My event-driven life

I apologize for my extended absence! I once again find myself with lots of drafts that I’d love to publish, but haven’t found the words to complete any of them to my satisfaction. I thought I’d take the time to publish something that is maybe a bit rambly and less focused than usual. I hope it’s not too rambly or unfocused.

I spend a lot of time building event-driven systems. JavaScript, as a tool, is based almost entirely around events. Any UI—by virtue of the nature of the DOM—is event-based. Communication between processes or threads in many languages and runtimes is modeled in an event-driven way. In a past job, one of our interview questions was to build an EventEmitter class. Working with AWS Lambda, my functions are triggered by EventBridge or events passed to a queue. Events and event-driven systems are all around us.

About a month ago, I started working with a new fitness coach. He’s had me getting ten thousand steps in each day. Ten thousand steps actually isn’t that much, when you think about it. But my phone was reporting less than half of that. After watching the little ring in the app tick up while walking on the treadmill, I became convinced that my phone was undercounting my progress by a significant margin.

I resolved to get a smart watch, which I understand to be much better at step counting. When I set it up, my step counts did indeed skyrocket, and I’ve been able to meet my goal almost every day. But the real utility of the watch hasn’t been in my fitness progress, it’s been in my notifications.

I’m the sort of person that mutes most notifications. For many years, I even kept my phone in “do not disturb” mode 24/7. Today, I am ruthless in silencing useless or low-signal notifications so only the most important bubble up: emails, comments, and other asynchronous communication don’t make the cut. Text messages, missed phone calls, and things of interest do.

Notifications are, in a sense, simply the interface for an event-driven system. Every service and app that I use pipes their notion of an “event” into a single set of APIs to show me notifications. I use rules to decide which of those get attention and priority. The sheer scale of notifications has made it a necessary part of life for almost everyone with a smartphone, and we form a sort of symbiotic relationship: we need our notifications, but the people sending the notifications also need us¹. Neither of us can easily live without the other.

Notifications can work well and benefit us: important notifications from loved ones let us feel close and connected at any moment. Just now, my mom asked for advice about air purifiers because her cat is shedding too much. Notifications can also cause us harm: we’ve all gotten onto an airplane and seen the person next to us with a notification drawer completely jammed with so much crap that you feel your blood pressure rise. Anyone whose parent says they’re “good with computers” has come home for the holidays to find that *someone* has clicked “Allow” every time a website has asked to send notifications.

Our lives and habits are shaped by notifications. At their core, notifications are the way digital systems make realtime information available to us. Which notifications we receive affect what we think about and how we structure our days.

Another recent purchase in our household has been a litter robot. It’s been a true luxury to have a robot clean up the cats’ poop for us, keeping the house smelling nicer and eliminating a regular scoop-job.

The new robot has an app that—by default—alerts you to cleaning cycles, which diligently occur a configurable seven minutes after a cat exits the litter box. Overnight, my awareness of when our cats poop went from a theoretical understanding of their bowel habits to an exhaustive one, complete with frequency and measurements of weight. Though I silenced the notifications (I really don’t care), it is fascinating and it’s given me a false sense of urgency to actually do something with this knowledge.²

It’s no surprise that companies will do damn near anything to find themselves in your notifications. eCommerce sites ask for your phone number for some trivial sweepstakes or menial discount so that they can find themselves in the most intimate human-machine interface that you possess. Hell, even this blog asks you for your email (not that I’m trying to get in your notifications—feel free not to subscribe).

I few weeks ago, I signed up for some premium frozen cat food to be delivered. My little orange boy, JT, had seemed bored with his food, and I thought that it could be interesting to mix it up with something new, healthy, and wholesome. The site I ordered from was quite good, asking some information about both of the cats and their preferences. They sent detailed tracking information to my inbox, and a text when the food had been delivered, reminding me to put it in the freezer before it thawed.

A few days later, I got an early 9am call from an unknown number. On the other end was a real human being asking me how my cats were adjusting to the new food. I don’t know, my dude, they ate it. I guess they liked it? JT didn’t hate it, though that isn’t high praise. The call ended pretty quickly.

I give them a lot of credit for caring, I guess. The goal, though, is almost certainly not to find out how my cats are doing, but to stay top-of-mind and ensure I renew my subscription. The currency that this company trades in is seconds of my life triaging things in my notification drawer related to their business.

I have been thinking about how we onboard new things to notifications. “Cheap” notifications are the sort you get from websites. Some are very useful, like Google Calendar notifications. Others, like the ones you might be prompted for on a recipe website, are disastrously bad. The meat and potatoes of notifications are the sort that tell you things you want to know about, like weather alerts, doorbell notifications alerting you to deliveries, or low-battery signals from a smart device.

One kind of notification I’ve considered onboarding with is not useful, but it is conceptually enjoyable to me. Instagram has been trying to sell me a camera so that I can watch the local wildlife eat in my backyard. They show videos of someone loading this little device up with bird seed, and the owner’s phone lights up with up-close clips of birds.

I actually don’t care about birds, but I do feed my local squirrels (perhaps to the disdain of my neighbors, but I don’t care). We buy almonds from Costco to throw out on our deck. The cats love to watch the squirrels scurry around and stuff their pudgy little cheeks full of almonds.

Max thinks that one of the squirrels is a bully and eats more than his share. He doesn’t want me to feed the squirrels because he wants to spite that one squirrel in particular. I just want to see the cats enjoy watching the fat fuckers eat. This has had consequences, though³. Getting squirrel drama pumped directly into my veins by way of a notification is highly desirable.

Advertising (and more broadly, marketing) is about getting you to think about a product or service. I think it’s probably pretty rare that companies think they’re going to show ads to people who are already in the market for their product: nobody thinks “yeah, I’m in the market for some cola, I should go get some Pepsi!”

Ad targeting lets companies narrow down the demographics that see their ads so that folks who are unlikely to purchase their product don’t see the ad. I’d expect folks under 21 probably see far fewer car insurance commercials than folks 22 and older: they’d probably just use whatever’s cheapest or whatever their parents use.

Having more targeted ads reduces the amount of money companies need to spend to market their products and services. But adding targeting to the top of the funnel also means that the attention you purchase down the funnel becomes more valuable: the expensive cat food brands I click on (because Instagram knows I have a cat) can collect my phone number in exchange for a small discount, and in turn they can directly advertise to me without needing a third party. They’ve plugged into the lowest-level interface for my attention that they possibly can.

Content farms, which produce no real product of substance, leverage this all the time: crappy web notifications aren’t free to deliver, but they cost very little. As long as the cost to send them multiplied by the conversion rate is lower than the ad revenue, they can pump out bad listicles and AI-generated nonsense and get the poor souls who clicked “Allow” in the Chrome permission modal to pay their bills.

A sidebar about VR

I’ve had a Meta Quest 2 for a little while now. Before that, I had an Oculus Go. I really quite liked the Go. The Quest 2, frankly, hasn’t been an improvement. I don’t care about the pixel density or lens quality or anything like that: I want to be able to lay down in my recliner and watch videos. The Go was great about that (and Firefox Reality was a real win). The Quest 2 does everything in its power to discourage you from being in any non-upright position.

The built-in browser is seemingly the only browser choice now. And the selection of apps somehow seems to have shrunk considerably in the past couple years. Holding the main button re-centers the view, but it doesn’t do really anything to adjust the pitch of the UI, so the home screen can only be practically used while you are upright. A “lying down mode” has allegedly been in the works for some time, and it’s frankly ridiculous that it doesn’t exist. I don’t want to get up and move. I will never play Fruit Ninja or Beat Saber. I want to watch ASMR videos while reclined on a screen as big as my house.

That said, I can see the potential of VR. The Apple Vision demo was exciting to watch. Ignoring the time between now and when the price becomes reasonable for the average consumer, I imagine all-VR Facetime chats become the future replacement of iMessage groups. You can imagine how this is more than just chat: I’ve played the Facebook Messenger basketball game⁴, and I remember a time when Facebook was about games (who remembers Mafia Wars?). It’s not hard for me to imagine a world where people casually share spaces to do stuff together—games or otherwise.

And that shouldn’t be surprising: the “metaverses” that we already have (Minecraft, Roblox, Second Life, etc.) have been doing this for decades, albeit with a keyboard and mouse. Even Animal Crossing⁵ is a great example of a place where you can do things with friends. VR, as I see it, replaces the general purpose computer (or, game console) with a platform that standardizes a place, and the apps/games implement the rules of the world. To use a bit of a contrived metaphor, I’d love VR devices to become the platform for VR apps in the way that Garry’s Mod is a platform for worlds/minigames/user-defined ways to be social.

I don’t think Apple Vision is likely to become ubiquitous like AirPods or the Apple Watch. I think the high quality bar of the product is going to carry a high price tag basically forever, and it’ll be out of reach for the market that Facebook and others are after. But I am glad that Apple entered the space because it saves VR, in a way, from Meta’s fever dream of a social product (Horizon Worlds) and the crypto bros’ capitalist endgame vision for a metaverse where plots of land on the infinite plane of virtual reality are owned with blockchain-purchased deeds.

I have definitely had a fair amount of experience with VR at this point and have owned two VR devices, but I’d hardly consider myself an expert. In theory, I should be an expert! It’s an incredibly cool technology with really hard engineering problems. It’s fun when it works. Hell, the Go blew me away and it had specs (on paper) that hardly rivaled a Dell Dimension from 2009, which is a feat. We’ve blown past this technical achievement that’s absolutely incredible and as of yet—at least in my opinion—haven’t found any compelling application for it besides somewhat more immersive video games.⁶

…which brings me back around to my point

I got my watch and I’ve been tracking my steps, and I find myself looking at it dozens if not hundreds of times a day. I go out of my way to find convenient times to charge it so I don’t miss out on steps or sleep tracking. Meanwhile, the device that makes me feel like I’m in another world is so shitty to use that I only begrudgingly use it. I’ll save my ranting for another post.

And it really comes down to utility. Looking at my watch and seeing notifications is second nature. There’s zero friction—mental or physical—to using it. Having my watch on is like having a sixth sense.

With my Quest 2, the battery is seemingly always dead, it takes a few minutes to get started⁷, and it limits me to uncomfortable positions. Moreover, I don’t feel more connected while wearing it, I feel the exact opposite. Maybe it’s because I’m not a Facebook user, but it certainly doesn’t help that my entire digital world (texts, email, calendar, and other notifications) simply don’t enter the headset except in the most basic way⁸. The mental and physical barriers to me using VR are enormous.

While disappointing, I think it makes a few things apparent:

• The notification interface is mostly platform-independent. If I threw out my Pixel and bought into iOS tomorrow, I’d still get pretty much all of the same notifications.

  • This is thanks to protocols like IMAP and WebCal, as well as cross-platform app availability and web technologies.

• Whether you think notifications as an interface are good or bad, they’re here and they’re what evolved as the lowest-common-denominator for multi-channel realtime messages.

  • Maybe this isn’t a global maximum, but it’s not the worst we could do, all things considered.

• For me (and I imagine a huge percentage of folks), the lack of real, interactive notifications is a dealbreaker for adopting a new mode of computing, like VR.

What worries me is the progression of this arms race for attention. Look at recent events: Twitter and Reddit are demanding first-party, signed-in access to your notifications to use their services. Messaging spam (SMS, instagram, Twitter DM, Telegram…you name it) is worse than it’s ever been. Telemarketers (and at least in my case—recruiters and software vendors) are as brazen about calling your personal cell phone as ever. Efforts to quash this nefariousness by governments and platforms will offer some temporary respite, but there is a fixed, finite amount of attention available for purchase. This will only get worse and there will be new ways to exploit your attention.

Could I unplug? Sure. Folks online would argue that plenty of people aren’t allowed to use their phones all day while they work, but that doesn’t make them less reliant on notifications (in the same way that sleeping 8 hours each night doesn’t make me less reliant on notifications). I don’t think being connected is a bubble or a fad, I think it’s a consequence of over half our species being connected to the internet⁹. Notifications are here to stay regardless of whether or not we like them or want them.

The future of VR has me even more worried. I’m not unconvinced that I’ll spend my days working in a VR headset in five years’ time. When VR becomes less disconnected and disjoint than Meta’s incarnation of it feels today, notifications will be more valuable than ever. The interface between me and the rest of the world won’t just be limited by whether I choose to glance at my wrist, it’ll be changing something in my field of view (regardless of what I’m doing). The notifications literally sit directly in front of your eyeballs.

Notifications will need to evolve (just like email had to evolve as spam became ubiquitous). Consumers like you and I will need to be more actively involved in choosing what to let in and what to keep out. And companies will do more and more to fight for their ability to get a slice of the notification pie. I can imagine a dystopia where you aren’t allowed to make an ecommerce purchase without opting-in to some kind of high-priority notifications. I hate the thought of that world. Unfortunately, I don’t see a way that it doesn’t become inevitable.

Thankfully, that world isn’t the one we live in today. Until then, companies will still need to persuade me that they’re more important than my cats’ shits.

1. Either because it’s other people relying on notifications to deliver their message, or because it’s a business that relies on attention for revenue.

2. I have yet to decide what use my cats’ bathroom habit data could yield.

3. Having a bunch of squirrels scurrying around on our deck does seem to have attracted some hawks, which have not been kind to the squirrels. The squirrels also seem to be very territorial and have formed factions, pissing all over the place.

4. I’m no longer on Facebook or Messenger and have no idea whether it’s been removed.

5. I’ve been drafting a post about Animal Crossing for literally months now. Hopefully it’s ready to read soon :)

6. I know that nausea has been a problem for some folks, but I truthfully don’t think this is what’s holding VR back. Modern headsets seem to have done a fairly good job of mitigating many of the physical problems of the headset. VR technology itself is not the problem!

7. Oh, Guardian wants to check whether my safety zone is still good? Fuck off, Zuck, I’m trying to watch videos in my recliner, not play Wii Bowling.

8. You can pair your phone with the headset to get your phone’s notifications on the device, but they’re essentially non-interactive and require you to take the headset off to actually do anything with them. You can’t, as best as I can tell, read an email if you get a .

9. We simply cannot ignore that a majority of all people now use this technology. To call connectivity itself a fad or something to be mitigated ignores the sheer scale of what we’ve created as a species. Even if many of those people aren’t “online” in the way we mean in the west, the sort of phones that many folks are using as their first connection to the internet still have notifications.