Blowing a 10 Year Lead

Google blew a ten-year lead - Second Breakfast

In 2010, I predicted that by 2020 Chrome OS would be the most popular desktop OS in the world. It was fast, lightweight, and $0.


Then something happened at Google. I’m not sure what. But they stopped innovating on cloud software.
Docs and Sheets haven’t changed in a decade. Google Drive remains impossible to navigate. Sharing is complicated. Sheets freezes up. I can’t easily interact with a Sheets API (I’ve tried!). Docs still shows page breaks by default! WTF!

I sympathize with all this this. Google undeniably has the best search and map data, and their Photos library is excellent, but I struggle to find a place for them elsewhere in my life lately. Apple continues to thrive in hardware and core services, while Microsoft excels at productivity and cloud storage, and new services like Hey have token me away from Gmail (the first time in 16 years Gmail hasn't been my main email service).

I think the Apple nerd community is harder on Google than they need to be, but I have to admit that over the last couple years, Google hasn't done much to get me to switch to their services, I'm slowly but surely switching away to what I find to be better solutions elsewhere.

How to Share Podcast Clips from Castro

This is likely old news to most people reading this site, but I made a short video over on my A Better Computer YouTube channel to show how simple it is to share fun moments from your favorite shows to social media (or just save them for yourself!).

I don't ever say it in the video, but if you enjoy these videos, like and subscribe, and tell your friends about the channel! I'd love to be able to grow the channel into something really great, and the more people there, the more I can justify investing more time to make even better content.

Apple Silicon Mac Will Have Touch Screens

Quinn Nelson put out a video yesterday about "Big Sur's Big Secret"

I should have shared this thought earlier, but from the moment this year's WWDC keynote ended, I was convinced every Mac made with Apple Silicon is going to have a touch screen. The UI changes in this fall's macOS release indicate this, the "larger" option for all UI elements implies UI will be boosted a bit more when a touch screen is active, and Apple very cleary talked about new form factors they would be able to make with control over the processing power in their upcoming Macs.

We're all expecting the first consumer Macs using Apple Silicon will be MacBooks. I have 3 predictions, ranked from most likely to least likely:

  1. These laptops will have touch screens
  2. The product name will not be "MacBook Air/Pro" It will either be "Mac", "MacBook X", or something totally different.
  3. The screen will be detatchable, offering a 100% touch-based input mode for the first time on a Mac. Look out for data miners to find references to some sort of digital keeyboard if this one is going to come to pass.

macOS will always work with a mouse and keyboard, so Mac Mini and Mac Pro will still work that way, but if you buy yourself a touch display, you can bet it will work with macOS very nicely as well.

Oh, and while I'm on it, how much do you want to be Apple's oft-rumored return to the display market will be a touch-based Cinema Display released just in time for those new Apple Silicon desktops?

Injecting Yourself Into the Relationship

Why AnyList Won’t Be Supporting Sign In with Apple - AnyList Blog

One problem is that most Apple IDs are tied to an iCloud email address. So most accounts created via Sign in with Apple will use an iCloud email address. But many of those iCloud email addresses are unused and unchecked, because a customer’s “real” email account is their Gmail, Yahoo, or Hotmail account. If we try to contact a customer using their iCloud email address, they may never see our message. We used to run into this problem constantly with customer support, back when AnyList used the built-in iOS email compose interface for sending support requests. This interface often defaults to using an iCloud email account. So people would ask for help, we’d reply, and they’d contact us again later, angry that we never replied. Our reply was going to their iCloud email account, but they didn’t see it because they only ever looked at their Gmail account, in the Gmail app.

The article lists a bunch of issues from a consumer and developer perspective, and while some of the arguments are true of all third party sign in options, right now I want to address a common thread among numerous conversations in the Apple community recently: Apple sticking itself between the merchant and the customer.

Programming note: I'm trying to refer to companies distributing their software as “merchants” and not “developers” from now on. I think it's more accurate and it avoids an unconscious bias when we frame these money/relationship debates.

Whether it's the right thing for everyone involved is up for debate, but Apple does a lot to put themselves in between the merchant selling a good and the customer. This manifests here with Sign in with Apple where there can be issues with merchants talking to their customers, and it manifests in App Store purchases, where you don't pay the merchant for their services, you pay Apple and Apple distributes 70% of that money to the merchant. There is merit to each of these decisions, but as AnyList lays out, and Hey brought up weeks ago, there are definite downsides to merchants and consumers as well.

It's interesting when you look at this in context of the greater history of software. It used to be that if you wanted to sell software to consumers, you needed to strike a deal with CompUSA, Circuit City, or Best Buy to get them to carry your wares, because if you weren't there, then no one would buy your product. This gave these retailers too much power over what was sold, what customers were pushed to buy, and ultimately, what software was able to be built in the first place.

The internet, as it's one to do, changed everything.

No longer did you need to strike a deal with Best Buy to stock your software, you could sell it direct to consumers from your website. No longer did you have to hope that CompUSA would stock your box on an endcap so customers saw it, you could run ads online and get your own cusotmers. No longer did you need to pay an exorbitant fee to Circuit City for every copy of your product they sold, you got to keep a huge share of that. No longer did you have to hope that customers registered their software after they bought it so that you could have a freaking clue who your customers were.

The internet made it so that merchants, large and small, could make more money, have better customer relationships, and ultimately have more control over their own destiny.

The App Store has done tons of good for the world of software, and more people buy software today than ever before, and the App Store gets tons of credit for making that possible, but I think that ignoring the costs to merchant/customer relationships is unwise.

We're now moving closer to the big big retailer model we had 20 years ago. The rules are less restrictive, but merchants must still work out deals with Apple and make software the way Apple dictates it should be made. You still hope for Apple to feature your app to drive sales, and if you want to take on customer acquisition and payments yourself, there's a good chance Apple will balk and force you to run that through them.

I know the arguments for this, largely revolving around privacy, and I get all that. Hell, it's likely a good thing in the majority of cases! But what we have is a commerce system where it's harder for merchants to have direct relationships with their customers.

Yes, it's easier to trust your credit card information to Apple than a random company online, but tools from companies like NMI (shameless plug, this is where I work and design simple, secure payment acceptance methods) and Stripe make it so that merchants can take payments securely and easily in 2020. Refunds are also harder when using the App Store rather than your own payments platofrm with makes refunding customers simple.

Yes, it's good for Apple to be able to prevent merchants from spamming you with emails and unwanted communication, but the cost can also be making it harder for merchants to provide support to their customers when issues arise, or notify them of genuinely useful information about the product the customer purchased.

Ultimately, I think the App Store is a net good, and I'm not suggesting we should aboilish it or anything like that. Putting Apple between the merchant and the consumer has benefits, but it's disingenuous to ignore its costs. And while things todat are far better for merchants and developers than they were in the big box retail days, I think the "at least it's better than 20 years ago" argument is pretty weak. One can acknowledge the progress we've made all while suggesting things can still improve.

Why Apple's New Sleep Tracking Doesn't Score Your Sleep

How the Apple Watch tracks sleep -- and why - CNET

Unlike other wearables such as the Fitbit or Oura, which measure how much time you spend in the various sleep phases and even give calculated sleep quality scores, Apple's sleep tech is more simplified. It just tracks duration of sleep, movement disturbances and heart rate. The content of your sleep isn't analyzed much at all. Instead, Apple's placed a big focus on the time you go to bed and what you do while you wind down.

I think this is a good start for them, as I often look at my sleep data through tools like Autosleep or Sleep Cycle and don't know what to do with it other than go, "looks like I was a little restless at this time for no reason".

My biggest complaint with the new feature is that it doesn't seem to have a way to edit your sleep history. This would be a minor problem, but I've also noticed that it always counts you as getting out of bed at your alarm time, whether you get out of bed or not. I can snooze the alarm a few times and it still shows me getting out of bed at 6AM on the dot. This weekend I slept in all the way til almost 9AM, but the watch says I got up at 6AM. This doesn't feel like a beta bug or anything, but a decision for this release, and I'm not a fan. Submitted feedback saying as much, so fingers crossed this is added.

It's Basically Just the Flu, Right? (Comparing Flu and COVID-19 Infection Rates, Death Rates, and Why You Should Wear a Damn Mask)

This is something you’ll hear all the time from Fox News viewers and those who get their medical advice from conservative political pundits on YouTube. I’ve heard it enough that I decided to look at the numbers myself and see exactly how much water those claims hold.

TLDR: They hold no water, are you kidding me? I kind of assumed this because medical experts disagree with this, it’s only political pundits who argue this is true, but looking at the numbers helps clarify how full of BS these claims are in the first place.

Also, this whole situation is more complicated than this article will get into, but this is simply an attempt to remove all political spin and just look at the numbers to see if the flu and COVID-19 are in the same ballpark.

Data sources:

Number of deaths

As of now, 125,000 people in the US have died from COVID-19. Maybe that number is slightly over reported, maybe it’s slightly underreported, but in general we can agree that at least 100,000 people have died so far, and that number is going to keep growing.

Now how many people die of the flu in the US each year? Well, these are estimates as well, as tracking exact numbers of causes of deaths is apparently harder than you’d expect, but here are the CDC’s numbers for recent years:

  • 2019: 34,000
  • 2018: 61,000
  • 2017: 38,000
  • 2016: 23,000
  • 2015: 51,000
  • 2014: 38,000
  • 2013: 43,000
  • 2012: 12,000
  • 2011: 37,000

So if we look at the worst year in the past decade for the flu, it’s still half as deadly as COVID-19. And on average, about 37,000 deaths is less than a third as many deaths as we’ve seen from the current coronavirus.

Interesting side note: in researching the numbers around flu deaths, there are some conspiracy theorists out there who think that flu deaths are inflated to sell flu shots. These claims are baseless, of course, and are centered around the fact the CDC uses the word “estimate” in their explanation on their reporting.

How many people get it?

By our current counts, at least 2.5 million Americans have gotten COVID-19. The number is some amount higher than that since not everyone who had it got tested, but we don’t have a good estimate for that.

Instead, let’s focus on how many people are estimated to get the flu every year in the US. According to the CDC, 8-20% of Americans get the flu every year, and 3-11% of them show symptoms (aka 5-9% are asymtematic carriers. The CDC estimates 2018 (the worst year for deaths) had 45 million Americans showing symptoms. That’s 14% of the population. The CDC doesn’t report asymptematic numbers anymore, but if we assume similar trends to previous years, that puts us in the 55-65 million range for all people who got the flu.

Put those together: how many die from the infection?

So if at least 45 million people got the flu and 61,000 of them died in 2018, the worst year in the decade, that‘s .136% of people in the US who get the flu, die from the flu.

Let’s give the “it’s just the flu” people the benefit of the doubt and say:

  • COVID-19 deaths are over-reported by the worst case 25% and deaths are closer to 100,000
  • Cases are not being tested for well and there are 10x more people who have had it, bringing it to 25 million Americans

Even with those stacking-the-deck numbers, that’s 3x more deadly than the flu.

But if we don’t stack the deck (since we didn’t do that for the flu numbers) then what we have are 125,000 people dying on just 2.5 million infections. That’s 5%, or 40x more deadly than the flu.

Long story short, we’re landing on COVID-19 being 3-40x more likely to kill you than the flu.

Mitigating circumstances

The background of all this is that while COVID-19 has killed more people in 3 months than the flu has in the past 3 years, we’ve also been in a nationwide “lockdown” to try and prevent its spread. Things are opening up, but there are tons of rules in place to make sure people remotely stay apart and wear masks to prevent them from spreading the disease if they have it.

Meanwhile, we do close to nothing to prevent the flu from spreading. None of us wore masks when we went out during flu season to make sure we didn’t spread it to others. Tons of people went to work with the flu because they wanted to work through it or their employer didn’t pay them if they didn’t work, so they needed to keep working. We didn’t clean surfaces or anything like that…basically, we lived with it.

You can split hairs all you want, but the fact is that despite our country taking the strongest steps maybe ever to try and stem the spread of a disease, it’s still killing at a much higher rate than the flu.


As I did last month, it’s been quite informative to look at the numbers. Doing so again for this was equally clarifying.

Wear a damned mask and only expose yourself to others when you need to do so.

iOS 14 Could Make Changing Your Wallpaper Much Easier

iOS 14 Could Make Changing Your Wallpaper Much Easier

In case you didn't know yet, iOS 14 lets you enter jiggle mode by holding down on any empty part of your home screen. This lets you delete apps, move apps to your App Library, add widgets, and even hide entire screens. It should also let you change your wallpaper.

For those who haven't used iOS 14 yet, here's the concept next to the current version of jiggle mode in beta 1:

Pretty subtle, but I'd love to see this be how you get to wallpaper settings instead of bundling it in the Settings app.

The Rejection of Design, and a Reversion to Checkbox Comparisons

One of the things I see often from people who don’t like the look of Hey is “I can do all of this in Gmail with labels and blocking senders.”

My response would be, “sure, you could probably do a lot of this in Gmail as well, but that’s not the point.”

I think that the thing people don’t bring up in this conversation is design. The way Gmail and Hey are designed is totally different, and the apps push you into doing different things with your email. Gmail is a complicated service by default, with 4 dort of inboxes, numerous labels (you probably have 100 labels if you’ve used a bunch of third party apps in the past), and a UI that’s built to work for many millions of people because Gmail has become the default email service over the past decade and half.

Hey on the other hand is designed around a specific handful of workflows. It’s designed around a screener system that makes you manually allow each sender to access your email at all (I’ve blocked several spammy “can we write a sponsored post on your site” 🤮 emails already), which is a mental shift from Gmail’s in-by-default view. And then little things like removing the need to archive messages, options to create mini-todo lists for things you will reply to later, and more.

These design details and suggested workflows make Hey a different experience from Gmail, and while it will not appeal to everyone, I think that suggesting that “most of this can be replicated in Gmail,” is selling design short and reverts to checkbox comparisons that are not always useful.

Here’s another example: macOS and Windows. I would wager that you could get your work done on either platform, especially if you do a bunch on the web where the functionality is exactly the same. However, despite sharing almost entirely the same features, you likely have a strong opinion on which platform you prefer.

The Point of Superhuman

Speaking of email apps…

Superhuman’s email app is overhyped and overpriced - The Verge

Part of the point of Superhuman is the exclusivity. It’s the email app for the 1 percent of people who use email in a very particular way for very particular business purposes. It’s built for the founders and CEOs and executives who are featured prominently on the company’s site and who’re willing to pay the price to be part of the VIP club.

This lines up pretty well with my understanding of the service. People who use Superhuman seem to love it, and that's great, but it really seems like something where half the appeal is that you're using something special that not everyone gets to use.

In a way I respect it, actually. They have a small (by Silicon Valley terms) user base who's paying them $30/month each, and all of their customers love them. They haven't gone mainstream, they haven't gone free to grow market share, they're just making a focused product that appeals strongly to a specific market (although I don't dig their tracking features).

My guess is we'll know things are going south for them if they start opening up to way more people all at once. Exclusivity isn't a bug, it's most definitely a feature, and if they do it right, it could be their way to a long running, sustainable business.

My Detailed Overview of Hey Email

I made this last week, but didn't get around to posting it here until now. Oops!

Quick update since I made this video: I have paid for my first year of the app. I was a little weary in this video, but I've enjoyed it more and more each day, so it only seemed right. You can send me anything you want at!