Me too, on the VIC-20.

I think something like the Commodore PET might qualify. Back in the day, I saw it used for everything from cash registers to accountants' workstations, but rarely for anything else.

I think that the original IBM PC was conceived and marketed as a business machine and only grew beyond that because of Microsoft's deep commitment to it as a platform and IBM's uncharacteristicly open specifications and design.

If not for that combination, the PC might never have left the office and most of us would have stuck with the companies who were actually breaking new ground, Apple and Commodore.

That doesn't surprise me. I have Haiku running in a VM, but haven't looked at it in 2 years, despite the fact I used BeOS as a daily driver back in the day.

Well, if you want "compile something unstable yourself," here is their official documentation for ARM64.

And here is someone's progress report on porting to RISC-V. They seem to have started in 2021, so maybe they were successful.

Whenever I price something, I look at the whole package. If I like what a company is doing, I don't mind paying extra to support them. Sometimes I win, sometimes I lose. With System76, I feel like I won.

They were the only company I found that was offering Canadians any laptop with Linux pre-installed. (I think Lenovo or Toshiba had something, but they weren't available in Canada.) Having fought mightily with various distros on a wide range of hardware for years, it was critically important that my new daily driver not suck up my time just getting it running and keeping it that way.

Nearly 5 years later, the laptop is still going strong. On top of that, my hopes for their distro have far exceeded any reasonable expectations. I was prepared for the likelihood that I would ultimately need to switch to another distro, but their ongoing development and contributions to the Linux ecosystem have kept me on board and excited for the future.

In the end, I wasn't buying a laptop. I was buying a system, and I've been extremely happy with the outcome.

That said, I suspect my next laptop will be a Framework. Again, it has less to do with the detailed specifics of hardware than in supporting a company in their attempt to do things the way I think they should be done.

Is that all? I bought my current laptop from System 76 3 or 4 years ago based on my perception that both hardware and Pop were mature enough to be the only computer in the house.

There have been some glitches along the way with the OS, but nothing to get excited about. Notably, I've never had to burn things to the ground and start over. :)

There are some ongoing annoyances with the track pad. I don't know where exactly the problem lies but I do occasionally get cranky :).

When I was looking at 3D printers, my wife asked me if I thought I could make money with it. I said "not in a million years." She asked me what I was waiting for, then. Two weeks later I was printing the test model! Sadly, I've made basically no progress with "real" CAD, so I do everything in TinkerCAD.

Fine woods and metals are crazy expensive, but if you keep your eyes open and aren't afraid to ask, a lot of pretty nice stuff is available for free or close to it as long as you're willing to put the time in on reclamation. I just got a nice Schubert chokecherry log for helping take it down and clean up. I got some sheet steel and light duty tubing from some discarded BBQs. And I've got more poplar than I know what to do with. Poplar is the go to tree for yards around here and there is hardly a week that goes by without someone taking down an old tree.

my hobby is finding new hobbies

Sounds a bit like me. I just retired in June and am busy setting up my workshop. The original intention was woodworking with a focus on boatbuilding. As I go, I realize that I'm actually headed more in the direction of hobby manufacturing: a variety of stuff in wood and metal with a side order of 3D printing. On top of rebooting my programming hobby of learning a new language every year from before I "sold out" to Visual Basic and Access.

Don't sweat it. I used to feel the same way. Then, somewhere in my 50s, I realized that I actually had already been doing what I wanted: everything. "Jack of all trades, master of none".

I did strive toward mastery of most things I tried that I liked enough to stick with for awhile. I like to think that I generally achieved competency, but I know that I had a few bosses and coworkers who would dispute that. :)

Learning new things and having new experiences is priceless. If you can find that within a particular career, that's fantastic, but it's not the only approach.

To answer the question a bit more directly, I would guess that demographics here skew a bit older than elsewhere. That is just a guess, based on the fact that sdf.org dates back to 1987.

Two big ones. I bought the VIC-20 shortly after introduction when I was 21.

Big memory 1: writing machine language programs without the aid of an assembler. I couldn't afford the assembler cartridge, but I wanted to break out of the BASIC sandbox.

Big memory 2: finding a military surplus acoustic coupler modem and using the schematics to make my own connector, then writing a terminal program so I could dial in to these crazy things called BBSs.

Are you sure that rounding was broken? Many systems use "Gaussian" or "banker's" rounding to reduce accumulation of rounding errors. Instead of always rounding to the next larger absolute value at .5, they round to the nearest even number. Although it introduces a bias toward even numbers in the result set, it reduces accumulation of error when .5 is as likely as as any other fraction and odd/even are equally likely in the source.

I was taught "banker's" rounding in school (graduated 1974) and have had to implement it a few times to reduce error accumulation.

If you are looking for a rabbit hole, Wikipedia has a pretty comprehensive article, including an example of how the wrong choice of rounding algorithm led to massive problems at the Vancouver Stock Exchange (Canada).

For anyone not sure what this is all about, CBC has a pretty good podcast covering the descent into medieval nonsense.

I had relatives living just a few miles away in one of the hotspots.

I recently retired from the local volunteer fire and rescue service. Several years ago the 911 dispatch service wanted to drop paging for notifications and move to SMS. I wrote a nice little technical critique of that plan. In addition to the basic issues regarding coverage (many members have no cell service in their yards, never mind in the fields, yet pagers basically just work), I learned from my technical contacts at the telco that there were a number of service guarantee problems. In addition to the lost and delayed message problem you discovered, things only get worse when crossing providers. As he put it, it's not so much that it works so good most of the time, it's that it works at all.

Dispatch did go with SMS notifications, but as an add-on to pagers and "robocalls" to registered phone numbers. We tracked notification channels for several months and found that with every callout, at least one member would report getting the SMS at least 20 minutes later than the page or phone call. Note that most members can get to the hall in 20 minutes or less. A couple of times over the years, we got a flurry of SMS notifications after we were on scene.

Friends don't let friends use SMS for urgent or critical communications.

I agree. I have no idea what it takes to run publicly accessible services over the long haul. Hell, I can barely keep my sorry-ass website up!

I know that lemmy itself is pretty new, but I have to assume that the people who've been keeping SDF alive and functional for over 35 years know what they're doing.

Could it be that Stacer and file manager are somehow reporting usable space instead of "absolute" space.

I recall from the early days that there was overhead in the process, so that useable space was always less than formatted space. Perhaps that is still the case.

Did you try logging in anyway? Maybe their email notification glitched. I've signed up a few places without providing an email, then come back a day or two later to find an active account.

I mostly agree, but I've seen elsewhere that the fediverse (or some corners of it) were set up with the explicit intent to be ad-free and privacy respecting.

My opinion is that it all comes down to two things:

1. Will Threads respect that intent?
2. Given the difficulty of moderating content, can we handle the expected volume?

The answers to those questions can guide the admins (and us, I guess) in the decision.

Edit: this comment changed my mind. In a nutshell, if we can't keep a large instance controlled by "the enemy" from destroying what we've got, then we just have to do better next time.

Yes, I would. Even if they are administered by people that have the best interests everyone at heart, sheer size means that they must be taken into account as the tools and clients evolve over time.

It's not that the system itself should be unable to cope with large instances, it's that the only reason for the system itself to gain that capability is in response to the rise or introduction of large instances. Some of what I've seen discussed is the need to change the development roadmap to accommodate the seemingly unexpected rise and possible introduction of very large instances. In other words, those instances are already controlling the direction taken.

Edit: this comment changed my mind. In a nutshell, if we can't keep a large instance controlled by "the enemy" from destroying what we've got, then we just have to do better next time.

I have been making a related point that we should be concerned about any instance capturing too large a fraction of the space. I'm less concerned about the fact that it's Meta than I am about any one instance having a critical mass that gives them a controlling interest.

History has shown that those with a controlling interest eventually use that control for their own benefit.

That's why I joined a small collection of focused instances and try to subscribe to communities that are hosted in their "natural homes" instead of those on generic instances.