would now arrive at 3 a.m. So we tried to sleep, failed, and then drove into Boston to get them. It was the quickest airport drive of my life, with no traffic, and when we arrived, there were only four or five cars in the parking garage. And we learned, quickly, that all of them were people there to pick up others on the same flight. We talked to other couples there, and we all relayed the stresses we had experienced. And then, finally, people started coming out from the plane. And when the kids arrived, they were smiling and looked triumphant, like they had accomplished something. Which I guess they had.

Years later, we were doing what would become our last home swap, in Amsterdam, and our son, Mark, had to fly separately because he lives in Rochester and couldn’t come for the whole trip. On the way to Amsterdam, his flight from Boston was canceled, again because of weather, so we were up half the night rearranging the two flights (Rochester to Boston, Boston to Amsterdam), with my wife on her phone talking to one airline, and me on mine with the other. Hours later, and after spending a lot more money and coughing up a lot of JetBlue points, we got the new flights booked and tried to go to sleep. Which was impossible, we were so wired.

Within 30 minutes of finishing that, I was still wide awake and my phone rang. It was a Pennsylvania phone number. At 3 am. What? I picked it up. A man told me he was trying to deliver the pizzas we ordered from Dominoes but couldn’t get to our house because of flooding related to the same storm that had caused issues for Mark. Confused by this, I didn’t think to tell the guy that he could have made it to our house using whatever alternate route: we do get flooding on the creek, which closes two bridges but there are other ways into our neighborhood. But I was too out of it. So I hung up the phone.

My wife was looking at me in the dark, confused. I told her it was the pizza guy. And then I called my friend Steven, from Amsterdam, who was in my house on the other end of that home swap and had obviously ordered a pizza. “Hey, that pizza isn’t coming,” I told him after he apologized because he had had to use my local number, which made sense. “I told you where to get pizza,” I told him, “and you called Dominoes? Come on!” But I told him it was no problem: we were up anyway.

My son flew the next day without incident. But unfortunately, and apparently because he is cursed, he had further delays on the way home: he was stuck in the Amsterdam airport for hours this time. There was nothing we could do, but he did get home safely after the delays.

Surface alternate timeline

ErichK asks:

Paul, here’s a wild and wacky thought exercise for you: if the iPad had not existed yet, what would have become of Surface, especially Surface RT?

I don’t believe that Surface would exist without the iPad, nor would Microsoft have ever entered the PC market. Tablet PCs had failed to make any dent in the market at all, and it’s likely that Windows 8 would have just included the many desktop improvements and none of the mobile UI and app nonsense.

Antivirus

andrew b. asks:

Do you recommend using any paid antivirus these days?

No, and I was just thinking about this because I was writing (well, updating) the Security section of the book. I feel like the paid antivirus packages are really about the things that are not antivirus/anti-malware. That is, you pay for them because they include additional services like identity protection, VPN, and so on, and you want all of that to come from a single source and be centrally managed.

But if your concern is keeping your PC safe from attack, the built-in protections in Windows 11 work fine, especially if you have a very modern PC that enables hardware-backed protection across the board. Looking at my various PCs, it was interesting to see the range of security capabilities, which is most easily seen by opening Windows Security and navigating to Device security. There can be a wide range of capabilities listed there, but some of my PCs only have one or two. (You will probably always see Core isolation and Security processor, the latter of which is the TPM. But new computers will often have even more hardware-based protections.)

MacBook Air 15-inch

sabertooth920 asks:

If the long rumored 15” MacBook Air becomes a thing, are you in?

This is the MacBook I would want. That said, I have two MacBooks here right now—an Intel-based MacBook Air that my daughter used to use and a first-generation M1-based MacBook Pro with a touch bar—and I don’t really care for or use either almost at all. I feel like I need to keep a Mac around for testing purposes, but as I don’t see myself ever going Mac—I really do prefer Windows—I guess not. At least not right away. But when it’s time to upgrade, that will be my choice for sure.

Regrets, I’ve had a few

j5 asks:

Hey Paul, what’s one or some of your biggest regrets in all the years you’ve been covering tech?

Hm.

I used to regret that I focused solely on Windows, go figure: as personal computing shifted away from Windows and the PC, I figured my strong association with Windows might make me less viable as a writer or whatever. But over time, I realized this wasn’t the issue I had thought it was, and—as noted above–­-I really do prefer Windows. I was happy to see a PC resurgence during the pandemic, and was really clear that this was temporary, so the current overreaction to the natural downfall we’re now experiencing is likewise worrying. It will be OK.

I’ve certainly gone down paths that didn’t pan out. But I don’t regret that. You learn by doing, and if you try something and fail, that’s valuable. To continue the Mac discussion above, I bought my first Mac, a white iBook, in 2001 so that I could test Mac OS X, and I’ve had at least one Mac here ever since (oftentimes more than one). What I’ve learned using the Mac is useful, including the fact that this platform just isn’t for me. But I couldn’t make some determination that Windows was “better” (in whatever ways, for me or others) without actually experiencing that. And keeping up to date with it.

I guess the thing I regret most is basically time management related. I walked away from programming because I was so involved in whatever was happening with my writing career in the early 2000s, and it was nice to formally build back those skills to a small degree over the past several years. That said, it’s easy to get caught up in big works—that Programming Windows series, for example, the book—and let other things drop by the wayside. I guess that’s not tech related, but I feel like my whole life revolves around tech, so it’s in my head.

More broadly, I do worry about exerting any influence over others to make technology decisions and about the negative impacts of technology, especially now in this Big Tech era. Am I implicit in causing others to spend money or waste time on things that become dead ends? I really do struggle with that. Again, not technically tech related. But I think about it.

I was a sort of anti-Microsoft Commodore fan in the early days, and after spending a stupid amount of money on a crazy Apple IIGS system and then finally getting an Amiga and going down a similar rabbit hole there, only to watch Commodore crater in the early 1990s, I wish that I had instead focused on the PC earlier. If I could have learned Windows programming in the early 1990s, for example, things might have been very different for me. My time with the Apple IIGS and Amiga wasn’t completely wasted. But if I had known then what I discovered just a few years later, I would have made better choices. Still, I am nostalgic for some of that. And I did get lucky in the mid-1990s when I belatedly made that move and met the guy who got me into writing.

To sort of turn this on its head, I feel like a did a pretty good job of seeing things that were important that did not come from Microsoft. The iPod and the iPhone, for example. The summer that the iPhone arrived, multiple people complained that the SuperSite should be renamed to the SuperSite for iPhone, but I just felt it was a big deal, and it has certainly gone on to change the world. Getting out of the bubble is huge, and it can be hard. I don’t always get that right, but seeing clearly, seeing past a bias, or whatever, is important. Imagine if I was advising people to buy PlaysForSure or Zune devices as those platforms were imploding. I would hate myself for that.

Well. There’s a bit of introspection I will now keep thinking about. 🙂 Would love to hear about others’ regrets.