Ask Paul: April 4 (Premium)

Happy Friday! It’s been a curiously portentous week for our industry, and we have nothing but questions. So let’s dig in.

? Microsoft is my Copilot. Or something

Anlong08 asks:

If you think to your favorite editor from your magazine and print book days and if money was no object, would have an editor today?

Short answer, no.

But we all need help. In The Shortest Path from Thoughts to Words (Premium), I glossed over that part of the writing process–it was more about a life-long goal of finding the most frictionless path between idea and expression, or whatever–but that’s important, too, and often overlooked. In the early, paper-based days, the book and magazine publishers I wrote for each had a specific process that involved multiple stages of edits involving several people. And I also did tech editing myself, both for the publishers I wrote books for and for the magazine publisher (Duke Publishing, later Penton) that I worked for.

This is like so many things in life. I went into this just wide-eyed and accepting of whatever process, assuming that everything was in place for a good reason and that experts who came before me knew what they were doing. And then I came to resent the bits that didn’t make sense. Ultimately, it all fell apart, and here we are.

The first book I wrote was about Visual Basic 3, but during that process, I also helped Gary, the professor who made this all happen for me, write, but mostly edit, an update to a previous Microsoft Excel book. So I don’t remember which manuscript came back first from the publisher. But I do remember Gary telling me that I needed to prepare myself for the disaster that was about to unfold. We had sent a six-inch stack of paper printouts representing the book to the publisher, that manuscript would be reviewed by four different editors, each using a different color pen, and it would come back with more colored writing on it than what we had written. He told me that the first time he saw a manuscript marked up like this, he just started crying. It seemed so overwhelming.

Those first manuscript edits were exactly as bad as he had described. But we plowed through them, together, one page at a time. And we got through it. That this was being done on paper amazed me, it felt like something that could/should have been done electronically, even then. And over time, of course, that is what happened: Publishers embraced the track changes features in Word, and we handled multiple levels of edits–and all the comments, which even in the paper-based days, were out in the margins–that way. I don’t remember when that started.

No one likes to be corrected. I’m sure I told the story of sitting across from Gary and him reading through some pages–yes, printed out, no one had laptops then–from the first (VB) book. He would scan each line, whether it was text or code, with his finger as he read, and he was going through some bit of code. And his finger stopped moving forward, reversed, went back a bit, stopped again. And he looked up at me and asked, “Are you sure about this?”

“I was,” I replied, knowing that Gary, an expert programmer who taught classes on this subject, had found a mistake. I can still feel that weird sinking feeling in my stomach now, just thinking about this.

Gary got up, walked into his home office, opened VB, and started typing. And sure enough, there was a mistake, or whatever problem. This scene replayed itself over and over again as we worked on this and other books. So even before we sent in our first co-written manuscripts, I had gotten a taste of what it was like to be constantly corrected.

You can get used to anything, I guess. But there were a lot of changes in the next few years. I created a website–today, I’d call it a blog, called the Internet Nexus, and one of the sections I had on the site was called The Future of Windows. I started a series of email newsletters at the college Gary worked at (and I attended and then worked at), one of which, WinInfo, continued past the others and was shared first among the professors and teachers and then the students and then just publicly. A guy from the publishing company we wrote for started his own small company in San Francisco and I worked for him for a few years. I brought the newsletter and site there and kept writing books with him. And then that company went under, and I was on my own, and the guys from WUGNET were nice enough to host the newsletter and site for me.

It was during this time that I created a site that was specific to Windows NT 5.0, which became Windows 2000, because I was excited by what I had learned at the Beta 2 reviewer’s workshop. That became the SuperSite, and it sort of replaced the Internet Nexus, even though I meant for it to be temporary. And then Duke Publishing, which made Windows NT Magazine, came calling. An editor at the publisher emailed me to see whether I was interested in writing the editorial for a weekly email newsletter, then called Windows NT Magazine UPDATE. And that turned into the publisher buying WinInfo and the SuperSite and bringing me onboard.

Duke had what was to me a familiar editing process by that point, with multiple editors. But it was particularly serious for the print magazine. None of this was paper-based, at least, the world had moved on by that time. But it was multi-level, and you had to address every concern any of the editors raised. These edits were also anonymous, which I found amusing, and I got to know who each was just by their styles and questions. This irritated my editors. There was an entire editorial team, literally dozens of people.

When I first came on board, Duke flew me out to Colorado to meet the team. I was a curiosity to them. Though I had published several books by then, I was seen as a lesser individual because Duke was interested in my digital/electronic work with a newsletter and a website. And I had a bizarre first meeting with the editorial team that was literally each person, one by one, criticizing me harshly. I’ve never endured anything quite like this, before or since, and I’m surprised even now that it went on as long as it did. It was like the scene in the movie Airplane where there is a line of people, each with some kind of weapon, waiting to beat on the frantic lady who can’t calm down.

The over-reaching issue, hilariously, was that I was too personal. Everything I wrote was “I …” or “me,” they said, instead of objective overviews of technology. To me, that was the point, this was my experience using these things. But they hated it. I wrote in a way that was unlike everything they did. What I had going for me was that what I did worked, and was successful, and eventually, they started asking me how to promote other authors in a similar manner. But that took a while.

That said, I had a terrific relationship with all the editors who worked on the UPDATE newsletter over the years–the name changed over time, as with the print magazine–and I’m still very good friends with the last person who held that job. But your question instantly reminded me of my first editor, Ronnie, an older woman and a sweetheart, and of some of the funnier interactions we had. I’ll just mention two here.

The UPDATE newsletter editorial was supposed to be “500 to 700 words” in length and “can be about anything germane to NT/Win2K — keeping in mind our target audience consists of IT administrators,” according to the original email reach-out. (I’m also still friends with the person who emailed that to me, go figure.) But I would routinely hit 1,000 words or more. And poor Ronnie, she finally told me, “you know, this only has to be 500 words long.”

No one reading this will be surprised to discover that I had trouble being succinct even then, but my take on this at the time was that certain topics just required room. I replied that “it was more bang for the buck for the reader.”

“Paul, it’s a free newsletter!” was her reply. Hilarious.

I loved Ronnie. One week, I had written my editorial, but our Internet had gone out, so I couldn’t get it to her. And I had to get to the airport for a flight, so I would be out for the duration. I called her to explain the problem, and she asked if I could just read it to her over the phone. I had to leave, but my wife was able to drive me to the airport–she had coincidentally just started working from home too–so I dictated the editorial to Ronnie over the phone from the passenger seat. Incredible.

When Mark Smith, who had cofounded Duke Publishing and started NT Magazine, convinced that it was the future, sold the company to Duke, he stepped aside. And we went through a series of leadership changes, most of whom I loved dearly, and am still friends with. But before the inevitable downfall that occurred across publishing because of digital settled in, we had this one dark period in which the editorial director over-emphasized that part of the business. She literally told us in an in-person meeting that editorial was more important to the company and its success than the writers. And the editorial process that had started off as onerous got even more brutal.

I can almost excuse that. The issue was that we had some of the best technical people imaginable, true experts, writing content for the magazine and newsletters. But most of them were terrible writers. This makes sense: When you’re really, really good at one thing, it’s unlikely you’ll be that good at other things. I was the exception in that sense, and I was allowed to do my own thing with WinInfo and The SuperSite, something the other writers resented. But the print mag and UPDATE edits became very trying for me. They were out of control.

That ended soon enough. Paying all those people to do multiple levels of edits became too expensive as faster moving blogs and websites started to dominate in this space. And our editorial process lightened up. And then lightened up much more when the print magazine ended. And then disappeared entirely. But that took many years.

To bring this full circle, I haven’t had editorial oversight for many years now and would resent and reject it today. But I still need help writing, in the sense that I am human and make mistakes, especially with typos and grammar issues. If you think again about that article The Shortest Path from Thoughts to Words (Premium), and the desire for this process to be frictionless, I think about this kind of thing in the same way. Like anyone else, I rely on the spelling and grammar-checking features of whatever word processor/editor I use. But I also sort of emulate, to a small degree, the multistep editing process of years past by using different spell/grammar engines during writing and when posting. When something I write is pasted into the editor behind the site, an AI-based web browser extension (in this case, LanguageTool, which I do pay for) kicks in for a second pass of edits.

Nothing is perfect. But time to publish is huge, and these tools are getting better. In the same way that I would never hire a graphic artist to create images for web articles, I’d never hire an editor. It would be too expensive. And it would be too time-consuming. It’s a different era.

☁️ Windows $365

wright_is asks:

I was looking at the Microsoft Link article and I just don’t get the appeal of the Windows 365 offering, it is very expensive, even compared to other hosted solutions. The Link costs nearly as much as our Core i5 desktop PCs with Windows and 16GB RAM, an equivalent Windows 365 offering is around $100 a month, so after 1-2 months the price of the Link + Windows 365 rental has paid for a local desktop.

If we lease those physical PCs, they cost around $17 a month. We are looking at using a Citrix server with dedicated nVidia pro graphic cards for our AutoCAD users works out cheaper per user than an equivalent Windows 365 offering.

Where is the value model for Windows 365?

For whatever it’s worth, I don’t understand Windows 365 Link or Windows 365 either, and cost has a lot to do with that. Even if one feels that Windows 365 solves some problem, anyone accessing this service has a device that would work just fine, and almost all of them wouldn’t tether you to a specific desk in a specific room of a specific location. Even if Windows 365 makes sense somehow, why not just use a laptop or iPad you already have?

But playing devil’s advocate, maybe we’re looking at this the wrong way. There are the types of costs you mention, cost of device acquisition and the cost of the monthly subscription, and those don’t make sense in a spreadsheet-type side-by-side comparison to me as an individual. But there are also other costs in there tied to supporting a full Windows install on a dedicated PC, and the value equation of Windows 365 Link, especially, is that you don’t have to deal with any of that or future hardware and software upgrades. The Windows install is in the cloud, hosted by Microsoft, and kept up-to-date for you and the user perpetually. There is a premium, perhaps, to just having Microsoft do this for you. (When people were still freaking out about the cloud and Office 365, I would semi-rhetorically ask people who on earth was more qualified to host and support Exchange than its maker?)

This is perhaps a bit simplistic, but this reminds me a bit of the conversation–or the complaints–around Microsoft 365 Copilot pricing. $30 per user per month sounds like a lot of money. But managed businesses with logical groups of users can see the value of paying that for some of those users, and can demonstrate how paying that is actually saving them money. The key here is that elastic nature of the cloud, that this isn’t a sunk cost you pay for every user regardless of need. You can pay for it only when you do need it, either on a by-user basic, or based on whatever schedules. Windows 365 doesn’t (to my understanding) actually support a usage-based pricing model–where you pay more when you temporarily need more capacity–but it should. So for now, it’s just a thing you pay for on a per-user basis. Like a Microsoft 365 subscription (which likewise has multiple tiers and cost structures.)

I don’t believe that Windows 365 Link will succeed in any way. But Windows 365 makes sense on some level, if you think of it as part of a range of choices with the end goal of connecting an employee with some number of Windows apps that you need for work. I struggled to see mainstream use cases, frankly, and the price points are tough. But given good connectivity, a reasonable expectation today, it can make sense … if the costs make sense. Right now, that doesn’t seem to be the case.

Basically, I’m with you. I don’t see it.

? You steal the stereo, I’ll steal the TV

helix2301 asks:

MacBreak Weekly had an interesting show talking about Apple’s 49th anniversary on April 1. They talk about how, without Apple, personal computing would have been much different. There would be no GUI, and there would be a more fragmented computer market. While I agree with that to a certain extent, I do believe eventually someone would have made a GUI, and I do think that Microsoft would have made an OS that worked on all platforms.

It’s a bit of a stretch to suggest that the GUI exists because of Apple. Or that Apple popularized the GUI in any way, beyond inspiring others to go in that direction as well.

They did get there first, of course. Xerox allowed Steve Jobs and others from Apple to see the work Parc (Palo Alto Research Center) was doing with graphical user interfaces, networking, and object-oriented programming because it wanted to invest in Apple. Xerox had spent $100 million on Parc research over a decade and had little to show for it. But the personal computer market was taking off quickly, and so Apple allowed Xerox to buy 100,000 shares in the (then privately held) company in exchange for a tour in December 1979.

Bill Gates and Microsoft didn’t tour Parc until 1981. But Gates had met Bob Metcalfe–who co-invented Ethernet while at Parc and was confused by Jobs and Apple weren’t interested in that technology–in 1979, and Metcalfe had licensed Microsoft’s Unix version, called Xenix, for his company 3Com. Metcalfe was buddies with a former Parc coworker, Charles Simonyi, who had co-created the Xerox Alto and Bravo, the world’s first WYSIWYG word processor. In 1980, Metcalfe recommended to Gates and Steve Ballmer that they should hire Simonyi because they were looking to expand into applications. And Simonyi was tired of Xerox’s inability to cash in on its inventions.

(Simonyi was the first of several people from Parc to jump ship to join Microsoft. This is similar to Larry Tessler, the inventor of cut, copy, paste, and others leaving Parc to join Apple, I guess.)

Gates got a sneak peek of the Macintosh that had come out of the Apple tour of Parch in late 1981, and Microsoft got its first Mac prototype in early 1982. And Simonyi was responsible for the Gates/Microsoft tour of Parc. But we should give Gates credit for understanding the importance of the GUI. He saw as quickly as Jobs did that this was the future. And unlike with Jobs, Gates had to fight IBM over this, as IBM didn’t believe in GUIs but was Microsoft’s biggest customer/partner by far. He was waging war on multiple fronts.

Put simply, the GUI was going to happen with or without Apple. And Microsoft contributed in major ways to the Mac GUI, which is something Apple fans don’t like to acknowledge.

Thinking back to Simonyi and Microsoft, and how Simonyi was responsible for the first two most important GUIs, that for the OS and that for the word processor, it’s interesting to remember why Gates and Microsoft were so taken with him. As a developer and futurist, Gates clicked with Simonyi immediately. Microsoft had started out creating the same developer languages for every computing platform imaginable, and it wanted to do the same with apps, and Simonyi had created a “pseudocode” backend for apps that made porting apps to different platforms easier. (This was a bit like the computer emulators Paul Allen had created, allowing Microsoft to create its first BASIC versions without the machines, but in software.)

But Simonyi knew that GUIs were the future, he had helped invent that future. It is telling that the first product he made at Microsoft, Word, was bundled with a (really good, for the day) mouse. That product was for MS-DOS at first, but it would come to the Mac and a GUI–Windows–would come to MS-DOS. Microsoft was preparing for that future.

One early observation I had about the Apple Mac/Microsoft Windows dynamic was that the Mac failed in the market, and it was Windows, not the Mac, that popularized the GUI. (This is a kind of innovation, when you think about it.) The Mac would have disappeared entirely had Adobe not shown up in the late 1980s with Postscript, popularizing the laser printer (which, go figure, was also invented at Parc originally). And Apple’s Mac and post-Mac products were always expensive compared to the competition, and never commanded large market share. Commodore outsold the Apple II series by a wide margin by being both technically superior and much less expensive. In later years, the PC clone market and MS-DOS and then Windows did the same to the Mac. Windows PCs weren’t just cheaper than the Mac, they quickly became better. (This parallels what happened later with Netscape Navigator and Internet Explorer, come to think of it.)

The Mac inspired many people and numerous companies to go the GUI route, and there were many text mode-based pseudo-GUIs in the mid-1980s, too, mostly because of the limits of the PC hardware at the time. But the GUI was inevitable.

One related thing I’ve often wondered about is what the world might have been like if Apple hadn’t ignored its cash cow at the time, the Apple IIe, and had instead put its GUI on top of a future version of that computer. This would have mirrored what happened in the PC space. And it’s fair to point out that Apple did try this, albeit half-heartedly and too late, with the Apple IIGS. But that machine was hobbled with a sad processor and other limitations so that it wouldn’t cut into Mac sales. (Apple does this now with the iPad, via software limitations. And IBM did this with the IBM PC Jr, which, like the IIGS, also failed.)

Tied to this what-if is that wunderkind Steve Wozniak would have played a major role in that product. Indeed, he was part of the Mac team when Steve Jobs took that over, but for about 15 seconds. He was in a plane crash, lost his memory, and almost died, and he ended up not really contributing to any more Apple products. It’s difficult not to wonder how different things might have been if Woz had not gone through that.

?‍⬛ Differentiation is hard

TheJoeFin asks:

I recently upgraded my Windows to Pro so I can use Hyper-V. It is always a little shocking how similar Pro and Home version seem to be. Do you think Microsoft should differentiate the two versions more? What differences would you make, adding more pro features, simplifying home more, both, something else, more ads??

This is something Microsoft has never done effectively, but especially with Windows. The SKU-ification of Windows, as I think of it, came about because Microsoft had been successful having multiple Office SKUs, or product editions. And one of the things they observed was that customers tended to buy the most expensive version, just as people might go into a car dealer because of a sale price they saw but drive out a more expensive model, having been upsold on its advantages.

There’s a lot to do this, but I just mentioned the evolution of Office as a product over time, in passing, in The Shortest Path from Thoughts to Words (Premium).

First, Office began as a bundle, but it became a suite when Microsoft started integrating the apps with similar UIs and cross-app capabilities. The selling point was more for less, so you could get three apps for the price of two. Customers fell for this in a sense, because it felt like a value, but I bet most people only needed one app anyway, so they were just paying more. Which was good for Microsoft.

Second, Office shifted from being focused on individuals to being focused on businesses, and this was an era in which the company was expanding in all kinds of ways, for example, into servers, but also into a provider of services that businesses could trust. So we got the 10-year product support lifecycle. But we also got volume licensing. And with volume licensing, there’s only one Windows product edition, really (now called Enterprise). But customers can also use any supported version of Windows.

Microsoft embraced the Office-style multi-SKU model for Windows with Windows XP in 2001, which is right when this shift was happening. Yes, they had previously sold consumer Windows versions based on DOS and NT versions of Windows with workstation and server SKUs. But it wasn’t until these things were unified that we got the notion of “Home” and “Professional” versions. In subsequent years, this went nuts, with Media Center and Tablet PC editions at first, and then the nightmare of Vista SKUs (and slightly fewer Windows 7 SKUs) But now we’re back to Home and Pro.

Apple infamously sold only a single Mac OS SKU through the OS X/macOS years, not counting the old Server version, and then just gave it away. So the Windows shift is partially a response to that, but also just an acknowledgement that too many SKUs are too confusing. But it is interesting that Microsoft 365, as the successor to Office, is still sold in multiple SKU versions, even to commercial customers. I suspect that’s tied to how enterprises work, and that there’s a demand for lower-cost monthly charges for those users with fewer needs. But I also bet a lot of enterprise “overpay” for higher-end SKUs because decision makers are human, and there is something very real about seeing the value of the upsell. Even if it’s sometimes imaginary.

Anyway.

Windows. Two SKUs for individuals. Why?

I don’t know, honestly. I think it’s a mistake, and I think that Copilot+ PC is a quiet third SKU that, ahem, skews things even further. Or at least makes it even more complicated. But looking just at Home/Pro, this is a problem of Microsoft’s own making. It’s difficult to differentiate these things. I don’t see why they bother. But they do. And the dividing line between the two is curious to me.

Granted, the biggest advantages to Pro are features most people don’t need: The BitLocker control panel with BitLocker to Go disk creation capabilities, Hyper-V, and Active Directory/Entra ID sign-in capabilities (for smaller businesses, typically). I mean, that’s about it for most people. It’s not a long list, and it’s not all that important to most. And maybe that’s the most important point.

I would love to see Microsoft just offer a single SKU to all customers. I don’t understand the point of this now. But I guess you could argue that there are multiple Android versions, in a way (even within Google, with “pure” Android/AOSP and the heavily modified Pixel Android). Apple’s OS all have whatever features, but they also work better when you have multiple devices, and that’s very expensive (while delivering real value, which is another upsell).

And maybe this is an issue that impacts almost no one.

Just as I argue that most people are better off signing in to Windows with a Microsoft account, I will argue here that most people don’t know or care that they get Windows 11 Home or Pro on a new PC, it just works. Those that do care, in both examples, aren’t just a tiny minority, but they’re smart enough to figure it out and do what they want. You can get cheap product keys for Windows 11 Pro online in many places and upgrade if you need Hyper-V or BitLocker, or whatever else Pro has that you need. But most don’t care and will never do that.

My Surface Laptop 7 came with Windows 11 Home and it’s fine. I did upgrade it to Pro, however, not because I “need” any of its unique features–I don’t, and I have other PCs I can use if I did–but because I have to document this stuff for the book, and on the site to a lesser degree, and I need to know about all that. But that’s about as unique a use case as I can imagine. Which is sort of the point. It’s not an issue for most people. But like you, I’m not most people.