macNchz

I imagine that, if attested cameras like this come into any sort of regular use, you'll see additional layers of metadata mixed into the signature—a depth map, GPS, accelerometers, operator biometrics etc, none of which are necessarily infallible, but which certainly create considerable barriers to faking things.

I've been able to do really cool stuff that I would never have otherwise bothered with by having an LLM generate Overpass queries + walk me through complex setup steps with QGIS.

Customizing tools is extremely common, and many manufacturers do ship moldable handled tools—they’re made of wood.

I frequently have multiple coding assistants going at once—Gemini 2.5 Pro via Aider as the workhorse for most standard changes, Sonnet 4.5 via Claude Code for question answering, documentation, test case development, or broad based changes to many files in a project, then GPT-5 for more complex diagnostic or architectural type things—I don’t generally like the code it writes, but it will often be able to fix situations where the other models get stuck in some kind of local maxima.

I think this is correct, but I also have a sense that smartphones and infinite scroll algorithmic social media feeds have acted as a one-two punch, significantly more impactful in combination than each would have been individually.

Modern social media provides effectively limitless opportunity for consumption, but smartphones unlock the ability to actually drink from that firehose regardless of time or place.

Rest is critical for recovery from any sort of stress, but for many people today there is zero rest from this stuff, it’s first thing in the morning through last thing at night, it’s in the car while driving, it’s during the thirty seconds between customers at work, it’s on the beach, in the pool, waiting in line at the post office, at the dinner table, during a movie at the theater, on the toilet, on an airplane, and a million other times and places that would otherwise provide “rest” if it weren’t for smartphones.

The cycle of car costs, repairs, and impacts of breakdowns for under-privileged people is a pretty serious poverty trap in places that have no other options. https://www.nytimes.com/2025/05/04/us/trump-tariffs-used-car...

Not clear to me what the "External organizations with access" actually means or who decided which organizations to add to the list, but it's curious to me that the campus patrol for a college 250 miles away (SUNY Old Westbury) has access to camera data from Ithaca.

The great majority of development in American cities over the last 75 years has been single-use, with neighborhoods of exclusively single-family houses separated from nearby commercial strips with big parking lots along wide roads.

The downtown/center of older cities may still have mixed use, and there have been changes happening in recent years to allow/build more apartments and mixed use areas, but, generally outside of the densest parts of the largest cities commercial and residential areas are required to be separate, with personal cars as the primary/only way to get between them.

This has been a bit of a self-reinforcing phenomenon, IMO, as car-first infrastructure puts people at the mercy of traffic congestion, and means that any apartment building or business in their vicinity will result in more cars passing through, more congestion, more competition for parking, as well as the presence of the large parking lots that cities mandate for any new construction, which themselves make it unpleasant to get around in any other way.

Businesses paying people already file copies of the W-2s and 1099s that they send to their employees with the IRS, meaning that, for a very large chunk of Americans, the IRS already knows everything needed to fill out their tax forms.

I felt a little bad but laughed out loud reading some of the other stories of peoples' problems with the Wagoneer—one person reported that the first night they brought it home brand new from the dealer, one of the headlights wouldn't turn off no matter what they did and just stayed on the whole night shining against the house. Spooky!

Looks to me that the browser version requires the targeted website to be iframed into the malicious site for this to work, which is mitigated significantly by the fact that many sites today—and certainly the most security-sensitive ones—restrict where they can be iframed via security headers. Allowing your site to be loaded in an iframe elsewhere is already a security risk, and even the most basic scans will tell you you're vulnerable to clickjacking if you do not set those headers.

I've been thinking in a similar way over the past year or so—we're seeing the emergence of more widespread access to custom software for use cases that previously never would have justified the investment.

There are so many situations where a little program can save one or a few people hours of work on tedious tasks, but wouldn't make sense to build or support given traditional software development overhead, but that becomes realistic with AI-assisted development. This is the idea of a sysadmin who has automated 80% of his job with a variety of shell scripts, borne out into many other roles.

We've seen the early phases of it with Replit and Lovable et al, but I think there's a big playing field for secure/constrained runtime environments for non-developers to be able to build things with AI. Low/no-code tooling increasingly seems anachronistic to me: I prefer code, just let an AI write and maintain it.

There's also a whole world of opportunity around the fact that many people who could benefit greatly from AI-built programs are simply not particularly suited to building them themselves, barring a dramatically more capable AI, where I think enterprising software engineers can likely spin out tons of useful stuff and address a client base they might never have previously been able to address.

Working on various teams operating on infrastructure that ranged from a rack in the back of the office, a few beefy servers in a colo, a fleet of Chef-managed VMs, GKE, ECS, and various PaaSes, what I've liked the most about the cloud and containerized workflows is that they wind up being a forcing function for reproducibility, at least to a degree.

While it's absolutely 100% possible to have a "big beefy server architecture" that's reasonably portable, reproducible, and documented, it takes discipline and policy to avoid the "there's a small issue preventing {something important}, I can fix it over SSH with this one-liner and totally document it/add it to the config management tooling later once we've finished with {something else important}" pattern, and once people have been doing that for a while it's a total nightmare to unwind down the line.

Sometimes I want to smash my face into my monitor the 37th time I push an update to some CI code and wait 5 minutes for it to error out, wishing I could just make that band-aid fix, but at the end of the day I can't forget to write down what I did, since it's in my Dockerfile or deploy.yaml or entrypoint.sh or Terraform or whatever.

I rented a Jeep Wagoneer recently and found it to have such comically glitchy electronics that this comes as no surprise. The second day we had it, the liftgate stopped latching entirely, it beeped and popped some error messages on the dashboard and simply wouldn't latch shut at all, no matter what we did. Searching the internet produced lots of people with the same problem, reporting that it required a software update to fix. There was no manual override to the electronic latching mechanism.

Luckily we were near a location of the rental car company—rather than deep in the middle of nowhere where we were headed—and exchanged it for another of the same model, which was all they had available. The next 1000-something miles we drove were filled with endless weird glitches:

- When a passenger plugged in their Steam Deck in the back, the entire infotainment system cut out and went black, including the instrument panel, and then started glitching in and out until they unplugged it.

- When parking, the driver's seat would retract slightly to make it easier to get out, but it never moved forward again, so the seat would get further back at each stop until it was manually repositioned.

- The entire drive the system flashed an un-dismissable error about a rear seat latch, which seemed completely functional.

- The TPMS light went on and off periodically as it seemingly lost and then regained signal from one wheel or another.

- The system flashed errors related to the automated cruise control being unavailable/broken at random times.

- The electronic parking brake kept applying itself while briefly paused in parking lots.

- There was something inscrutably wrong with the climate control that we never really figured out where sometimes it'd just get hot inside the car despite no change to the AC settings.

When we got back I found tons of people online talking about similar (often worse) issues. Incredibly terrible for any new vehicle, never mind one that costs $80k.

This is a misconception—research has found that people entering hospice often live longer than those who do not:

https://www.sciencedirect.com/science/article/pii/S088539240...

There are many who will raise their hands with anecdotal counters to this, but I think much of that is borne from misunderstandings about end of life generally, which is a charged and difficult topic lots of people would rather not learn more about.

I highly recommend the book Being Mortal by Atul Gawande for anyone who wants to explore the topic further—or really for anyone who has loved ones at all!

Depends a lot on what kinds of games you play, I think—I built a PC in 2020 and originally set it up to dual boot Linux and Windows, but over time I used the Windows partition less and less and wound up deleting it last year.

I realized recently that at some point I stopped even checking ProtonDB before buying games on Steam, I guess because its been so long since I've run into one that didn't work. I play a pretty wide variety of games, but not so much the type of competitive multiplayer FPS that seems to have the worst Linux compatibility due to anti-cheat.

This is pretty location specific—in the US 42% of households had home internet in 2000.

https://www.statista.com/statistics/189349/us-households-hom...

There are too many food (and personal care and clothing etc) chemical additives for the average person to remotely be able to keep up with the details of each, especially given not all products even need to disclose them–the charitable, or simply non snarky, reading of that kind of comment is more like "I don't want to eat food with unnecessary/under-studied additives in it."

For anyone interested in a deeper dive into the history and impacts of Lyme Disease I highly recommend the podcast miniseries Patient Zero: https://www.patientzeropodcast.com/

Minitel comes to mind as a genuinely popular predecessor: https://en.m.wikipedia.org/wiki/Minitel

I’ve heard it blamed for stunting France’s later adoption of the internet, because people were able to do many useful things on it and didn’t have as compelling a reason to get online to the internet as they did in other countries with no similar system.

This is even a somewhat less aggressive flow than it used to be, I think. I dug up this screenshot of a part of the cancellation process as it existed when I went through it in 2018 or 2019, which confused me at the time: https://i0.wp.com/ebookfriendly.com/wp-content/uploads/2019/...

I had an annual subscription, and the options on that page made it seem like if I were to cancel that I'd actually lose access immediately, forfeiting the remaining value I'd already paid for. That wasn't in fact the case, but clearly it was designed to guide you into using the "Remind Me" button instead, which I imagine is a very leaky bucket—surely many people who fully intended to cancel would wind up missing the notification three days before renewal and get billed for another year.

Additionally, the text of the buttons and lack of explanation of what's going to happen gives that screen some finality: if I click "End my Benefits" does that mean they end immediately with no recourse? The next page actually showed that you could end and get a refund, or end at the end of the already-paid period, but it was obviously designed to make you uncertain.

Source of screenshot: https://ebookfriendly.com/how-successfully-cancel-amazon-pri...

It seems like they've been adjusting some of the language on the screens in the years since, but still in super confusing ways: https://www.reddit.com/r/amazonprime/comments/wdy84t/cancell...

In my view a lot of this hinges on how well the results are identified as ads, and whether they’re vectors of fraud.

I’m not inherently opposed to ads that are relevant to a user’s search query, but I am opposed to watering down their visual differentiation until they look just like regular search results. Once upon a time Google put ads on a yellow background labelled “Ads”. Now they’re “Sponsored results” and they look mostly identical to the rest. This is simply not about providing interesting and relevant alternatives, but about tricking the user into clicking the ad so Google can charge the advertiser.

What I truly can’t abide, though, is the volume of fraudulent and malicious advertising circulating their network. Given Google’s $100 billion profit in 2024, the amount of fake/scam versions of real websites that they allow to appear in search ads, or deepfake Elon Musk bitcoin giveaways they allow in YouTube prerolls is a calculated choice, not an inability or lack of resources to prevent it. At the end of the day it would eat into their profit if they were to make it harder to post deceptive ads.

Armonk and Greenwich don't really make sense if the idea was to create a perimeter around Manhattan.

I feel like advances in manufacturing and materials science have really made some massive strides in the past 10 years or so—my sub-2lb 2p tent feels just as durable as my 7lb backpacking tent from the late 90s did.

Recording videos to send to our AI Tamagotchis so they don’t start complaining that they’re bored.

People are adaptable. Likes and dislikes and comfort zones are all malleable. I never liked working out in the slightest. Never stuck to any sort of "gym routine" more than a few days. Did most of the Couch to 5k program in college but never kept running. Just...never liked it. I had lots of friends in high school and college who ran Cross Country, and was always a bit baffled about the appeal. It seemed terrible, honestly.

When my friend randomly suggested that we try a very ambitious hiking route, I knew it would absolutely suck if I didn't train for it. I got a gym membership and told myself I'd at the very least set foot in the gym 7 days a week for the first few weeks, just to build the habit of going. I was motivated to make sure I didn't slack off and ruin the hike for the group by being undertrained. A few months of that and the hike went great.

When we got back, though, I found it felt weird to not go to the gym in the mornings before work (as a decidedly NON-morning person my friends and family looked at me like I'd grown a second head when they heard me say I was working out before work). I started running outside on days the gym was crowded, and it felt good! In the nearly eight years since then, there have been only a handful of weeks where I didn't go for a run—I genuinely really enjoy it, no motivational tricks required.

I have been heavily using the Gemini API via Aider for a few months and it has been absolutely stable. Claude, in comparison, has been much flakier. OpenAI somewhere in between.

Tesla...certainly isn't top of mind when I think about makers of technology products that permit true ownership of the hardware / respect their users' privacy.

It sounds to me like the distinction here is that the ban in NY specifies the entire school day, as opposed to just during class.

The horizontal rear waistband zipper pocket on Patagonia Strider Pro running shorts genuinely makes my phone not noticeable at all during runs, unlike any other shorts I’ve tried. My experience is limited to smaller phones (6S, 12 Mini) without any cases, though.