don-code

RIT back in the late 2000s was practically overflowing with hardware in the dumpsters. I easily made a few thousand dollars through daily visits over four years - enough to keep me eating well (not out of the dumpsters themselves, like the article), or at least better than I would have on a college budget.

I also lived with someone who did this - he called himself a "freegan". My roommates and I frequently worried about him unintentionally ingesting rat poison.

I'd argue that restaurants _don't_ have the manpower to do this. Restaurant margins are small as it is, and while I don't have concrete numbers to back this statement up, it feels like designating (or even hiring) someone to upcycle products like chicken bones into stock would be prohibitively expensive, in comparison to just buying prepackaged stock in bulk.

The author differentiates between "serverful resources", like RDS and OpenSearch clusters, and "things that should be a platform team's responsibility", like VPCs.

I'm currently running a shared RDS cluster, for instance, that's used in a platform capacity. Workloads are isolated from each other; there's a clear pipeline for my team and others to update it; and the eventual goal is to build out monitoring to make it evident which workloads are putting stress on the system, so that the owners can respond. We chose this architecture to save on infrastructure costs, at the expense of marginally higher operational costs, as we explicitly wanted to avoid a proliferation of tiny RDS clusters with each new service. The expectation is that the higher operational cost will ultimately be lower than what infrastructure costs would be.

As a New England native, I'm ashamed to admit that outside of television and movies, I had never seen tenpin until I went to college. I rationalized it as some kind of Hollywood in-joke, rather than a real thing.

Nowadays, bowling alleys in the area probably skew two-thirds tenpin, one-third candlepin. Luckily there is a candlepin alley about two blocks away.

Home pesto maker here. It's not for the faint of heart. Each year I grow four basil plants - it takes close to an hour to completely de-leaf and de-stem just one plant. I end up with a stock pot full of leaves. Those are then ground with pine nuts (close to $7 for a small jar - I need 4-5 jars), close to a gallon of olive oil (the brand I use comes in gallons, so about $30), and lots of cheese (I honestly don't remember how much I spend on cheese). The result is great, but it's definitely a day's work plus close to $100.

This feels similar to backdooring encrypted systems. The upsides are similar: some trusted entity has the ability to act in a supposed best interest - national security in the case of encryption, reducing e-waste in the latter. Unfortunately, the downsides are also similar: it's necessary to put an outsized amount of trust in an entity (government, Apple, or whomever) that not everyone wants to place that trust in.

It feels easy to think that it's the case that your employer and coworkers are a family.

In my case, I stayed six years in my first job out of college - as did most of my coworkers (the company was pretty stable up until the very end, when I left). I made a lateral shift within the company, and also made it to senior, under their wing.

The illusion started to come apart after a layoff (35% company-wide, 45% in my department, 60% of my team). I started looking for another employer at that point, because it was pretty obvious I'd only been retained to keep the lights on. When I gave my resignation, I wasn't met with any attempt at retention like higher pay. I _was_ told that it'd be personally detrimental for me to leave, because the company was my family (they knew I'd had some unrelated domestic issues that year), and without them, I'd be lost.

And of course, I agreed with them - for one week, after which I shook my head, said "what on earth am I thinking?", and resigned again. It took some time to realize that it was mostly a manipulative attempt on their part.

Back in high school, I fulfilled my community service requirement at a local nursing home. Several of the residents there had no local friends or family, and as a result, no visitors. I was tasked just with giving them someone to talk to.

What surprised me is that most of the residents did not want to talk to someone - they wanted to be left alone. Many assumed they'd be going home, and didn't want to build any friendships. The one exception who stood out was a man in what I believe was his late 70s, functionally deaf but still able to speak. I'd communicate with him by writing on a whiteboard, and letting him speak in response.

This man over the course of several weeks gave me his life's story - if I recall, he'd worked most of his life as a Boston firefighter. He had quite a bit to say, and I could tell that given the nature of interacting with him, no one else had really spent time talking with him since he'd become deaf. I don't know how long he had been deaf for, but I imagine this wasn't terribly different from being "locked in".

My second to last week at the nursing home, he told me that he'd said everything there was to say - he didn't have any more stories, but was happy to have seen me come back.

My last week there, I went to go see him anyway, only to find the bed clean and his things moved out. It turned out that he had passed away earlier in the week.

I'm happy that I was able to give him those few hours.

In the past seven years, the only consumable I've bought for my moka pot is a new seal - $5 on eBay.

The pot was made in 1947. It was my father's. Before that, it was my great aunt's.

These are not statements you could make of really any product made today.

I've "tried" cassette tapes three times now.

The first, which at this point is insignificant, is when they were period-accurate, and I didn't have a frame of reference. They were pretty good then.

The second time was around 2009, when I bought my first car, and the steering wheel had buttons that could control (pause / seek, not just rewind / fast forward) the cassette deck. So I bought a recorder and started moving some music onto it. When I got it into the car, though, I learned about "wow and flutter": the motors in the cassette deck being ten years old, they no longer rotated at a constant speed. Instead, they'd speed up and slow down, sometimes at unpredictable frequencies, so the pitch of the music would be increasing and decreasing all the time. Sort of like a truck horn, constantly going by you.

The third time was just this year, when I had the opportunity to buy an album digitally for $9, or digitally and with a cassette for $10. Naturally, I had to see what I'd get. I put it in a (different) car's cassette deck, and this time was greeted with music that sounded like it was recorded by Alvin and the Chipmunks. This car doesn't get driven much, and when the alternator is charging the battery, 12 volts is more like 15 - and the motors in the cassette deck again care about that.

Digital has its own shortcomings (I own too many devices with horrid DACs), but I vastly prefer being able to listen digitally and get a better-sounding result.

It took awhile before PCs really caught up with the demands that a full-fledged Active Desktop would put on them. I remember trying it on my 83MHz Pentium with 16MB of RAM, and later a 233MHz Pentium II with 32MB of RAM, both without much success.

Towards the early 2000s, I was using it successfully on an Athlon XP with 256MB of RAM, displaying some live-updating weather widgets. That worked wonderfully, and actually felt like a great use of Active Desktop. Screenshot: https://hardwarehacks.org/lt/bcweather4tn.jpg

I've noticed this as a cultural difference between "self-improvement" type engineers, and "dark matter" type engineers.

Self improvers are the ones reading Hacker News, going to conferences, and generally trying to hone their skill.

Dark matter developers are a class I've shamelessly stolen from https://www.hanselman.com/blog/dark-matter-developers-the-un... - which is definitely worth a read for the context, but the gist is that they're in this as a 9-to-5.

The trouble comes where the two groups meet. We can all learn about how to ask better questions and more generally improve our communication skills, but it's vastly more difficult to _make someone care_ to do so. We've all seen and empathized with https://www.nohello.com/, but if you link it to a dark matter coworker who behaves exactly this way, chances are they will see it as an attack, not as a recommendation - and that attack reflects poorly on you, rather than on them.

The North American owner's manual for the BMW E30, which being an 80s car would have been contemporary with the Franklin Ace, has a similar humorous admonition:

> Caution: Although the ABS is very effective, always remember that braking capability is limited by tire traction. Always adjust your driving speed according to the road and traffic conditions. Do not let the extra safety afforded by the ABS tempt you taking extra risk. The ABS cannot overcome the laws of physics.

While I don't agree it's a Spendesk problem, it's definitely a hard position to take that Spendesk shouldn't fix it themselves. When a large (I'd guess, at minimum, double digit) percentage of your virtual card traffic goes through Amazon specifically, having your product trigger Amazon's fraud detection system, however broken it is, isn't a great look, and borders on being a bug. At the end of the day, the customer doesn't really care whether what Spendesk does is perfectly reasonable - they only care if it works.

We have some other examples of this in software:

* There was an article here on HN a couple of days ago, regarding how Windows 95 detected SimCity and treated memory allocations differently if it was running. (https://arstechnica.com/gadgets/2022/10/windows-95-went-the-...)

* Excel still treats 1900 as a leap year, which it was not, for greater compatibility with Lotus 1-2-3. https://learn.microsoft.com/en-us/office/troubleshoot/excel/...

* The Linux kernel has an extremely strict policy of not breaking userland - to the point of keeping buggy behavior in place (I can't find a reference for this one, but I believe it one notorious case involved PulseAudio and one of Linus's earlier "moments").

Is that, by chance, a T500 with the T60 keyboard swap? I ran with one of those for some time, before the lack of graphics horsepower (I think it had a Radeon 3650?) ultimately did me in.

If you're looking for another cool upgrade, I also did a 1920x1200 screen upgrade on mine. I wasn't able to find a Lenovo-native part, but I did find a pin-compatible screen out of a Dell mobile workstation that fit mostly perfectly, after sawing off some tabs with a Dremel.

I'd be very interested if you chronicle this somewhere. A part of me long wanted to do the same with an O2 workstation, but in more recent years it became much less tenable due to the modern Web (half the web is already broken in minor modern browsers like Firefox), modern languages (Rust? or even Python, now that half of its ecosystem is dependent on Rust by way of cryptograpy), and modern tools (I'd be fired for disuse of Slack, which isn't open-protocol so that it could be used with a portable client).

I bought it used, and with the expectation that I'm going to be on the hook for my own long-term maintenance and support, as is true with more or less everything I buy. (Full-time Linux user here.)

The sad reality is that the experience of running an off-brand phone with an easily-replaceable battery would, long-term, be much worse than running a locked-in flagship that LineageOS builds for.

Interestingly, the Pentium Overdrive mentioned in your linked article will actually run relatively recent OSes - the latest release of Slackware, for instance, should still run on it.

I have the 83MHz version in my old 486 (though it's really not a 486 anymore - ha ha), and got Slackware booting on it out of sheer curiosity a few years ago. What surprised me is that, modern software aside, 14.x wasn't much slower than a period-accurate Linux - Slackware 14 (2010s) and Mandrake 7 (1999) booted at more or less the same speed.

This may work for common phones (aka: Apple and Samsung phones). I had to retire a perfectly workable, but five years old, HTC phone after three such shops refused to fix it, citing unfamiliarity and cost.

I now have an ASUS phone, and fully expect the same to happen in several years.

Is this something that's documented publicly? I'd love to read more.

I self-host on a 3-node Kubernetes cluster in my basement. It runs on three off-lease HP DL360 1U servers that I bought for about $120 each. I work heavily with Kubernetes in my day job, so this also gives me a playground for things that I may need to know about for work, or also things I'll _never_ need in my day job, but are good learning opportunities anyway.

I have an overspecced solar install on my roof that make the electricity used by the cluster essentially free to me, so the only recurring cost is business-class Internet service.

I personally use a black desktop background - and a black shell background as well - as a way of establishing negative space. It feels much cleaner and less distracting than some of the other common backdrops I've seen: colorful wallpapers adding art to my desktop, semitransparent shells distracting me with what's behind them, and I'd be remiss without mentioning Aero on Windows, which turned the entire window stack, titlebars and all, into layers of transparencies.

In a former role, I used EC2 in us-east-1 to host the front door e-commerce site for a consumer electronics company. AWS suggested that we go through the Infrastructure Event Mangaement process (https://aws.amazon.com/premiumsupport/programs/iem/) for Black Friday and Cyber Monday, so that staff on Amazon's side could guarantee that they'd have capacity to run our system at its forecasted peak.

The strategy they helped us arrive at was two-pronged:

1. Pre-launch all needed infrastructure. Yes, for all their "cloud scale", it was actually suggested that we preallocate all of our servers the week before, rather than rely on autoscaling.

2. Order capacity reservations for all of those instances (https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/ec2-capa...). This ensure that, if any of those instances go bad, we'd be able to relaunch them without going to the back of the line, and finding out that there was no more compute capacity available.

I'm deeply reminded of MindMaze, which shipped with old versions of Microsoft Encarta in the 90s. While I'm sure the questions were canned and not programmatically generated, it behooved you to go read up on the subjects that it was being asked about, using none other than Encarta itself.

With regard to:

> in principle you could create a “mirror human” with everything flipped, which is good because it’s insane and they would be immune to infectious disease but also to food so to keep them alive you’d need an entire mirror life ecosystem, although where is my mirror life science fiction

I highly recommend Neal Stephenson's _Anathem_.

About 14 years ago - before I'd taken as much as an intro to CS class - I wrote some software that helped a bar keep track of who'd drank what. They were the type of bar where, if you drank every beer they had available, you'd get a free mug. Prior to it being computerized, the staff used index cards in shoeboxes. Lots of the wait staff's time was lost fumbling through those boxes, unsticking them from each other (gross!), etc.

I've since gotten a degree and written software for a handful of companies.

When I think of how many people are actually _using_ my software, though? Fourteen years later, the mug club software is still live in a production environment, used every day by wait staff who turns over every few months. No doubt hundreds - potentially thousands (it got deployed at a few different bars) - of people have interacted directly with it. That code embarrasses me nowadays, but as far as impact goes: that's probably it.

I have the same FM transmitter that the author mentions. If you're worried about an extra digital-to-analog step somewhere in your audio pipeline: it's a moot point. The transmitter's acoustics leave _a lot_ to be desired: there's background hum, bass goes very flat, there's tons of intermodulation around center frequency, and center frequency itself tends to jump around when amplitude increases. For small Part 15 transmitters, this sort of thing seems like the norm.

Yes! This was actually my first shot at home servers as well - same rationale and all. Old Pentium laptops seemed to get thrown out every few weeks, easily ran Windows Server 2003 (I hadn't seen the light yet), were reasonably easy to repair, used very little power, and were practically silent in operation.

Someday I'd like to chronicle how my homelab evolved, but at the end of the "laptops" generation and immediately prior to the "VMware on a desktop" generation, I had an old DEC (Intel), an AST, and a Gateway laptop, all running under my parents' couch.

I run my home site off a 3-node Kubernetes cluster with 16-core CPUs and some tens of GBs of RAM each, off a business-class Internet connection with a static IP, and I'm still under no pretense that some arbitrary new restriction on behalf of Internet vendors won't render my setup useless one day.