Earlier this year I was on Mastodon, opining casually on Apple being so tight-lipped about their strategy with their highest-end pro Macs, and suggesting that Apple just speak openly about the long term plans because there isn’t really much of an Osbourne effect to be concerned with, and it will help pros trying to make a long term home on the Mac.

I had mentioned John Gruber in the toot because it was discussing an interview he did with John Ternus, so he replied to me and said “Easy. Because Apple does not talk about future products.”

Which was a really funny response to get from John Gruber because he literally participated in a roundtable discussion hosted by Apple where they specifically talked about future products, specifically pro Macs.

But this is less about me dunking on John Gruber and more about this kind of thought pattern I see a lot now from prominent bloggers and podcasters who cover Apple. You criticize something about Apple and their response is that that’s just how Apple does things.

The most recent such thing to make the rounds has been surrounding criticism of Apple for releasing a new MacBook Pro in late 2023 that still has 8 gigs of memory in its base configuration.

Pundits are taking to macsplaining that this criticism is just futile because this is just how Apple does things. Apple specs these base model Macs to hit a price point that sounds good, but they will be stingy in how it’s configured so that you need to make a hundreds of dollars upgrade to get something good.

Of course, my issue here wasn’t that I didn’t understand Apple’s strategy here (and neither is anyone else who is calling Apple out for this). I know this is their strategy, and I’m saying it sucks. A computer with the “Pro” moniker should be pro-level without qualifiers. People buying a pro Mac shouldn’t be buying one with just 8 gigabytes of memory in 2023, because even if that suits their needs now, needs grow, and a MacBook Pro with 8 gigs of RAM will, on average, have a shorter useful life than a MacBook Pro with 16 gigs (since you can’t upgrade down the road).

It’s not a pundit’s job to defend a position Apple takes that’s so indefensible their own PR teams don’t even try to defend it out loud.

Apple’s products have historically had their differences that have been misunderstood, sure, and it’s not unreasonable to explain that (and I continue to appreciate that macOS has a distinction between an app being quit and an app being open with no open windows).

But to just plain defend Apple’s right to be shitty capitalists extracting every last basis point of profit margin just for its own sake? When that very same Apple was talking a big game in September about their ambitious environmental goals? I just don’t get why you’d want to waste paragraphs or air on it.

It’s true that this strategy of under-specced products at the low end has been a cornerstone of Apple product strategy under Tim Cook, and it’s fine to explain that historical context to your audience. And you can tell me that I shouldn’t be surprised at behavior like this because it’s consistent with years of their behavior. But you do not get to tell me I just need to accept that. These practices are worse for Apple customers, and if your primary audience is Apple customers, what good are you doing your readers and listeners telling them there’s no more in expecting more from Apple? What good is your loud voice if all you want to use it for is to help a $2.8 trillion company work their way to $3 trillion and beyond?

icanthascheezburger isn’t here to shill for Apple; I am here to understand Apple and to call them out on customer hostility. If Apple wants to make record-setting revenues they can do so by earning the revenue by making great products, not extracting it through nickel and diming on entry level configurations.

2023 was a relatively quiet year for icanthascheezburger.

It wasn’t a quiet year for me, though. I spent much of the year adjusting to the new reality that the most important person in my life decided to pursue another relationship and moved out. I had experienced a previous loss in momentum back in 2021 when I set the ambitious goal to publish 100 posts, and it was related issues that was taking the wind out of my sails.

The details and aftermath are mundane and not likely to be that meaningful to anyone but me. If you’ve experienced the same, you probably know what it’s like. In retrospect, I’m impressed I stayed as on top of things as I did. I was close to being a functioning adult as one could reasonably expect, but I was a depressed functioning adult.

The theme of 2023 for me has been “radical self-care.” Sometimes just continuing to exist from day to day was all I could handle. There were a lot of evenings where I remember just sitting in the hammock chair in my office, doomscrolling on TikTok and generally feeling numb, save for the comfort of being cocooned in the chair and having some cats curled up with me.

But I also did a lot of activities this year. I biked something like 150 miles this year which probably isn’t that much to a biking enthusiast but it’s more than I’ve ever biked in a year. I kept up with a commitment to regularly play retro games with one of my friends and we have completed a ton of games this year (more on that in a later post). I landed four different fixes into Ruby on Rails and even experimented with making a code change to Ruby itself (the PR didn’t land but I’m not mad; I got to learn how to play around with the Ruby C code).

I went to therapy. I relaxed in soaking tubs. I floated in sensory deprivation tanks. I even did some mushroom trips. I met new people and made some friends along the way (some people things didn’t work out with, but that’s okay). I went on dates.

And in the last couple months especially, I felt myself starting to turn a corner. And I found I had more I wanted to post on here, and so I just started posting more stuff.

Writing on icanthascheezburger is a virtuous cycle; it makes me feel good to post a lot of stuff on here, but I need to feel adequately good in the first place to be able to have ideas on what to write, and to write stuff that I like enough to post. And it was hard for me to do that in the last couple years while I watched the most important relationship in my life steadily decline from being really healthy and a solid foundation I could count on I knew was always going to be there, to becoming uncertain, to eventually not being there at all. When my employer announced in February (the week my ex left, actually), I joked to my manager that I could handle it just fine because I was accustomed to spending months uncertain if a relationship (work or otherwise) has a future.

icanthascheezburger’s direction

I still don’t think that icanthascheezburger has any particular major theme aside from “stuff Aaron’s interested in talking about.” I plan to overall keep doing that, but hopefully with more regularity. One of the biggest lessons I learned this year is that home is what you keep showing up to, and I want to make icanthascheezburger feel like a home to me.

I kind of look at John Gruber’s legendary Daring Fireball blog as aspirational. That’s not to say I want to make a DF clone. But as I look at the Apple bloggers and podcasters I’ve looked up to, I see this generation of now middle-aged straight white dudes whose values aren’t quite lined up with mine. I still respect their work, but I realize now that it’s okay for new voices to sprout up. It’s not like the internet has a limited number of column inches.

In particular, my adulthood has seen Apple ascend from being a relative underdog to being an incredibly powerful force, and though I know Apple’s history and humble roots too, I want to be louder about the stuff I think we should be worried about more, both Apple related and otherwise. The generation of Apple literati I’ve grown up reading and listening to were shaped by a different era of Apple’s history, one where you could unironically refer to Apple as beleaguered and in need of people to champion them.

So I plan to write about that more. I want to talk about major issues facing the world and the tech world. I’ll talk about topical stuff without feeling an urge to weigh in on every single event. I will be buying a Vision Pro and I expect I’ll be talking about it a lot. I may not have a post for everything, but I’ll post when I feel I have something worth saying.

I don’t expect to ever amass a following of tens of thousands or even thousands (if I can pull off three figures I’d be delighted, to be honest), but I’d love to put out some good writing and get people thinking more about stuff. But icanthascheezburger doesn’t have to become big. If it got big enough that I could make a living off of it, that would be wild. But I’m happy to keep it a hobby, sharing thoughts with my dozens of readers (even “dozens” is probably a stretch).

And barring that, this is a journal where I am sharing my body of writing work with the world, and people can watch me work on my writing craft.

Expect an uptick in posts as 2023 nears an end and I am getting more relaxation in, and let’s see if I can’t really hit that 100-post mark in 2024.

Finally, if you are reading this, thank you. It’s humbling to get to spend a little time in your brain.

As it turns out, “scary fast” was referring to the fact that the event itself was contained to a tight 30 minutes.

I noticed a few things in the event:

• Macs are available with a dizzying array of chip options now. Not only do you get to choose between M3, M3 Pro, and M3 Max, but also within those families you can choose different amounts of CPU and GPU cores. This is in stark contrast to just a few years ago when you got one or two chip options for later Intel Macs. Is Apple selling Macs with all these options because it loves offering lots of choices to customers? I doubt it; this feels more like Apple being stuck doing this because there are no other computer companies to sell M3 chips to and Apple needs to sell every possible chip on the die, and the best way to do that is to sell the ones with fewer good cores for cheaper.
• Last month during the iPhone event people complained that Apple was often comparing the new iPhone’s performance to some other years-old iPhone instead of last year’s. For this event, they compared the M3 models to the M1, but in their charts they provided benchmarks comparing them to both their M1 and M2 counterparts, which is a nice touch for those trying to keep track of the year-over-year improvements. The practice of comparing to a 2 generation old chip feels deceptive but it’s a reasonable comparison to provide; someone with an M2 Mac probably isn’t considering upgrading yet, but an M1 Mac owner might be. Apple even threw in a few comparisons to their fastest Intel models of some Macs, and the comparisons were staggering (on one benchmark the new M3 chips were 11x faster than their last Intel Mac counterpart). That wasn’t to knock Intel; it was more to point out to Intel Mac owners the gains they can expect.
• The M3 generation sees better speeds and efficiency gains overall, but the gains are uneven. Memory bandwidth on the M3 Pros got slower than on the M2 Pros, for instance. Also, interestingly, you will get six efficiency cores on an M3 Pro, but on an M3 Max you only get four of them in favor of more performance cores. As a user of iStat Menus on my own M1 Max Mac, I can see that the efficiency cores are very hard working, usually pretty occupied with work, since they are usually the default chip for most work unless there’s something harder to be done. It’s interesting that you lose those extra efficiency cores on the Max, meaning that if you choose a Max, it better be because you intend to put it through its paces.
• I’m really delighted to see that Apple is launching a new chip family with three different variants on day one. There are still other chips to be announced in this generation. The Ultras, which are made by stitching two Max cores together, will come later, and maybe this will be the generation where we see the “quad” chip made of four Max cores.
• In a similar vein, I’m happy to see Apple releasing new pro computers with new chips as soon as they have them instead of waiting. If you bought an M2 MacBook Pro that might be a little disheartening that it didn’t even last a year without getting bested, but the computers are still plenty wonderful.
• Apple specifically called out that the 24“ iMac is a great upgrade for both older 24” iMacs and 27" iMacs alike, which is a bit discouraging as a 5K screen lover. I think that bigger screens are still coming eventually (I have been patiently waiting for a 6K version of the Studio Display), but it seems like Apple is subtly trying to tell iMac fans that if they want a Pro iMac they should be buying either a Mac Mini or Mac Studio with a Studio display. And I do appreciate that such machines are at least a bit more modular, allowing the computer and monitor to be upgraded separately, but it’s a bit of a shame to see an end to higher end iMac configurations, especially since Apple Silicon could make them great overall computers.

I’m really happy with where the Mac is at overall. I can recommend just about any Mac model now, and the only caveat I need to provide is to avoid the base models for their lack of memory and SSD (which I’m sad to say is a caveat I’ve needed to provide for awhile now).

Also, as someone who used to own the black Intel MacBook, I have to say the new space black color looks really sharp, and I am delighted that Apple designed it in such a way that it is less prone to fingerprints.

About a year ago, I laid out some ideas for how Elon Musk could take a company like Twitter, which he recently acquired, and turn it into something that makes money and retains its value.

Elon Musk didn’t exactly go with that approach. Instead, he’s kind of been running it into the ground. There’s little value in rehashing his shenanigans, but suffice it to say, Twitter’s value is a fraction of its purchase price value, and what’s left of it seems like a miserable place to participate in, and the company seems like a miserable one to work for.

When Musk decided to shut down third party Twitter apps, Twitter was useless to me. If I couldn’t use Twitter via Tweetbot, I wasn’t using Twitter. After that, I started participating more actively on Mastodon, where I’ve had an account for several years.

Two exciting things then happened:

• Tapbots, makers of Tweetbot, decided to make a Mastodon client based on Tweetbot’s codebase. A couple months later, Ivory was born.
• A sizable contingent of the people I followed and interacted with on Twitter moved over to Mastodon.

Some days, it feels like I didn’t miss a beat, and when I scroll through things on Ivory I almost forget that it’s not Twitter.

But let’s dive into what life on Mastodon is like.

Mastodon isn’t about hyper-growth

Mastodon has been gaining users at a steady pace for a while now and there are now a total of around 15 million accounts.

If Mastodon were a startup funded by venture capitalists, this would be panic-inducing.

But instead, Mastodon is happy and financially healthy. Their latest annual report shows healthy financials, buoyed primarily by the fact that Mastodon as an organization runs super lean.

Mastodon (the organization) leads the Mastodon software project as a whole, and they run two of the largest Mastodon instances. I’d love to see them getting more money, and they are going to be getting it from donations from ordinary users like you or me.

For me, I don’t care if Mastodon hits 100 million or a billion users. Twitter didn’t become more valuable to me when it gained hundreds of millions of users. If anything, Twitter got less appealing to me as it got more growth-obsessed.

I hear a lot of nerdy-ish users lamenting that Mastodon isn’t just a little bit more like the incumbent social networks, being a little more algorithmic to help the onboarding experience and to keep users.

I hope Mastodon continues to reject that thinking, even at the cost of more users. I want Mastodon’s user base to be made up of people who value having agency over what shows up in their feeds. It’s still possible to toot something that goes viral; it just needs the help of people with lots of followers boosting your toot. And honestly, that’s a more honest way of going viral than getting lucky and having the algorithm tip the scales in your favor for something you made.

Instance drama

I’ve previously described how Mastodon addresses a lot of issues by structurally being different by not being one monolithic Mastodon service, but many Mastodon instances that work together.

Big monolithic networks like Twitter want growth so badly that they try to appeal to everyone, and aren’t always the best at moderating extremists who make others feel not safe to exist. I’m happy to say the worst kinds of users Twitter tolerated aren’t tolerated on the mainstream fediverse in general. Sure, you can make a Mastodon instance composed of white supremacists (and people have), but no one’s going to want to federate with you and you’ll be isolated and not able to recruit people from the mainstream.

To be clear, I’m not saying this solves all of Mastodon’s problems and that it’s free of racism and such. It’s not, but this feels like a more solid foundation.

Instances on Mastodon have agency over how they moderate, and that includes the freedom to defederate from other Mastodon instances, preventing the instances’ users from being able to communicate. Defederation is a sharp tool and personally, I feel like it should only be an absolute last resort. But I’ve seen instances defederate others because of beef between admins.

Now, if you run the instance and you’re paying for the infrastructure, you are well within your rights to defederate as aggressively as you like. But I’ve seen more than a few users end up in situations where they’re on an instance that either defederates or is defederated a lot, and then they got fed up with losing connections with people.

If you don’t like a decision your instance admin makes, you can move over to another instance pretty smoothly, but the migration process isn’t perfect (for instance, your historical posts don’t move with you, and if your instance defederated with another, those follower relationships won’t be able to move because they aren’t there).

I’d like to see Mastodon making some bigger investments in account portability for the sake of robustness of the fediverse overall. Not only is there drama among people who run instances, but also some Mastodon instances just don’t stay around. That saddens me; lots of toot URLs will eventually be dead links. One advantage Twitter had was that every public tweet was accessible by a URL (though recently Elon Musk ruined that too).

The reality of Mastodon is that the people running Mastodon instances aren’t paid professionals and they aren’t backed by a big community safety team. To some people, hearing that is a relief and there are finally some instances where some groups of people feel safe (I know a lot of instances that are substantially safer for LGBT people to use, even when they’re posting racier content). But also, the fact that these are run by amateurs means that we’ve got a lot of instance admins who are not prepared to handle the issues of moderating a community, and in many cases instance admins aren’t adequately protecting people of color.

For now, my recommendation for most people is to just make an account on mastodon.social, and if they want to move later to something that suits them better, they can.

Bluesky and Threads

Most of the recently announced microblogging apps don’t have much traction, but two seem to have decent backing: Bluesky, and Threads.

Bluesky was originally an idea of Jack Dorsey’s back when he was running Twitter, and the idea was to create an open protocol that would support Twitter-like features, and then Twitter would migrate to becoming a participant of that open network.

Personally, I’ve always been a little suspicious of the motives behind Bluesky; if Dorsey really did want to move to something open and decentralized Mastodon and ActivityPub had existed for years when he announced Bluesky.

It’s also become clear that Dorsey is hoping that with Bluesky he can wash his hands of having to moderate and community manage the network, which is always the kind of material work that tech bros wish didn’t exist, so they like to just pretend it doesn’t have to exist. But the truth is that online communities need moderation and care, and that takes real human effort that you can’t just replace with an algorithm. People continue to try and fail, and in the age of large language models I’m sure people will try again.

Bluesky opened up a beta app this year, and it’s got some traction, but nobody’s making third party apps for it so far, and it seems to be focused on an algorithmic timeline (albeit one that has some customizability, and there is a promise that you will be able to pick a chronological timeline). If Bluesky does indeed take off I hope Mastodon adds interoperability with it or vice versa so that I don’t need to have another separate account just to see what my friends on Bluesky are ~posting~ skeeting (sighs yes, really) about.

Facebook released Threads as an Instagram product earlier this year. Threads initially made a big splash, in part because when you have the engineering resources of Facebook you can crank out a large scale app like that at incredible speed, and also it seems like they flat out lied about engagement and numbers by just counting Instagram users as Threads users, even if they weren’t using Threads yet. Recently Mark Zuckerberg has claimed Threads has 100 million monthly active users, and that sounds like a big number but that’s just the number of people using it at least once a month. That’s an… odd metric to cherry-pick and it’s likely it’s only being highlighted because it’s the only one that sounds good.

Interoperability

One reason I’m ostensibly rooting for Threads is that they are promising ActivityPub interoperability, meaning in the future we can (theoretically) follow and interact with Threads accounts on Mastodon (and vice versa).

That’s actually kind of exciting to me. Mastodon is still a little bit on the nerdy side, and a lot of non-nerdy people I know are probably way more likely to become Threads users than Mastodon users. With ActivityPub support my friends can follow me from Threads and see my musings, and I can follow them, and neither of us needs another account or to even leave our respective social apps.

Better yet, if a friend is ever talking to me about Threads and complains about something like the number of ads, I can make a great case for switching to Mastodon, and they can actually move but still be able to follow the people they currently follow on Threads.

Over time I’m hoping that interoperability will become table stakes for social networks, and I’m likewise hoping that new social networks coalesce around ActivityPub (which is what Mastodon uses).

Here are a few different projects that interoperate using ActivityPub:

• Pixelfed (picture sharing network like early Instagram)
• PeerTube (video sharing like YouTube)
• Lemmy (link sharing, like Reddit)
• micro.blog (a microblogging site like Twitter or Mastodon)
• WordPress via the ActivityPub plugin (icanthascheezburger is actually using this plugin now and you can follow it via ActivityPub at @[email protected]).
• Pleroma (microblogging)
• Calckey (microblogging)

Tumblr (also owned by Automattic, makers of WordPress) has made public commitments to add ActivityPub support but it’s unclear whether that’s still happening.

Clawing back control

In short, I was able to replace a product made by a company worth tens of billions of dollars (as much as $44 billion if Elon Musk is to be believed) with one made by a company that brings in less than a million a year in donations.

And the products I’m using (Mastodon via mastodon.social + Ivory for Mac and iOS) are better than what Twitter can offer. There are no ads. The apps aren’t littered with tracking code and other useless.

And because Mastodon is philosophically different, it’s fostering a different clientele; one that is more aligned with my values on these things.

There are still areas where I want to see Mastodon mature, but when I step back I’m really proud of what we accomplished as a small community. Through community action a chunk of us have managed to reject a mainstream social network and make our own better one.

I look around at all of the other ways the tech world has gotten shitty. Streaming services were great but they’ve all been consolidating and jacking up prices like they’re cable companies. Every web site I visit is littered with ads and modal popups, and the dreaded “we noticed you’re using an ad blocker” notices. Too many of my apps are sending me push notifications that are nothing more than “hey, open up this app!”.

But Mastodon’s not trying to sell me anything. It’s not pushing for me to follow more people or interact with people just to juice up engagement numbers. It’s just there, humbly at my service.

Sometimes I feel like so much of software is so entrenched that we’re stuck with it as-is, but it turns out that isn’t true. Sometimes a company will just make its stuff unusable for you and you leave (Twitter, Reddit), or you just learn that this is all made of software, and it’s actually not as hard as you might think to build something new and use that.

And I take great comfort in that.

Ozempic’s been getting a ton of attention in the media over the last year or so. If you’re not familiar, Ozempic (generic name semaglutide) is a relatively new drug that came on the market a few years ago to treat diabetes. One of its side effects is loss of appetite and subsequently weight loss, and as a result, a flurry of doctors started to prescribe it off label for weight loss. Now, Novo Nordisk is marketing semaglutide for weight loss under the brand Wegovy.

And as with anything that gets popular, every news outlet has been running pieces about Ozempic nonstop in a rush for clicks. Clickbait factories like Forbes are naturally name-dropping Ozempic in articles not actually related to Ozempic. We’re getting pieces about how Ozempic is killing industries. Because Ozempic curbs appetite, some stock pundits have been worried that food sales are going to drop as a result.

And those are all kind of stupid and overblown, but they’re not nearly as tedious as the thinkpieces that suggest things like “the end of obesity” or perhaps even stupider, “the end of obesity as an identity issue”, or that we’re about to usher in a “super skinny era”.

Ozempic is absolutely a great drug. I’ve taken it to help manage diabetes, and it has been a really effective tool to help me keep my A1C in check. I have experienced Ozempic’s weight loss side effects, and I suspect that the weight loss in turn helps make the diabetes easier to manage, a nice virtuous cycle (and one that a lot of diabetes drugs fail to achieve because they decrease blood sugar at the expense of weight gain). My digestion gets slowed down, which keeps me feeling fuller longer as well as slowing down blood sugar spikes after meals. I get full a lot more easily. My fasting mimicking diets easier to do when I’m taking Ozempic. And in some ways, I feel like Ozempic has changed my relationship with food to be one that’s less obsessed. Food remains enjoyable, but I feel like my joy for food is more tempered and less obsessive. I can lightly snack on something tasty and not find myself mindlessly consuming it. I can be satisfied more easily. Ozempic also reduces cardiac events, and that’s not just a side effect of weight loss (semaglutide is also available in pill form in Rybelsus, where it causes weight loss but doesn’t have the same cardiac benefits)

But this idea that Ozempic is going to make obesity disappear is absolute bullshit. Even if it was priced to be ubiquitously available and there were zero side effect concerns, overall people see modest weight loss (think like 10% of body weight). In clinical trials there were participants who lost a decent amount of weight, and I have lost a noticeable amount of weight, but I’m still plenty sizable. Ozempic isn’t going to decouple us from obesity the way the pill decouples sex from pregnancy risk (a real comparison from one of the above linked articles). When we’re so awash in weight loss snake oils that do nothing, when something with any real level of effectiveness like Ozempic comes along, it’s heralded as a game changer, modest as its effectiveness may be.

But of course, the pundit’s muse is not fettered by such inhibiting factors as reality. When something is popular and you put it in your headline, you’ll get views, no matter how inane your article is. When pundits see technologies that are obviously transformative in hindsight, they can’t help but want to be the one to call the next big revolution, so they just start applying that to everything that looks like it has traction, in the hopes that one does hit it big, at which point in time they will look smart. And as a result, we end up with outlandish claims about how Ozempic will transform society; after all, if you turn out to have made the right call, you’ll end up looking really insightful in the future, even if it was just a lucky guess.

I’ve been using ZSA’s new Voyager keyboard for a few weeks now.

The first thing that hits you when you unbox it is how damn thin it is. It’s not just thinner because it’s using low-profile Choc switches (and similarly low-profile keycaps), the actual case of the keyboard itself is considerably thinner than even the Moonlander (whose thinness I was impressed by as well).

This keyboard was made with portability at top of mind. Its footprint is tiny and minimalist compared to the Moonlander (which itself is capable of being quite small on a desk when the wrist rests are removed), and it’s quick to set up with tenting legs that magnetically attach to the bottom of the keyboard with a decidedly satisfying click.

Build quality is quite possibly ZSA’s best yet. Everything feels built to tight tolerances, and the keyboard strikes a good balance between feeling solid but remaining light, with the case’s bottom being metal and the plastic of the keyboard case and the key caps feeling perfectly nice. The keys all have a solid feel with minimal wiggle or play. The keyboard feels properly premium.

If you are accustomed to full-sized keyboards you might find the lower profile to be a bit jarring. The keyswitches have the premium feel and full-depth travel that you would find in other mechanical keyboards, but the keycaps are flat and even, not at all unlike a laptop’s keyboard.

Customization: Layout

If you are looking at a photo of the Voyager, you might look at the key layout and think that makes no sense to you.

And that’s okay, because the Voyager’s key layout is whatever you want it to be. It’s a fully programmable keyboard. And when I say the keyboard is fully programmable, I mean the keyboard is fully programmable; this isn’t some janky third party software you have to install on whatever device you want to use it with. When you program the Voyager keyboard, the layout you program becomes the keyboard’s real layout.

You might also look at the keyboard’s spartan 52 keys and find that intimidating, and you might be imagining a weeks-long learning curve to get up to speed typing. But in reality, you’ll pick it up very fast! On a small keyboard like this, different keys will be doing double duty (or more) and since there are only 52 keys to learn, your fingers will develop muscle memory fast.

You can customize the keyboard’s layout using ZSA’s online configurator tool, or you can use the new Keymapp app (which doubles as the app you use to flash the firmware to the keyboard once you’ve decided on a layout).

Keys can be customized in a number of ways. Of course, you can assign keys on any position on the keyboard. But the real power in the Voyager’s programmability is in the extra capabilities you can give keys, which is important on a keyboard with just 52 physical keys.

You can have multiple virtual keyboard layers and assign a key that will switch the keyboard to that layer (and you can make that work like a shift key, or a toggle where tapping the key switches to another layer, and tapping again switches back). Every layer is its own full layout, and if you want a layer where just a few keys are different, you can do that and leave the remainder of the keys as-is. By default a key on another layer is transparent and will just fall back to whatever the same key does on the layer below. If you typed a lot of German, for instance, you could have a sort of shift key that shifts vowels to umlauts but leaves everything else untouched.

And even within a layer keys can do more than one thing . You can have a key that does one thing when you tap it and something else when you hold it, like my key that’s Esc when I tap and Ctrl when I hold. You could set up a key that does one thing when you tap it and another thing when you double-tap it (again, potentially handy if you type in languages that have letters with diacritical marks). You can even have keys that can perform an entire keyboard shortcut in one keystroke. No finger acrobatics for you!

And the amazing thing is that once you realize that you have the power to just change how your keyboard works, it gets really addicting to just pop into the configurator and make that change. When devices are this willing to be customized you can’t help but feel delighted.

And since each keyswitch has its own RGB light beneath it, you can fully customize key colors if you like. My favorite use case is my number pad RGB layout that helps me easily tell which numpad keys are which.

My Particular Layout

Of course, with a fully programmable keyboard, you know I can’t make things simple, and I am not using the Voyager’s stock layout at all.

My layout is based on the Planck keyboard’s standard layout. The Planck was the first ortholinear keyboard that I really got into typing with, and all of my ortho keyboards use a layout that is based on my lightly customized Planck layout.

This is a pretty typical Planck layout:

(image courtesy of https://github.com/Felerius/planck-layout)

The Planck lacks a dedicated row of number keys.

In its default layout, the Voyager does have a number row:

But I don’t need a dedicated number row, so my Voyager layout is based on a Planck layout, but I make use of the four extra keys.

At first this felt a little weird and unnatural, because the keyboard was made with the intent that the home row is a row lower. The thumb cluster keys feel like a bit of a reach, and the inner bottom corner keys on the main cluster are easy to accidentally hit with my thumb.

After about an hour of using it, though, I was accustomed to my modified layout and I don’t accidentally hit those corner keys much, and I assigned them the / on the left and ? on the right (and I have keys with similar roles on my Moonlander).

Customization: Hardware

The Voyager offers versatility, just like its bigger Moonlander brother. Out of the gate you can buy a tripod mount kit for your Voyager, which attaches magnetically to the bottom of the keyboard. This is handy if you want to have a bespoke setup at your home desk, but also want to be able to pick the keyboard up and pack it quickly.

I haven’t yet gotten ambitious enough to try some of these more advanced mounting options (though the chair arm mount option sounds interesting to try).

The switches are hot swappable, so if you decide you want your typing to have a whole different feel, you can. You can even mix different switches on different keys (I’ve done this on other keyboards where I use a switch with a heavier spring on the easy-to-reach keys, and a lighter switch on the ones further away, so it feels even). You can switch out your keycaps too, but Choc switches use a different style of key cap and there isn’t the same vast selection out there. But that’s okay because the stock key caps feel excellent.

The Voyager’s magnetic feet actually use magnets that can be removed from the feet to be used as donor magnets for custom mounting gear you might want to design yourself or 3D print.

So many consumer electronic products are designed by companies that just want you to blindly consume them, exactly as-is, never thinking about the other things they could do. Cracking open a product’s case usually voids the warranty. But the Voyager, like all of ZSA’s keyboards, bucks that trend. It wants you to customize it. And I really welcome products that challenge the status quo of our relationships with the everyday tools we use.

The Moonlander, To Go

If you’re in the market for a great ergonomic desktop keyboard and you’re not afraid of a learning curve and you like to be able to customize just about everything, the Moonlander remains my default recommendation (and I am happy to say I know several happy Moonlander users that got one on my recommendation). I think if you’ve got the space for it, you’ll get the best typing experience on a keyboard with full-height switches. I also like to use sculpted key caps to help my thumbs differentiate the more far-reaching keys from one another.

But if you want a travel keyboard or prefer a smaller footprint, the Voyager may well be for you. It’s a smaller keyboard, and it does have fewer keys, and that might be intimidating at first, especially since the Moonlander’s 70 keys already might feel like too few, but believe me, you will be surprised at how little you miss the extra keys once you have a layout you like (in fact, when you go back to a fuller-sized keyboard you’ll likely be annoyed with the extra keys getting in your way).

But the Voyager manages to retain just about everything that makes the Moonlander a great keyboard, retaining incredible customizability and comfort while shedding a lot of volume. It’s a keyboard you’ll happily set up at your desk, and because it’s so travel-friendly you’ll be happy to be able to bring one more comfort of home with you.

The Voyager isn’t an impulse purchase (unless you’re a crazy keyboard enthusiast like I am); at $365 it’s a commitment of a purchase, but as soon as you unbox it you’ll immediately feel the build quality, and ZSA backs it up with a solid 2 year warranty with support people who are real human beings who are always a delight to talk to when you have trouble with something.

And best of all, this isn’t some crowdfunding campaign run by someone who never made a keyboard before where you’ll pledge your money and then get signed up for a newsletter about a team who learns from first principles that manufacturing stuff is hard and prone to delays; ZSA was taking orders the day they announced the Voyager and they were shipping by the end of the week. As of right now if you order one it’ll ship within a few weeks.

Overall, I really enjoyed this WWDC keynote, and it feels like one that belongs to the ages (after all, it’s not every WWDC that gets a brand new platform announced). WWDC events nowadays are a massive info dump sure to overwhelm every time simply because there are so many platforms now.

I want to focus on the two specific parts of the keynote that I made posts about this past week.

Mac Pro

I figured there was about a 20% chance we’d see an Apple Silicon Mac Pro announced today, and we did!

Sadly, it’s a bit of a disappointment. I outlined three different approaches Apple might take to transitioning the Mac Pro to their own chips, and Apple sadly took the least ambitious approach of all, keeping the Mac Pro with the same general compute and memory limits of its other machines.

It’s a real disappointment that the very best Mac Apple sells now has a maximum memory capacity an eighth the amount of the Mac Pro that Apple announced four years ago (to say nothing of the fact that you’re stuck with memory soldered onto the board instead of being in slots for expandability). Gone are the highly innovative MPX modules that brought new dimensions of expandability.

Instead, Apple is apparently content to let the Mac Pro not try much of anything new, and spend the next several years catching up to its Intel counterpart in key ways.

This could very well be a transitional machine and Apple may very well have something more revolutionary up its sleeve for later, and I hold out hope that we might see that in a couple years.

Vision Pro

First off, kudos to Apple for coming up with a marketing name that was in no rumors at all.

For not having given much thought to AR/VR though, I currently feel like my prediction from yesterday is going to hold up well.

Even Apple can’t get past the dorky factor, though I believe it’s a nice piece of hardware. The fact remains that it’s awkward to have a headset on, and you need a damn compelling use case to convince the masses to wear something like this.

And to Apple’s credit, the OS does indeed look beautiful and it looks like the UI elements and interactions are well thought out.

But that’s not what’s holding AR/VR back. If there was indeed a use case that only AR/VR could serve, and if it was compelling enough, people would be buying VR headsets already and suffering through the clunky interfaces. But interest remains tepid.

Apple doesn’t seem to have put forth a killer app either. They seem to hope their killer app is “take general-purpose computing to have a radically spatial UI that’s nicely executed.”

Things like 3D video recording (and viewing) and substantially more immersive Disney movies sound nice, but none of them are nice enough that you’d buy dedicated hardware just for them.

Apple is marketing Vision Pro and pricing it like a brand new general-purpose compute platform. Apple’s track record with building up brand new compute platforms is a little shaky, though. The iPad is Apple’s most recent attempt at building a brand new general-purpose computer platform, and iPadOS has evolved at a glacial pace compared to the Mac’s evolution. Some concepts were a slam dunk for Apple (like their cursor support), but other things, like general-purpose multitasking, are these multi-year blunders Apple still hasn’t quite gotten right. And this is for a computing device that has tons in common with the Mac, in that both are 2D interfaces using a computer screen and a few different input methods. iPad has had close to a decade and a half to evolve and I couldn’t depend on it as a platform to get much of my real work done on. I have serious doubts about Apple’s ability to get VisionOS to a point where it’s a contender, to speak nothing about my hesitancy to fully embrace a platform where the only way to get apps on it is probably the App Store.

And looking at this OS, there’s still a lack of fundamentally new UI ideas. Apps are largely stuck in 2D windows that are projected into 3-space. That’s a bummer! If you are spatial there’s nothing stopping an app from being split up into many things to drive the metaphor home (hell, early Mac apps like Photoshop did that too). I don’t need a 3D headset to use apps whose UIs are primarily 2D; I can already do that on existing computing devices.

I’m not dismissing this product out of hand because I’m wiser than Steve Ballmer (if I’m being honest, I want one now more than I imagined wanting one yesterday). But I remain convinced of what an incredible long shot this product is for Apple. It will probably be a few years before there’s even a version of it I’d consider to be a proper v1 for an average consumer (it would need to come in under $2000, and probably even under $1000 to have a chance to sell in volume). And to use one is a bigger behavioral change than any computing device to date.

On the eve of WWDC I’m going to share my hottest of takes: there’s a decent chance that Apple’s VR headset ends up being a product that Apple tries but will ultimately exit the market.

It’s not that I think Apple will do a bad job with this product; if anything I think their execution is going to set the bar for the product category, and like the iPhone I think that future VR headsets from competitors are going to magically start looking and feeling like whatever Apple makes.

I’m bearish on VR itself. Specifically, I’m bearish on VR as a mass market product.

It’s interesting to me that Apple is entering this market. They’re very conservative about entering new markets, even though I think there are dozens of product categories they could do great work in. Apple may be the only company capable of getting VR to properly take off as a technology for the masses.

But a lot needs to happen for VR to really take off. Naturally, it’s going to need to get smaller and much cheaper, but I am confident Apple will solve those tech problems in the coming years if VR shows traction. But VR needs two very big things to happen to succeed, and they are tall orders.

First, VR needs a killer app. There needs to be an application that’s 1) only possible with VR, 2) is so compelling that people will literally buy the hardware just to use the app, and 3) is widely popular. VR will probably need a few killer apps. But a killer app for VR hasn’t surfaced so far elsewhere. And VR headsets have been pretty competent for years now, to the point where it’s been possible for someone to build a killer app on them. But collectively, I don’t think anyone has ideas for something that’s any good, and I don’t think that a better headset product from Apple will fix that.

Second, we’d need to make the technological leap past these headsets and toward something like the glasses Apple is rumored to also be working on. If there’s a good enough app people will tolerate the relative clunkiness of a headset, but eventually people are going to want something that’s more portable and able to integrate into everyday life. Not only is that going to help make the device feel less dorky, it’s going to open up use cases that aren’t feasible with the headset. With today’s tech, even with Apple’s access to very fast custom chips, just the headset itself is proving to be challenging. It’s hard building something with the display quality Apple demands at the tiny size needed for a headset. Transforming from that to something that can project an image onto transparent glasses lenses feels like the kind of effort that’s at least a decade away.

Apple historically has been great at taking products that exist, have traction, and that incumbent companies suck at making, and then reinventing the entire category. They did this with computers multiple times. They did it with personal audio players. They did it with the smartphone. They did it with tablets. They kind of did it with the smartwatch.

With VR, Apple needs to kickstart the entire category. And it’s odd because VR isn’t brand new. Companies have been working on it for years, and a lot of VR tech is quite mature now.

The other thing: Apple’s a big, big company. Their minor products have the sales of Fortune 500 companies. Around 9.6 million VR headsets were sold in 2022. If Apple sold 10 million a year, all for $3000 a pop, it would still be a relatively small business for them. When you’re bringing in upwards of $400 billion in revenues annually, your threshold for success is remarkably high. Hell, given the years this product has spent in the R&D phase it will face an uphill battle recouping that investment.

I’m smart enough that I know not to bet against Apple. However, if they are announcing a VR headset tomorrow at WWDC (and at this point I have to assume they are, because if they weren’t they would have quietly leaked that to the press to tamp down expectations), then this will be Apple’s biggest and riskiest bet to date.

And even if Apple doesn’t succeed at this, I like them being bold like this.

In the “free idea for Apple” spirit of my SwiftScript post from awhile back, I want to share a vision for the kind of AI-powered product Apple could incorporate into macOS and iOS: Spotlight Bionic.

Spotlight was pretty revolutionary when it was first announced back in Mac OS X Tiger. With really good indexing technology Spotlight made it possible to do full-text search on basically your entire Mac, and it was fast.

Apple has augmented Spotlight with some extra functionality over the years, but fundamentally the core Spotlight feature is pretty rudimentary. You can search your entire Mac for things, but you have to be searching for pretty much the exact thing that’s in the files in question to be able to find it.

Throw a sophisticated LLM-based AI into the mix, though, and that changes the game substantially. You’re still ultimately querying the same dataset (i.e. the data on your Mac), but an LLM lets you query the data like an actual human being.

One of the early ways we could explain the magic of computers was stuff like full-text search. You could be editing a big text file, trying to remember where you used the word “blue,” and because computers are fast, you’d be able to find the word almost instantly. Originally this was very inflexible and you needed to search for the exact string, but it was still a big leap forward and search has gotten more flexible as computers got faster, allowing you to make “fuzzier” queries that may not be the exact matching text but are close enough that the computer can find the match anyway.

AI brings us to a point where we can do that same kind of searching, but without the “you need to search for the exact thing” restriction. And that’s revolutionary.

That means, for instance, you could ask Spotlight Bionic for all of the emails from your mom where she sent a recipe, even if your mom didn’t specifically ever use the word “recipe” in these messages.

We already have machine learning powered image search in the Photos app, but Spotlight Bionic could supercharge it, letting you ask “what’s that Keynote presentation I made that had the Simpsons meme in it?” (As if I would have just one…)

Traditionally we’ve worked to overcome computers’ rigidity on search by attaching metadata to files, or by putting files in folders with a strong organizational structure.

But that strategy has limits. Maybe as a high school student you keep all your American Lit book reports in a folder of that title, but if you failed to one time and you don’t remember what you named the file, you might not be able to find it unless you had something concrete to go off of. There’s nothing wrong with that, but AI can help you find stuff without a need for that. Maybe it could even help you see relationships between different files on your computer! Imagine how handy it might be if you’re writing an essay on banking regulations, and you know you’ve got some prior work where you wrote about similar stuff, and Spotlight can just show you those documents.

AI search can get more capabilities too, if it’s trained to understand the relationships things on your computer have with one another. Maybe you want to do a search for recipes from your mom, but you only want it to show you recipes where you replied to the email and said you wanted to make that recipe. In theory, a robust enough email app could give you a querying language that could let you search for specific emails, like ones from your mother, then from that search show you a list of your replies, but it would be a complicated query (and without AI you’d still need to query for words you know the emails have, like “recipe”, for instance). But if you give the AI the ability to understand and see those relationships and map those relationships to human language, then this becomes something a computer can easily automate.

Apple’s really well-positioned to offer this, too. Apple Silicon Macs have chips specialized in machine learning, so all of this can be indexed by your own Mac instead of in the cloud, which is a win for privacy and it’s a win for Apple because they aren’t paying for all the server infrastructure that would be needed for that.

Thinking of this kind of AI as a search capability is still just scratching the surface of the capabilities it unlocks. If you generalize this AI to “a general-purpose way to translate human language into something a computer can execute” (which is essentially what it would be doing when you make a search query; it’s just executing a search), you can actually start to talk to your computer like a human and have it perform complex tasks for you. Siri goes from being a tool that can set timers and give you sports scores to being able to take those recipes from your mom and build a Pages file of a cookbook that you could then edit and send to have printed as a gift to family members.

At Microsoft’s Build conference this week while showing off features like this, Satya Nadella said that with the help of AI, every user could become a power user, and that’s really the thing that excites and energizes me about adding AI into existing products. If we can make AI good enough (and this is still a big “if”; AI demos today are super rough), we will be living in a world where my specific skill of being a computer power user might become somewhat obsolete (or radically evolve). I’ve always been able to do cool and useful things with computers because I took the extra time to be able to work with my computer and communicate to it what I need, and the computer can reward me with productivity. But soon we might be able to cut that middle man out, and people will just be able to ask their computers and devices what they need and the computers will do the heavy lifting of figuring that out and translating it to computer-speak.

Everyone’s busy thinking about Apple’s upcoming VR headset, but I’m more interested in a product Apple has publicly stated is coming: the Apple Silicon Mac Pro. I would be surprised if it’s actually getting announced at this WWDC, but I hold out hope that Apple’s making something great.

The current Mac Pro was released in late 2019 and is vastly different from its contemporary Macs, in that the entire system is a modular tower-based desktop. The video card is removable. The memory is removable. You can add third party hardware cards to the computer at will.

The Mac Pro has this modularity for two reasons. First, it’s vastly more expandable than the other Macs. You can configure your Mac pro with up to an astounding 28 processor cores, and up to 1.5 terabytes of RAM. You can install incredibly powerful graphics cards in the machine. The second reason for this modularity is to accommodate pro workflows that involve installing custom hardware in the Mac.

These are use cases reserved for the most demanding of pro Mac users. Apple doesn’t break down Mac sales figures, but I would be surprised if the Mac Pro ever topped 1% of the Mac’s total unit sales.

John Siracusa makes the case more eloquently than I ever could in The Case for a True Mac Pro Successor. The Mac Pro isn’t a practical computer for Apple to make in terms of the money it earns them, but it’s important to the Mac overall that as a software platform it will scale from a being a simple personal computer to being used in advanced scientific research. It’s aspirational in many ways, but from a practical standpoint it’s enticing to know that you can use the same computer to do all of those things.

In the years between Apple’s roundtable discussion about pro Macs and June 2019 when they finally announced the machine, there was intense discussion among Apple bloggers and podcasters wondering whether Apple was going to bring a truly no-compromises machine to the market, or if the next pro Mac was going to be akin to the 2013 Mac Pro (lovingly and not-so-lovingly referred to as the “trash can”). These tone of these discussions tended toward pessimism after seeing years of Apple making missteps with the Mac lineup, compromising utility of pro Macs in favor of aesthetics that pros weren’t asking for.

Pundits were right to be extremely skeptical of Apple’s commitment to pro users at the time. Not only was the trash can Mac Pro poorly received for not suiting people’s needs, it was languishing because its design was suboptimal for more powerful GPUs. The MacBook Pro got revamped in 2016 after years of stagnation only to be worse for pro users in key ways (not least of which was the choice to use the butterfly switch keyboard in these laptops). Apple had been showing some promising signs that they were listening to concerns, but it remained to be seen whether they would make a pro level machine that would satisfy the most pro users.

That pessimism turned out to be unfounded because the 2019 Mac Pro was revealed to be a machine that compromised on absolutely nothing (with a price tag that also didn’t compromise). It checked off every box, with plenty of expansion slots, support for high-bandwidth MPX module cards to support hardware that had more data to move, a processor with up to 28 cores, and even a rack-mount option. It threw in a couple USB-A ports in addition to a handful of Thunderbolt ports (including two on the top of the computer) and it was highly user-serviceable. The entire machine was designed for thermal load with a beautiful lattice of round holes in front to maximize airflow and internals that were designed to ferry air through the machine.

The 2019 Mac Pro was a love letter to pro users, and it makes no sense to me for it to have been a one-off computer. After hearing John Ternus confirm after announcing the M1 Mac Studio that they were still making an Apple Silicon version of the Mac Pro, I became even more convinced that Apple’s next Mac Pro is going to be free of compromises in the same way.

Apple Silicon Architecture

All of Apple’s existing Apple Silicon Macs are a tightly integrated affair. The M1 chips are entire systems on a chip (SOCs), with CPU cores, GPU cores, neural engine cores, and memory all in one unit. That makes sense given that the roots of Apple Silicon are in iPads and iPhones.

This is an incredible design for all of the Macs Apple has transitioned so far to Apple Silicon. Laptops get incredible battery life due to the power efficiency. The computers themselves get to be incredibly sleek, like the new iMac. Having one big pool of RAM that is used by both the CPUs and the GPUs saves space and money, and it’s faster too because there’s no time wasted copying data from RAM to the GPU.

For Apple’s more powerful laptops and desktops, Apple scales up their chip design by fusing together multiple SOCs.

With this approach Apple can make machines with as much as 128 gigs of memory, 20 CPU cores, and 64 GPU cores. That’s a lot, but it doesn’t come close to how generously you can configure a Mac Pro.

So what’s Apple to do?

Option 1: Use the same architecture and take the L

The simplest thing Apple could do here (and very much a prevailing theory among those who speculate about Apple) is to release a Mac Pro that contains the M1/2 Ultra chip, perhaps with an optional Extreme chip that is 2 Ultra chips to double the number of cores and maximum memory once more.

The machine could still be offered in a Mac Pro looking case and maybe there would be PCI slots still for pros who need to add more cards for connectivity and such.

The problem, though, is that Apple would be replacing the Intel Mac Pro with a Mac Pro that is substantially less capable in key ways. Yes, the unified memory might be faster, but that’s not going to mean much if your workloads demand a terabyte or more of memory. There’s also a GPU issue here too; although Apple’s in-house GPUs are really power efficient, they aren’t as fast as the best graphics cards you can get from AMD or Nvidia.

This would be the most cost effective strategy for Apple, though, since it would mean they can leverage their existing chip designs without doing any radical new designs that will only be used in one machine. If Apple went this route again we might see them revisiting the 2013 Mac Pro philosophy to upgradeability and modularity, where customers were encouraged to augment functionality by attaching external devices via Thunderbolt.

Option 2: Use a traditional PC architecture

Another option Apple has here would be to make a Mac Pro that is essentially identical in architecture to the Intel Mac Pro, but it swaps out the Intel CPU with an Apple Silicon CPU.

In this case the chip could be pared down, maybe losing or deactivating the integrated GPUs and memory in favor of the RAM slots. Maybe it would be an option to have a mix of unified memory and extra memory in slots so users who don’t need a big GPU wouldn’t need to purchase one and could rely on the GPU built in. There would continue to be MPX modules for installing high end cards and several PCI slots for other hardware.

This makes an Apple Silicon Mac Pro that’s no-compromises on specs and upgradeability, but it would be missing out on the benefit of unified memory which Apple has been touting as a huge advantage of Apple Silicon. It would seem weird to me to build a machine without that.

Option 3: Deconstruct the SoC

Finally, there’s the more ambitious approach: instead of trying to cram everything into these tiny SoCs, Apple could instead build a custom Mac Pro logic board that applies everything Apple has learned about making fast, tightly integrated chips, but implemented on large footprint with modularity.

Instead of having memory soldered on, Apple could instead build a very fast interconnect to a bank of memory slots, saving us from needing to choose between having memory that’s fast and without latency, and being able to have an absolute crap ton of memory. In the same way that Apple’s MPX modules have high bandwidth connectivity to the rest of the system, Apple could make a successor to MPX that allows installing high-performance graphics cards with full-bandwidth access to all of the system memory. Having slots be this high-performance also gives brand new options for expandability. If you want to be running a lot of ML workloads, you could install a card with a massive array of Neural Engine cores, all of which have direct memory access.

We could take this a step further and make every major part of the SOC slotted, including the CPU cores, giving customers total ability to customize whatever part of their system is needed by their workloads. Need more CPU? There’s a card for that. More GPU? There are cards for that! Need more disk I/O? There’s a card for that too!

What will Apple choose?

I speculate that we’ll see Apple choose Option 2 or 3, or maybe some hybrid between the two.

Of course, Option 3 would be the biggest R&D undertaking and it would be the biggest stretch for Apple because it would mean developing a completely custom architecture just for the Mac Pro, and that architecture is not likely to be one that would scale down to their other machines very well. Also I’m not at all well versed in computer hardware design; it’s entirely possible that what I’m suggesting won’t work within the laws of physics because Apple’s current chips rely on physical proximity to be able to achieve the low latencies they enjoy (after all, Grace Hopper taught me that a nanosecond is about 11.8 inches).

I still look back to the 2019 Mac Pro announcement (it was the best WWDC), and although Apple didn’t say it in so many words, that announcement felt like a recommitment to the kinds of pro users who would be buying a Mac Pro. Apple thinks for the long term, and they weren’t going to make the investment they made into designing the 2019 Mac Pro if it was just going to become a one-off product. And it similarly doesn’t make sense to go all out with the 2019 Mac Pro if the intent was to make subsequent Mac Pro computers successors in name only.

It’s taken Apple many years but I think they have finally come around to the idea that they need to let pro Macs be pro Macs, and it’s safe for us to expect a Mac Pro that not only has specs that lap every other Mac (and almost every other computer on the market), but modularity and flexibility that the most demanding and tiny subset of Mac users need.