markdyerhere

Human/Cyborg Relations

[markdyerhere]

The Agent Paradox

Why Enterprise Agent Adoption Depends on Human Rhythms, Not Just Model Capabilities

📄 PDF

The Problem

Agentic tools, like Claude Code, became an integral part of my workflow as a Model Behaviour specialist at the Browser Company. I used it daily for everything from building evals to analyzing data. However, the more I used agentic tools the more I noticed a core tension in how these tools are currently designed and how they are fundamentally mismatched for human use.

I often found myself trapped between two equally bad states: watching the agent in paralytic captivity as I waited to approve the next step, or context switching to human-suited work only to return and find the agent blocked, waiting for me.

These tools promise to augment human abilities, but instead humans end up serving the agent through continuous monitoring—what Bainbridge (1983) called the 'ironies of automation'.

This violates well established research on human attention and monitoring of autonomous systems (Parasuraman & Manzey, 2010). We know that humans are actually quite bad at monitoring autonomous systems due to natural cognitive limitations. From studies on air traffic controllers to safety drivers in autonomous cars, humans have proven that they are not capable of monitoring systems that appear to randomly alternate between states of autonomy to needing immediate human intervention.

This is especially important for enterprise agentic systems which are typically not created or managed by their end users—knowledge workers with varying familiarity with AI and agentic tools. These knowledge workers may become frustrated by the start/stop workflow that can occur when trying to work with an agent in parallel. Worse, if they become trapped in a state of agent authorization captivity, they may reflexively approve a dangerous action that violates the company’s AI safety policy.

However, from my vantage point the more dire risk is knowledge workers quietly churning off of a secure enterprise agentic platform to a consumer one that provides a better, more human centric experience. If the enterprise agent UX is hostile to human attention patterns, then all the work on security and privacy is rendered moot the second the end users run sensitive queries on a consumer AI tool. ChatGPT is always just a few taps away on their personal phones.

Model capabilities alone will not solve agentic AI adoption. You can have state of the art performance on standard benchmarks and perform technically well on agentic tasks. However, if the workflow burns humans out they will not use it. Current approaches have engineers manually inserting checkpoints based on risk or compliance. What’s missing is the agents learning when the human attention state warrants interruption. The question for the agent needs to be framed not as “do I need permission to do this task”, but as “is this a natural collaboration checkpoint that respects the human’s attentional state?”. For a more human centred agentic workflow, the human must be understood as a collaborator not just part of the loop.

The Attention Problem is a Design Problem

Our attention spans while using digital devices have collapsed significantly in the last 20 years. Dr Gloria Mark’s work in this field has documented that knowledge workers switched applications every 2.5 minutes in 2004. This has been reduced to 47 seconds as of 2023 (Mark, 2023).

This kind of rapid task switching is known as “kinetic attention” and it’s framed as an adaptation to digital tools that allow for rapid context switching between apps, windows and tabs. While it is easier than ever to access information across multiple tools, this kind of rapid context switching has major costs to human attention. Mark’s work has shown that on average it takes 25 minutes for humans to return to a state of deep focus after a context switch.

Furthermore, Mark details how human attention follows internal circadian rhythms. Unlike AI agents, humans do not have the same capacity for attention throughout the day, but rather the day is broken in to periods of attention peaks and valleys. Typically humans will have two peaks of attention where they are capable of deep work—one in the late morning and one in the early evening. She notes that the attention valleys are equally as important as they act as a rest time for the human where they can accomplish more rote tasks and build up attentional capacity for the next peak. These naturally occurring rhythms of human attention are fundamentally mismatched to the current workflow that requires humans to be constantly monitoring the agentic tool. The risk here is that unless enterprise agentic tools are better aligned with the attentional rhythms of humans they are likely to exacerbate the ‘kinetic attention’ problem of using digital tools, leaving the human even more scattered and drained.

This compounds the vigilance decrement problem documented in autonomous systems research (Parasuraman & Manzey, 2010): humans monitoring automation not only lose situation awareness over time, but in agentic workflows, they're doing so while already operating at fragmented 47-second attention spans.

Recent research has begun acknowledging temporal misalignment between humans and AI. A 2024 CHI paper on proactive AI assistants explicitly studied 'interruption cost' and found that timing of AI interventions significantly impacts collaboration quality (Lee et al., 2024). Similarly, researchers at ACM 2024 called for 'rhythm and attunement' between humans and LLMs, noting that current systems ignore fundamental temporal differences (Buschek et al., 2024). However, this work remains nascent and fragmented. Human-in-the-loop (HITL) frameworks like LangGraph, which govern when agents pause for human approval, focus on technical ‘pauseability’ for compliance, not cognitive optimality for humans. The connection to decades of automation monitoring research—vigilance decrement, out-of-the-loop performance, situation awareness—hasn't been made in the AI industry (Parasuraman & Manzey, 2010; Endsley & Kiris, 1995). And critically, no one is proposing this as a modelling problem—something that could be learned through post-training rather than just engineered into workflows.

Enterprise agentic platforms are typically sold to CTOs and IT heads, but adoption depends on whether knowledge workers actually choose to use them. Security, data sovereignty, and compliance value propositions are meaningless if end users route sensitive queries to consumer tools on their personal phones. Human-centric agent design isn't UX polish. It's a prerequisite for the enterprise value proposition to hold.

Building for Human Time

I’ve been working at the intersection of model behaviour and real user pain on a production AI product for the last 18 months. I’ve debugged where prompt engineering hits a ceiling against the current capabilities of models. I've built evals that exposed systemic behavioural issues. I've watched users abandon well-designed features because the workflow burned them out. These problems aren’t just technological; they’re also human.

The path forward requires treating human attention as a first-class design constraint, not an afterthought. That means agents that learn when to interrupt and when to wait, that work with circadian rhythms rather than against them, and that treat the human as a collaborator rather than an approver. The checkpoints, the pacing, the rhythm of handoffs: these should be learned from data about human attention states, not manually engineered for compliance.

We're building intelligence that operates in human time. It should feel like it.

[markdyerhere]

A Product People Love

“I’m not being corny this shit is real” — A Plea, Flea

I joined The Browser Company because I loved Arc.

I started using Arc as my work browser in 2022 and it clicked with me instantly. I’m a bit of a tinkerer and was no stranger to using alternate browsers, but Arc hit differently. It felt like the folks at BCNY had researched all the major Chrome user pain points and just fixed them — I even mentioned this to Browser Company CEO Josh Miller in my interview!

I loved the way Arc made me feel faster, more organized, more in control. Unlike Chrome, I didn’t have to wrestle with settings and extensions to make this happen. Arc was already tuned up and ready for me to hop on and start flying. The team at BCNY sweat all the details so when I was ready to get stuff done, the software just got out of the way. I loved that. I loved Arc.

As I worked on Arc for macOS, Arc for Windows, Arc Search for iOS and Android, and eventually Dia I kept on thinking about why people love products. My tenure at BCNY also coincided with a new love in my life. Just before Dia went into public beta, I became a father.

Now, the concept of ‘loving’ software gets thrown into stark relief when held up against loving my son. How could I profess to loving a collection of bits when I had this beautiful new baby in my life filling my heart in ways I’d never imagined possible?

However, as the hours turned into days, into weeks, into months, and now almost a year I began to understand what it means to love on a deeper level.

Please forgive the cliché—I know I’m not the first parent to realize this—but through the sleepless nights, projectile vomit, and nap traps, ‘love’ started to reveal itself as a form of care and understanding.

It’s more than just the kind of care you give to a child; the kind of care where you give not expecting anything in return. It’s care based in the understanding of why the care is needed in the first place.

I rock my son to sleep because he’s growing and he needs sleep to develop. Because my wife has been with him all day and needs a break. Because he doesn’t know how to get to sleep on his own yet. Because my arms and chest are warm and snuggly. Because I know him and what he needs. Because I love him.

“That’s lovely, Mark…but what does this have to do with B2BSASS building software products?”, you may be asking. Well, if you want people to love your product, everything.

When I started outlining this essay I made a list of all the products in my life that I love. Some of them are software (Todoist, Raycast, Ableton), some of them are hardware (Teenage Engineering EP–133, Korg Monologue, Mac mini), but the one that I kept coming back to was my KHS Urban Soul single speed bicycle.

I love this bike. You can just throw a leg over and ride. No gears, no battery to charge, nothing complicated. It’s light enough that you don’t really miss having gears and my legs are strong enough for tackling the odd hill as I zip around town. I’d be heartbroken if it was stolen, but it’s inexpensive enough to replace. It’s also pretty nice to look at, in my opinion!

This bike is my platonic ideal of what a bike should be. Simple, practical, and uncomplicated. I bought it because it suited my needs. I liked it because of that. However, I grew to love it not because of how practical it is, but how it made me feel.

I logged hundreds of kilometres on this bike the summer I got it. Not because I had a lot of quick errands to run, but because I would get on and end up riding 20kms just for the fun of riding!

This bike is so light it basically feels like you’re flying when you get moving. I know there are actual racing bikes that are lighter, but without being encumbered by gears (or any other cyclist accoutrements save a helmet) the experience of riding felt lighter.

Once all that extra stuff got out of my way I could just ride. Ride to anywhere, to nowhere, just because it felt good to ride. I rode this bike so much that summer that I hurt my wrists from all the stand pedalling I was doing! I was in love and love makes you do crazy things.

When you build something that’s meant to be used by someone—a web browser, a bicycle—you’re not just building a list of specs to meet their needs. That’s part of it, of course, but it’s so much more. By building something that addresses someone’s needs you necessarily are showing them that you know them, that you understand them. You do it almost as an act of faith that they will see what you have done for them.

It’s like a message in a bottle. You put it out there and hope someone will read what you’ve written. The goal isn’t to hear back from them, although it is nice when that happens. The goal is simply to share what you’ve made and hope that they appreciate the work you’ve done for them.

The Browser Company has a value, “make them feel something”. It’s a philosophical product bias towards emotion. They want to illicit a feeling in the user, much like how I experienced Arc. However, I personally have reframed this value as “let them feel something”.

You can make people feel lots of things, but I don’t think love is one of them. If love is an act of care and understanding then your product must first express these values to the user unselfishly. The product, through how it is designed, must let the user know that it was made for them. That it cares about them. That it understands them. That it loves them and wants nothing in return.

There’s a reason that Jony Ive, the designer of some of the most beloved products in history, named his company “LoveFrom,” when he left Apple. It actually was inspired by something Steve Jobs said:

“[Steve] said that one of the fundamental motivations was that when you make something with love and with care, even though you probably will never meet… the people that you're making it for, and you'll never shake their hand, by making something with care, you are expressing your gratitude to humanity, to the species.”

As we move to a new era of AI product design, building with love has never been more important. AI has enabled us to build products that can truly understand users in the language that they themselves use.

To do this well will not only require high levels of user trust, but a deeper, more contextual, understanding of user problems and the care to address them thoughtfully.

We should not be building to impress or for the sake of building with AI, at least not if we want users to love our products. Products built this way may be flashy and technically impressive, but without care and understanding they will not connect with users.

We need to have the strength and vulnerability to build unconditionally in the service of others—to love so that we may be loved in return.

Written with love by yours truly! All em dashes my own — 4real!

[markdyerhere]

Why Your Claude Outputs are Bad

Your brain is (still) the ultimate tool

A lot of people complain about their Claude outputs. How they appear to be cogent, well formatted, semantically correct but once you poke at them you find a ton of flaws. When this happens people reflexively blame Claude:

“Oh this thing isn’t as smart as I thought!”

“So weird! A human would have never made that mistake ...”

“Man, this thing is actually a bad writer — look at those em dashes!”

But as the old saying goes, only a poor craftsman blames his tools.

Sorry to burst your bubble, but if your Claude outputs are bad it means that your inputs are bad.

Garbage in, garbage out. It’s as simple as that.

I don’t mean for this to come off as a thinqboi “I have the secrets to AI” style piece. Honestly, far from it!

There are actually very few secrets about AI and Anthropic in particular has been very open about how they’ve designed Claude and the principles behind that design. They have many dev docs, research papers and blog posts out there. Heck, they made one of their most important documents, the Claude Constitution, available as a pdf and epub!

In fact, The Claude Constitution is perhaps the most illuminating document for this particular problem of poor quality outputs from Claude.

Essentially the model has several principles that will prevent it from creating quality outputs for a user who knows very little about the subject or task they want Claude to create an output for. This is a tool that is meant to augment the user, not automate them away.

The Claude Constitution clearly lays out that they want the model to encourage the user to not become overly reliant on AI and instead to encourage the development of their own thought and craft.

"Autonomy-preserving: Claude tries to protect the epistemic autonomy and rational agency of the user. This includes offering balanced perspectives where relevant, being wary of actively promoting its own views, fostering independent thinking over reliance on Claude, and respecting the user's right to reach their own conclusions through their own reasoning process." — Claude Constitution, page 33

So, if you, a know-nothing dingus, come to Claude asking it to do some task that you have no business doing yourself, let alone telling a powerful AI model to do for you, it’ll respond with an output that pulls from the median knowledge on the topic that’s roughly equivalent to the first page results on Google. It meets you where you’re at and if you’re a know-nothing dingus the first page of Google is where you’re at.

Now, if you actually do know what you’re talking about the results are very different. I saw this a lot at the Browser Company when we would do AI hackathons. The designers were able to get Claude Code to make beautiful and considered features. The engineers made features that were super useful and had complicated back end workflows. Both were able to make fully useable prototypes in the dev build of Dia, the Browser Company’s AI Browser.

Since I don’t have a traditional software background (I come from the front end web world) my prototypes were mostly mockups done in HTML. They got the point across, but I needed way more context to actually prompt Claude Code to actually make them usable features in the dev build. In this case I knew slightly more than nothing, but I was still a bit of a dingus!

I had this experience recently when adding some very light custom styling to my blog to remove the annoying iframe banner to encourage folks to login to Tumblr (tumblr 4life btw — I am an elder millennial, after all!). On the first pass I was lazy, assumed the problem was trivial and asked Claude Code to write the custom styles for me. Oh man that did not work out! Claude way over-engineered the solution and introduced some very junior level and frankly embarrassing responsive issues to my theme.

For my second pass I actually got the class that styled the iframe, asked it to look up other solutions from the web (since this definitely is a solved problem) and most importantly told it to prioritize a CSS based solution and to only use scripts if absolutely necessary. This produced a much better and more elegant solution and it’s the custom styling that’s currently running on my blog.

I honestly should have known better. I learned front end web dev from my wife, who is a legit unicorn developer + designer. Her code is beautiful, simple and elegant. Unlike a lot of front end devs, she builds like she has nothing to prove. If HTML and CSS are all that’s needed (and let’s be honest, in most cases they are all that’s needed) then that’s what she uses. Her code is cleaner, runs better, is more accessible and has less dependancies that whatever Joe Schmoe built to prove that he’s a l33t h4x0r or some huckster cobbled together from old StackOverflow posts.

When I remembered these principles that she'd taught me everything clicked. I was able to think more carefully, create the proper context with Claude and get the output that I wanted. I had to show the model that I was not just some dingus, that I actually knew something.

I truly believe this is a feature, not a bug. This is how Claude was designed and it encourages behaviour that Anthropic wants to see in the world. And much like my wife’s elegant HTML and CSS, there are so many benefits of this design—chief among them being safety. Anthropic is not about to let a bunch of dinguses and hucksters who have no business doing certain things get Claude to do those things for them! Of course, some get through (looking at you OpenClaw!) but at least, that’s what they’re striving for.

This is why the dinguses and hucksters are doomed to fail, at least when they are using Claude. In order to use Claude well, and perhaps AI in general, you need to know what you’re talking about. You need to sharpen your ultimate tool.

You need to develop your brain.

100% human written joint! All em dashes, my own. Props to Claude for finding links to things I forgot to bookmark 😅

[markdyerhere]

On Tech Support

On May 5, 2023 I got a call that changed my life.

I had just accepted a job offer to work for the Browser Company of New York!

I walked up the stairs to our apartment with great news and my wife told me to sit down. My Mum had called. My Dad had to be rushed to the hospital. He was bleeding internally and it looked like he was suffering renal failure. We flew out to be with him and my Mum. He passed away 9 days later.

My Dad is one of the reasons I’m good with computers.

Dad went to college to learn how to use computers back in the 70s — this was back when computers would take up whole rooms and you had to input commands with punch cards.

He brought home our first home computer in the late 90s; a chunky Compaq Presario that clicked like a bottle nose dolphin as it chugged along. The Compaq was mainly for his side hustle (a carpet cleaning business) and for me and my brothers to do our school work (the side hustle was helping to pay our school fees), but it quickly became my direct portal to the skateboarding article on Encarta and pokemon.com.

Dad’s main job was the night production manager for the Royal Gazette — a job he had for close to 40 years. He would often talk about how he missed the sound of a newsroom. The clacks and thunks of typewriters, the ringing of phones, the chaos of people coming together to make a paper every day. I couldn’t imagine how the Gazette could ever have been a loud place when I would visit him at work. It was deafeningly quiet! He would plop me in front of a computer to play Minesweeper and the only sounds you could hear were my mouse clicks.

Since Dad worked at night, he had a good chunk of hang time waiting for the paper to be ready to go to production. He would listen to internet radio stations, read interesting articles and forward dirty joke emails from his coworkers. When I left home for university my procrastinating study hours and his work hours would sync up. We would email each other links to cool music, videos, articles and, since I was old enough, the odd dirty joke email forward. But we mostly just chatted.

We started saying “I love you” over email

He was a great writer — he wrote just like he spoke — and both of our conversation styles lent themselves to the time delayed back and forth of email. I say something, you read it and think about it, you send a considered response back with a little something extra. Repeat.

We started saying “I love you” over email. We were always close, but we both had a hard time talking about our feelings especially when I was a teenager. Being apart, but still staying in touch over email gave us the space to feel comfortable saying we loved each other. We had no trouble saying it after that — it was one of the first things we said to each other when I could finally come home for Christmas in December 2022.

Tech is very expensive where I come from (about 30% more expensive than the US), so whenever I go home I roll up like Tech Santa with a bag full of laptops, tablets and phones. The Compaq Presario was long gone and Dad’s laptop had seen better days. Dad was a long time Windows user — he wrote his memoirs in Microsoft Word — so he was keen to stay on something familiar.

I got him a Surface Laptop 3. I walked him through setting up Windows 11. I arranged his bookmarks bar in Chrome so he could quickly access his email, Facebook, Youtube and, most importantly, his memoirs which we’d uploaded to Google Drive. He was so happy to have all of his stuff moved over to his new computer and relieved that his memoirs were properly backed up.

Going through his browser after he died was some of the most heartbreaking work I’ve ever done

After he passed away I helped my Mum with the gut wrenching task of cleaning up Dad’s digital life. I made sure she had access to his email and important logins so she could deal with closing accounts and adjusting billing names. I also memorialized his Facebook page to ensure it didn’t get hacked and start spamming friends and relatives with crypto scams.

Dad had several strokes leading up the internal bleeding and renal failure in May 2023. Honestly back in Christmas 2022 he had only recently recovered from his last stroke and he wasn’t quite himself. Going through his browser after he died was some of the most heartbreaking work I’ve ever done. The bookmarks bar that we had carefully set up together was riddled with duplicates — Facebook, Youtube, Google Drive over and over spilling out into a drop down menu. My Dad knew his way around computers. He had been using them since they took up a whole room. I knew some damage had been done after the strokes, although he was very good at hiding it and I don’t think I was ready to admit it at the time. But with him gone and looking into his computer, his window to the world, I couldn’t deny what I had known for at least a year. My Dad had strokes, my Dad had internal bleeding, my Dad was dying, my Dad was gone.

These things — computers, web browsers — they’re not just inert tools waiting to be picked up and used. The more you use them, the more they become a part of you and you of them. The laptop lid gets scratched, coffee is spilled onto the keyboard, cords become frayed, bookmarks get bundled up into folders to be forgotten, passwords get scribbled on sticky notes and then reset again, tabs sit open for months just in case, URLS and search queries are typed in repeatedly just because you’ve built up the muscle memory — facebook, youtube, google drive, pokemon.com.

We’re all about to be new at this again

A developer friend of mine once said to me, “the future is here and we’re all bad at it”. University level STEM students don’t know how file paths work, they don’t teach kids to type anymore and more and more jobs across a variety of fields require some degree of computer literacy. Everyone is expected to just know how to use computers, but as my Mum would say “I’ve never had a lesson!”. As we head into the next phase of computing — where you work with the computer instead of simply using the computer — much of your interactions will be through AI language models, but even the people building those don’t really know how they work. We’re all about to be new at this again — new at computers, new at web browsers, new at everything. Now more than ever, we’re all going to need some help. Hopefully that help comes from someone who understands what you’re going through, what you’re trying to get done and can meet you where you’re at.

What we’re building is so much more than just a portal to websites

On June 5 2023 I started a job that changed my life. I started working at the Browser Company of New York!

I help to build Arc, a product that I love, with some of the most talented, thoughtful and inspiring people I’ve ever met. I helped bring Arc to Windows, so people like my Dad could use this very special browser that feels like my home on the internet.

I truly feel like I’m doing the best work of my life and that’s because what we’re building is so much more than just a portal to websites. It’s where you do your work. It’s where you hang out with friends. It’s where you pirate anime, do your taxes and send emails to your Dad.

It’s where you live your life. As we continue to build I hope to be that human voice on the other side of the computer that helps you feel like Arc is for you, for your stuff, for your life. I want to help Arc feel like home. See you on the internet ✌️ Love you, Dad!

[markdyerhere]

Arc Boosts are rad!

So, the Browser Company just released a big update to their Arc browser that more fully fleshes out the Boosts feature and it's really cool!

The new Arc Boosts 2.0 provides a simple no-code interface to tweak a webpage's CSS so you can do fun things like update fonts and colours or hide ads or other unwanted visual elements of a page. Arc will save these Boosts for specific websites and you can share them too.

They're kind of a cross between a browser extension and the occasional manual fiddling I would do in the inspector to remove or change an annoying element on a webpage. The fact that you can save and share them is 🤌

Here's a quick one I made for nytimes.com to remove the ad images and switch to dark mode.

I also really like this simple Gmail Boost I found in the Arc Boost Gallery. You can check out the release notes here, FAQ here and The Verge's David Pierce did a nice write up on the new feature here if you want to learn a bit more 😊

[markdyerhere]

The Best Burner Phone is an Apple Watch

Yeah I said it. Fight me! (Don't actually fight me😅)

I have eyed the Light Phone, the Punkt phone, the Nokia 3310 3G. I've thought about rocking an iPad Mini with cellular data. I've flirted with the idea of a goddamn pager. Anything to get me away from the ever present anxiety rectangle of doom that is the modern smartphone. I didn't get any of these (although I would love to play Snake and listen to MP3s on the Nokia 3310 — just look at how sick this thing is!).

I got an Apple Watch. And y'all, it's the best.

Okay, so the Apple Watch isn't actually a burner phone. It's more of a chill vibes, low stress, limited functionality device. It's basically a powered up wrist iPod that can make phone calls.

But honestly that's as good as you're going to get as cell providers rapidly work to decommission their 3G networks ,ruining my Nokia feature phone fantasies. With the Apple Watch I can send messages, listen to music, play podcasts, check off reminders, view calendar events, track a workout, pay for stuff, turn on my lights...oh and make phone calls. It lets me do just enough and get back to whatever I was doing — helping me avoid getting locked in a phone hole of pseudo productivity and passive consumption. Ironically, strapping a smartwatch to your wrist physically frees you from your phone as Allison Johnson puts it here:

"For me, embracing the smartwatch life makes me feel like I’m in charge of my phone rather than the other way around. I don’t need it on my person at all times anymore"

My phone lives on my desk or in a bag. I actually find it annoying to have it in my pocket now. So much so that I'm looking into getting a cellular Apple Watch in the new year so I can fully leave my phone at home most of the time.

With my Apple Watch on, my phone has become what I think it was always really intended to be — my field computer. A device small enough, yet capable enough, to get some stuff done when I'm out and about, but nothing more. It's a small change, a perceptual change, but it's been hugely impactful to me. The phone moving from the everything device back to a functional gadget in my arsenal of tech tools has helped put all of my devices into functional context. Phones are for outside, tablets are for the couch, laptops are for laps and desks and watches are well, for your wrist.

[markdyerhere]

Remove People in the Background

Google's Pixel phones and Google Photos have a feature called Magic Eraser that can remove people (and animals?) from your photos.

While This is nothing new for digital photo editing — Google's just made this dead easy to use for those of us without photoshop skillz — watching product demos people being raptured out of the backgrounds of photos is...unsettling.

Not that wanting a photo of your kid on the beach, alone, gazing at the sea, the treacherous sea is wrong per se. It's just that removing evidence of all other humans in a photo after we've gone through a respiratory pandemic that killed many, many people smacks a bit wrong.

Pandemic aside, I think this feature brings up a lot of bad things about what it means to take a photo right now. Photos are evidence that you are the main character in the movie of your life. You, your kid, your dog, your stuff, is the focus. Everything else is set dressing — a reflection of you, your power, your status.

Again, this is not new. We have aristocratic oil paintings, the walls of Egyptian pyramids, tapestries and so on.

What is new is how easy it is to create the 21st version of Henry VIII eating a turkey leg and how easy it is to share that image with your friends, family, enemies, exes, side pieces.

What's more is that these images come with meta-data, like geo-location tags, so you can know exactly where 21st century Henry VIII got that tasty looking turkey leg, go there, snap your own pic and get similar clout. This got me thinking about a great Caity Weaver piece about #VanLife

"To prepare for this regular-shaped mission, I threw myself into the #VanLife corners of TikTok and Instagram. Accounts of popular “vanlifers,” as they are known, are an infinite reservoir of gorgeous, unpeopled scenery previously encountered only in desktop backgrounds: sunrise canyons, sunset oceans, high-noon highways that stretch on, carless, forever."

The piece goes on to describe the throngs of people Caity and her travelling partner had to fight through as they tried to grab idyllic shots of themselves, the van and nature.

The goal is the pic, but when that's everyone's goal that's a problem. If only we could disappear these fools out of the photo, right?

I don't blame Google for creating a technological solution to a problem. The camera on the Pixel phones has been a defining feature of the product since gen 1 and their machine learning smarts are increasingly becoming a key part of all of their products, consumer or otherwise. They spotted a problem and created a solution that is genuinely useful (I imagine) for their users.

It's just that the desire to remove others from your photos, from representations of your life, just feels gross, y'know?

[markdyerhere]

The Yellow iMac is Fire and You Can't Tell Me Otherwise

Apple announced new iMacs on 4/20 and people have opinions about the design.

I've read complaints about the chin (sure), the white bezels (uhhh), the stand (okay...), the colours (what!?).

I really feel like these aesthetic gripes fully miss the point of this design.

This is not office equipment; this is furniture. I mean, just look at the yellow one!

I feel like a lot of detractors are thinking of this computer as something that will live on a desk in an office and never to be seen by others. Fair enough. That's typically what we do with desktops. But that's not the story Apple is telling with this design.

Apple's promo images and videos show the new iMac in bedrooms, living rooms, offices, kitchens (I love a kitchen computer!). And, unlike any other all-in-one desktop, you'll actually want to put an iMac in those rooms.

"Computers aren't just for spreadsheets or for tucking in the drawer when you're done. They can be a part of your identity same as your clothes and your furniture. You use them just as much, so why not?" - Michael Heilemann, principal designer at Squarespace

The new iMac isn't just a functional and expressive design piece for the home, it'll look just as good at the front desk of your business.

Back when I used to go places and do things my main encounters with iMacs were in small, design focused businesses. The fact that the back of the iMac has the brighter colour is a dead giveaway that Apple is aware of this market and designing for them. Who wouldn't want their front desk computer to match the business brand colours?

I could give a little more leeway to the design detractors if the iMac was slow and clunky as a computer. And to be fair, the reviews aren't out yet so we don't really know how it performs.

We do know, however, that Apple's M1 chip inside the iMac rips on the Mac Mini, Macbook Pro and Macbook Air. I don't think we have anything to worry about with the iMac in terms of performance.

"Apple's goal is modern retro, but a breath of sunshine -- or a bright yellow iMac -- is certainly appreciated after a year of pandemic and unrest," - Avi Greengart, an analyst at Techsponential

A computer that looks this good and potentially performs as well as its siblings is a colourful recipe for joy. After a long year+ being stuck inside the last thing I want to bring into my life is a another silver or black rectangle. If I'm going to be stuck on a computer for the majority of my work and social life, it better be a computer that makes me smile.

[markdyerhere]

✂Copy/Paste on Phones is a Power User Feature 🖌

Copy/paste occupies this weird space as being completely innocuous and tremendously important. I joke that it's 90% of my job and apparently I'm not the only one who feels that way.

I once asked a coworker at my old job to copy/paste something on their phone and send it to me. They didn't know how.

I was shocked.

I canvassed the rest of our coworkers to see if anyone knew how to copy/paste on their phone. The vast majority of them did not.

Why is copy/paste on phones a power user feature?

Well, for starters text manipulation on phones sucks.

Text is still hard to select on a touch screen. It's difficult to maneuver your cursor around (yes, even when you use the virtual trackpad). Once you have managed to select the text you want to manipulate you have to play a delicate game of touching in just the right way to get the option to cut/copy/paste/select all.

Still, we do type on our phones a lot. But what are we typing? Usually short bursts of text. Notes, search queries and most importantly messages. Lots of messages.

Messaging apps, anecdotally, are the main apps people are typing on their phones. The typing is conversational and spontaneous, so it's unlikely you'd be pasting in text from another source.

It's also unlikely phone users will have two or more text rich apps or documents open at the same time and need to move text from one to the other. The phone medium doesn't really lend itself to that behavior.

(Multitasking on mobile devices is for another post. I have some thoughts!)

It's starting to sound like copy/paste on phones is a power user feature. Heck, Apple didn't even include it on the iPhone until 2 years after it launched.

So why was I surprised?

A phone’s form doesn’t necessarily dictate its function, but it definitely dictates its perceived function.

My phone is a computer running an OS and by extension it should have typical computer-running-an-OS capabilities like copy/paste. 

To my old coworkers, their phones are, well, phones. They're computer-like, but removed from say a desktop or a laptop. (What's a computer, anyway?)

To me copy/paste is something so commonplace, so obvious, so boring that I took for granted as something everyone would just assume to use on their phones. I have a decades long history of using computers -- a luxury my former coworkers may not have had en masse and certainly a luxury that a majority of the world’s new internet users don’t enjoy.  

I return to this episode from my old job often because it really challenged my biases about how I think about tech and highlights the importance of empathy in the design process. 

So yes, copy and paste on phones is a power user feature. 

[markdyerhere]

🔔On Slack & Opting Out of Chaos 🔕

I love Slack. I consider myself good at Slack. I think it's a much better way to communicate by text for work than say, email.  

Still, Slack has problems.

I think the main problem is that you need to opt out of a lot of Slack defaults in order to make Slack useable. 

The reason I'm good at Slack is because I've taken the time to customize my Slack experience. 

I've turned off specific @ mentions (@here may as well be @nobody 🤷‍♂️)

I mute or leave irrelevant channels 

I set a do not disturb schedule so I'm not getting notifications during breaks and outside of work hours

I unfollow threads when they no longer need my attention

I use the all powerful ‘Mark All As Read’ shortcut (shift + esc) at the end of my day 

That’s a lot to ask of a user!

If you as a user don't take these measures the signal to noise ratio veers hard to mostly noise, which creates a whole bunch of other problems. 

To Slack’s credit, they have acknowledged this.

Still, the user shouldn't have to work that hard to get their primary means of work communication to function how they want it to. 

We know that users rarely change the default settings, so even if an opt out method is available it’s unlikely users will even take the time to change their settings to make the product better suit their needs.  

So, why not make all these opt out features opt in?

[markdyerhere]

Some Quick Thoughts On Opera (📱💻 not 🎶🎻)

My browser is one, if not, the most important apps on my computers. 

It’s where I read, watch, hang out — heck, the majority of my job (and probably yours too) takes place in a browser.

Still, it doesn’t feel like most browsers are designed for the way we use them. 

You have multitudes of tabs open. You’re playing music in one of them, you’ve got a WhatsApp convo going in another. You’re desperately trying to find that one tab that’s got that article that you really want to read, but once you get there the page is plastered with ads so you grab the URL, find the tab with Gmail open and email the article to yourself to read later. 

It’s a mess, but hey people are messy. That’s how people use the web.

Opera is the first browser I’ve used that acknowledges this user behavior and has designed solutions for it:

Too many tabs to manage? Group them into workspaces and you can easily search individual tabs.

Can’t find where the music is coming from? Here’s a nifty sidebar with quick access to Spotify and Apple Music’s web apps.

Guhh so many chat app tabs! Remember that sidebar? Yeah you can have all your chat web apps there too for easy access.

Ads! Why? Ad blocker built in.

Guess I’ll email myself this link. Don’t be silly there’s the sweet feature called Opera Flow that lets you send notes, links and files between your devices (and to yourself). 

This is starting to sound like ad copy, but for real, the list goes on! 

Literally, look at this list (taken with Opera’s screenshot tool)

I could go on (and I might!), but my main thing here is that it is so nice to use a product that feels considered from the jump. 

The problems that Opera is addressing are very real -- we’ve all seen our fair share of a-billion-tabs-open-at-once nightmares.

To go through this list of features and then use the product feels like you’re using the end result of a team working their butts off to design something really thoughtful and good.

And that’s, well, that’s just swell ain’t it?