ssl-3

It is legal for a person with an appropriate license, yes. But laws (in the US) about RF transmission differ from how many other laws are structured: We're only allowed to transmit what we have permission to transmit -- and by default, we're forbidden.

But right now, I've got an ESP32 WROOM module on a dev board sitting in my windowsill reporting some sensor values over wifi.

The WROOM itself is registered with the FCC[1]. If I stick to using the documented operation, then I have permission to use it in those documented ways. As a practical concept, I don't need a unique license to use it in the way that I am; I'm granted the ability to use it by Part 15 of the FCC rules.

And that's pretty great: The hardware was inexpensive to buy and very easy to find. I was able to spin this sensor up very quickly and get moving on the software side that deals with those readings. It has been a very rewarding little project with a fantastic cost:benefit ratio.

The fear being expressed here (which I don't necessarily subscribe to) is that things like this WROOM module will become less-available when/if methods to use it in undocumented ways are published and start becoming used.

[1]: https://fccid.io/2AC7Z-ESPWROOM32

I used to do this kind of thing with TiBackup when Android phones were easy to play with and experimentation was more common.

I'd back up the phone to an SD card and/or send the backup offsite over the network.

After that, I could nuke whatever I wanted -- or factory reset it, switch to a different OS (we often called them "ROMs" back then), play with that for a bit, switch back and forth, or whatever.

When the time was right, I'd use TiBackup to restore the backup. It didn't even have to be the same phone. Apps and data were restored to where they had been. Home screens looked the same. Settings and preferences went back to how they were. The same web stuff remained in browser cache, alongside the same cookies.

There were occasional outliers that didn't come back perfectly, of course, but broadly speaking: It worked very well. Those outliers were few and easy to deal with.

This required root access. But for a time (years), that was easy.

(I will never accept any defense of any proclamations that we must be restricted from doing whatever the fuck we want with the hardware that we own. If I can do things with a desktop PC, then I should also be able to do those same things with a pocket computer.)

Backup camera is cromulent. Reversing camera is also cromulent.

A reversing camera is handy for people on one side pond when they take their estate down to the tip to chuck out a load of rubbish.

Backup cameras are useful for people on the other side of the pond when they take their station wagon to the dump to get rid of some trash.

> Because thats a reduced redundancy period

So move the bar up to raidz3 or the equivalent? This adds cost, of course, but enhancing redundancy isn't free.

It was something I kept in-mind when I made the decision to use raidz2 for my own purposes: It tolerates two concurrent drive failures before local data loss happens, which I feel is acceptable for my purposes. raidz3 moves that tolerance to three concurrent drive failures.

I mean: What's the alternative for storing X amount of data? A greater number of relatively small drives? That's expensive in its own ways, too.

And adding complexity in that way tends to nosedive metrics like MTBF: If a system relies upon one widget and this system has an MTBF of n, then a system that relies upon two such widgets has an MTBF of n/2.

If the drives all have the same failure rate, then: A system with raidz2 and 10x 10TB drives will tend experience more disk failures than a system with raidz2 and 6x 20TB drives. Both systems can store ~80TB of data, but the one with 10TB drives has a worse ratio of redundancy and includes more parts that will eventually fail.

KISS, etc. :)

> and the longer it takes the more likely another drive will fail during that time.

You mean the old light-weight RAID-5-esque fear? A drive fails and redundancy drops to zero. So the failed drive is replaced or the hot spare is rotated in (or whatever), and the rebuild process begins. Another drive crashes during the rebuild because of the stress of actually-being-read and now the data on the array is presumed to be trash.

That process seems to actually work more like this: Old data that hasn't been read for years finally gets accessed (because RAID rebuild), and that's when some other drive starts reporting errors. But those errors were already there; they just weren't detected until it was way too late.

It doesn't really seem to work that way with filesystems like ZFS. The contents, whether old and stale or brand new, can (and should) be read routinely and checksummed to confirm integrity as part of a regularly-performed scrub.

The weak drives that can't stand to be read will be discovered during this read-only exercise, but they were already dead. We just didn't know it until we ran a scrub. So we scrub early, and often.

That's the same as with RAID, except: With RAID, the same disk still dies in the same way. We just don't know it was dead until a recovery is in-process. RAID, as commonly practiced, can therefore function as a problem-multiplier in ways that other systems can avoid.

When I start a new game of Factorio, I want the run-around-and-chop-down-trees Minecraft-esque beginning to be over as quickly as it can be. I race to get bots and logistics going as soon as possible. Being limited to only building things that are near my character is something I find very tedious, and bots relieve that tedium almost entirely.

And that lets me focus more on the stuff that I do like to work on in the game.

Figuring out a universal assembler that keeps a list of mundane odds-and-ends in-stock. Working out that shipping from the bottomless seas of Fulgora to Vulcanus where oil is expensive can be a net positive, even though many assume that it must be a bust. Sorting out the unusual steel barrel management problems this creates. Tweaking asteroid collection, and conversion, and eventually working the kinks out of the huge space casino ships that start becoming essentially inevitable.

And the best part of those space factories: Once I have a design that is finished, I can copy and paste it. Replication is simple.

There's other parts that I avoid, like: Main buses. They're efficient, but I tend to lean towards organic spaghetti belts because that's more funner-er for me.

Or trains. I don't use them much. They're cool and all; they just don't usually scratch my itch. They promote vast expansion, but when interplanetary logistics and the space factories are going full-swing then my need to expand on any one planet dwindles considerably.

Lots of people play the game differently than I do and that's perfectly fine. Some people like to program in assembler, and some people might prefer to work in Python.

Others can't even get the syntax right. When they need software written, they can get help from a friend, hire someone to write the code for them, or (with enough of the right kinds of supervision) get a bot to do it instead.

Some people enjoy running around and chopping trees.

It's OK to operate at a very low level. It's also OK to work at a very high level. If the results are rewarding, then chances are good that they're doing it right.

(Anyway, the birds are singing again.)

It could have, sure. That would have been easy. But it would have added weight and complexity. That would have reduced speeds and/or altitude and/or range between in-flight refuelings.

We fixed it the other easy way: In the event that an SR-71 makes an unscheduled stop somewhere that wasn't prepared for it, then we send out a bunch of dudes with a cargo plane and some starter carts. That just takes money, which even easier.

It also takes time, but time was not a big deal: The plane was going to be there for awhile anyway. It could go over 15,000 miles between takeoff and landing; it wouldn't have landed where it did if it weren't broken somehow to begin with.

It'd need repaired by the crews who actually knew how to work on it -- who would also need flown in.

It would also need fuel before it could leave. That's problematic since the JP-7 it required was shared by ~nothing else in the sky, so: More deliveries.

Waiting for some Buick V8 starter carts to be delivered from the other side of the world wasn't really a roadblock at all.

I've got the same kind of nostalgia about it.

It's been probably 40 years since I first became aware of the SR-71 myself. I'm definitely not a kid anymore; if I want to feel some sizzle in my SR-71 beans, then I can just get in the car, drive to Dayton, and see one in person.

If that's not nostalgic enough then maybe I can spend some real money on a new-old-stock copy of the first new Lego kit I was given -- set 497 -- and bask in the rose-tinted glory of the past while I put it together on the rug.

But sure: We can make tech cool again. Part of me thinks that tech should be amazing and cool to learn about.

Kids deserve their own new generations of amazing technological spectacles to marvel at, not a resurrection of ghosts from 60 years ago.

Remember how old and archaic it seemed when you were a kid, 40 years ago, and you saw something that was high-tech 60 years before that? We're talking 1926-level tech, here. I sure did love the local history museums that had areas that covered that timeframe, but I recognized them for what they were: Displays of really interesting very old stuff.

We can do better than that for today's kids, but we can only get there by moving forwards -- not backwards.

That's a very logical theory, but it doesn't really seem to hold up in real-world data transfer speeds.

For instance: 7200 RPM SATA drives have been stuck in the peak-performance realm of between about 250 and 300 megabytes per second for a decade, even though maximum capacity has more than tripled over that time.

What's "far too long"? Why is it too long?

I keep my important stuff with ZFS raidz2.

It doesn't matter much to me if the overall process of replacing a dead drive can be done in 40 minutes or if it drags out to 40 hours. The storage system is still alive, still available for reading and writing, and still redundant for that entire time. It can take as long as it takes.

My break-glass-in-case-of-emergency offsite backups are done with restic over the network, so I only had to send 1 initial copy. That first copy took quite a long time. But restic is pretty clever about how it handles things; all subsequent daily backups are what might best be described as incremental-ish. They contain only recent changes and are pretty small with the usual churn rate on this particular pile of data.

(This matters because it keeps the cost of keeping it running quite low, which means that backups actually happen instead of being just a good idea. :) )

---

I'd probably pick different methods if I had a real budget for this stuff, but even then: Why would I be motivated to care about how long it takes to replace a hard drive?

Old news. The first EV cannonball was run in 1968. 210 hours, 6 minutes. :)

The current EV record was set in 2024 at 39 hour hours and 29 minutes.

https://en.wikipedia.org/wiki/Cannonball_Run_challenge#Elect...

A Bolt EV can already easily be the only car that a person would have to own.

Those goalposts are behind us.

I work inside of these plants, but only just barely. I'm not turning wrenches and cutting pipes and stuff; instead,.I work on the back ends of their radio systems.

I just drive over to the radio shanty near the tower, park, and walk inside to poke at a computer or move a wire or whatever.

A friend of mine does work on the corpo AV systems for some of these same places. This doesn't ever put him inside the plant; he's always separated from it by a magical chain-link fence.

He was rather surprised when I told him how arduous the drug testing was, as he was never required to do anything like that (and wouldn't stand a chance at passing if he were).

I wish I could do anything with THC. It's not illegal under Ohio (where I live) law anymore.

But my job involves chemical plants and oil refineries, and they use DISA, and DISA drug tests everyone thoroughly.

I can count on getting a test that involves hair, urine, and breath samples at least twice per year. (It's supposed to be annual and random, but "random" seems to be rather heavily-weighted.)

THC hangs out way too long for that.

It's stupid: A person can be completely ripped on booze every night and cocaine occasionally, or live entirely in a world of psychedelic drugs like LSD or mescaline and that's... fine, I guess?

But smoking a bowl will ruin them on those tests.

Where you at?

Around me (Ohio, USA), nearly every car I see is gas-fired. I only see an occasional EV; they tend to stick out, and they're very unusual.

Things look different when I travel to different areas of the US, though. Poorer areas have even fewer EVs than I see around here, while wealthier areas have a lot more of them.

It's a little bit unusual for me to encounter an EV at all during my ~daily driving (often about 40 miles, all said).

---

For my part: My lifestyle requires a minivan that can serve as a work truck. There are zero EV options for that here.

But if I could get one with 250 or 300 miles of capacity, then I'd have a new goal. :)

For most people's regular driving, level 1 charging is often adequate to cover their daily needs. They can just plug it into a wall outlet. (If they have one that can be used for this; many apartment dwellers do not.)

A Chevy Bolt EV gains around 5 miles per hour when charging at 12 amps from a 120v socket; it's similar to an i3 in this way.

If it's home half of the time and it sits plugged in for 12 hours every day, then it it can gain ~60 miles during that time.

60 miles per day is not enough to cover everyone's needs all of the time, but it's more than enough for all kinds of folks who would otherwise be afraid of having some kind of electrical upgrade required before considering an EV.

Please don't mistake my observation of a thing's presence as a defense of that thing.

Indeed.

I can only speak for myself, but I do get surprise trips sometimes. My phone could ring at any time and the person on the other end will tell me that I need to go to Detroit, or Columbus, or maybe some corn field in Indiana. This is a thing that happens in my life (and I do get paid for these drives).

But I never have a surprise trip that takes more than about 2 hours. And that's... well, that's well-within EV range. I might need to spend a bit of time charging up for the return trip, but whatever: The car has a high chance of starting at ~80% charge anyway because it is plugged in while it sits in my driveway.

I do take plenty of far-longer trips, but they always have some planning. Often days worth, sometimes whole weeks worth, but always at least a few hours. I'll have plenty of time to juice the thing all the way up beforehand (if that's even useful) and also to look at the anticipated route.

And besides: While I can eek ~500 miles out of a tank of gas if I work at doing so, I never do that in one shot. I take breaks. Even when I've got a co-driver who I can swap with on a 1,200 mile jaunt, we walk around, hit the bathroom, refill on libations, and et cetera when we stop to trade spots. We make time to kill.

Since I'm already taking breaks and it's not a race, then the addition of an EV to the game only affects where I take those breaks. But, really: Those intermediate points are not a big deal -- it doesn't really matter where they happen. It just matters that they do happen.

If it is a race: If the whole point of the trip is to catch a show or a game or a ferry or some other such thing that happens at a very specific time whether I'm there or not, then I can just leave earlier. Maybe I'll get some free time to explore the area when I show up, or maybe I'll edge into a parking lot with 5% battery remaining and a growing sense of fear about getting into the venue on time, but whatever. For all 1 or 2 times per year when that's actually important and also outside of normal EV range, I can accommodate them with a grant of some extra time.

But for the every-day stuff? Man. I could do very well with an EV. I could even do very OK with an EV if I lived in apartment, as long as I could plug into a regular wall socket for a few hours while at work (which is cheaper for the boss-man to pay for than the space heater that sits under Cindy's desk).

(I guess an EV might cut off my access to certain routes in places like Wyoming or the Dakotas, but I'll worry about that ability after I gain a desire to go there to begin with.)

An every-day diesel-electric locomotive solves the gearing (torque) problem, and can let the engine run at the most-efficient speed that can produce the required power output. The required output power is dynamic, though, so in practice the diesel engine runs at a variety of speeds along a given route across the plains or over the pass or whatever.

It's hard to make an engine that runs at peak efficiency at a range of speeds and power demands. But when batteries are added, the picture can change quite a lot.

The engine can be built and sized to run efficiently at a fixed speed, and its power output never needs to exceed whatever the generator that it drives is able to accept.

So when it runs, the combined engine+generator set can be run full-assed. No half-assing required. It's a binary state. On, or off. Full-staff, or flaccid.

By reducing variables in this way, all of the fuel that is burned gets converted to electrical power as efficiently it can be for the process at-hand. (It's also easier to quiet an engine that runs at a fixed speed than one that runs at variable speeds; things get a lot simpler when there's only one set of identified frequencies to bother with.)

While the engine is on, some of that electrical power can get used for driving an electric motor for propulsion and the rest can go into a battery.

If it is on and the vehicle is stationary, then all of that electrical power can go into a battery while the engine is remains at full-chooch.

When the battery gets full-enough and/or demand is low-enough, then the engine can just turn completely off. In this mode, all of the electrical power used for propulsion then comes from the battery.

It works. There's even some boats[1] that use these concepts.

It's just an unusual arrangement for now. (It may always be an unusual arrangement if the goal is to eliminate localized combustion of fuel. But as we keep being reminded over and over on these pages, whenever a headline about vehicular electrification comes by: We aren't quite uniformly there yet.)

[1]: https://en.wikipedia.org/wiki/Hybrid_ferry

That a person can see such endless pages of people having various forms of disagreement, and yet somehow manage to conclude that this observed chaos constitutes a clear exhibition of cohesive groupthink is just...stunningly amazing to me.

I don't know why I find it so amazing since it happens with such regularity, but I'm always amazed by it anyway.

In my experience, it always takes 4 interactions, including the initial prompt, when this stuff happens.

It doesn't matter if I verbosely explain my query 3 more times or just tersely and succinctly reject the responses 3 more times.

The number is always 4.

After that, Google's dumbest-of-bots will finally respond to the actual-fucking-thing that I typed to begin with.

I wanted to find a specific web-published story that I'd read years ago.

I knew the website (and it's old; it predates the term "blog"). I knew the context. I just didn't know the URL and I didn't want to page through entries manually like some antediluvian pleb.

So I used a search engine; I Googled it.

Zero results. Nothing. The search terms were common and there should have been thousands or millions of hits, which I would then be able pare down with better terms until my desired result was near the top. But there was nothing.

Zero response from the Google bot, too.

(The lack of bot response isn't unusual; the subject matter involved a man on a balcony and his gentleman sausage, and Google's bot tends to avoid non-Puritanical concepts. But the Google search engine of Not So Long Ago, Really would've been very pleased to present results that match the terms.)

That's an interesting example.

I detest the Proud Boys for a wide variety of reasons and I really wish that they would just go away.

But if they exist on the web, then I want to have the unfettered liberty to find them on that web and see what they've got to say without having it digested, ruminated and filtered before eventually being regurgitated through someone else's mouthpiece.

Censorship restricts my ability to know my enemy.

Right. I've done the same.

And when the new distro-provided nVidia driver does something garish like, say, break XFCE, then it's easy(ish) to roll it back using distro tools and pin it there. After that, just wait until some other more-functional combination of shakes loose in the distro channel.

When a new nVidia driver shows up that promises a fix but the distro hasn't packaged it yet, then the temptation to just download and run the installer that's on nVidia's website. But using nVidia's special-sauce installer taints the system in ways that distro tools deliberately seek to avoid.

And: Oh, man. I'd almost forgotten about the binary tarballs that were intended to be simply dumped into /, where they'd just tromp on whatever. Sure, it was fast. And for some people, some times, it even worked. Sometimes, it didn't work. Other times, it broke other things that had been working. And it left a mess behind every single time.

A little bit of a mess isn't necessarily devastating on a personal system. But the mess accumulates every time such an unmanaged installation process happens until it eventually overwhelms to the point that even the most functionally-disorganized of people become dysfunctional.

It says this this at the bottom of every one of the dumb responses that Google's trash-tier bot puts above the (deliberately awful, these days) search results:

AI can make mistakes, so double-check responses

Oh yes. I stepped through most of those, including the region-hackable Apex and a plays-anything Phillips.

Previous to that, I was doing some process or other on the PC that converted VCDs into a format that could be played on a modded PSX. The quality was terrible and I did it anyway.

When DVD was still a very new and expensive-to-use format, I scored a very cheap SCSI DVD-ROM drive that was a pull from some Sun machine or other. I never did succeed at playing a glitch-free film with it.

Part of that disappointment was due the 3dfx Voodoo 3 card lacking the MPEG 2 decoder that it had, at one point, been advertised to feature. In its place was a mail-in coupon that could be sent in for a software decoder...which never showed up, either.

It was an interesting time in digital video.

It's not just you. Objects that blink are hard to track (our brains aren't good at it), and working to process it anyway creates an unnecessary distraction as a secondary effect. And as a tertiary effect, this added distraction leads to a reduction in safety.

Having some movement can be good, though. This is why I went out of my way to mount reflectors on the back side of my aftermarket pedals.

Along with a constant red tail light. And a fixed red reflector.

(A grown-ass man with reflectors on his bike? You betcha. The dork disc is still intact, too. I don't like crippling surprises, and I'll look way better with a safer bike than I ever can in the hospital.)

Much of that (I'd like to say most) is due to a bad beam pattern and/or improper aiming.

For instance: Round beams suck. The first LED front light I ever bought was nothing more than a flashlight that took 3 AAA cells and attached to a handlebar mount.

It produced a round beam. If I aimed it up/far enough ahead that I could see the road ahead in any meaningful capacity, then it was also blinding to other riders and people in cars.

I could aim it down some and limit others' exposure, but then it was useless at letting me see what I needed to see, myself.

The most recent front light I have came from Walmart, and was sold in Schwinn packaging. It's rechargeable. With this light, other people have zero direct exposure to the LED because it is occluded by the housing.

It uses optics to shape the light. A curved mirror, and AFAICT a lens are inside. The beam it produces is shaped to evenly illuminate my path and it has a very sharp horizontal cutoff.

It's a very sensible design, in that when I aim it at the road so I can best see, other people's eyes are above the cutoff so they aren't blinded by glare. Everyone wins.

But not every bike light has a sensible design. And not every rider has sensible intent.

(The Germans have standards for this stuff enshrined in StVZO, wherein: Stuff like beam shape and cutoff is governed by law. Searching for StVZO lights tends to yield good results.)

VCD was in-keeping with the style of the time. General-purpose compute was expensive, so it was avoided.

Like the red book audio CDs that preceeded them, white book VCDs were a way to make playback a light-weight task on very specialized (maybe even single-purpose) hardware.

That's not necessarily a meaningful goal these days, but our limitations were different back then.

The problem, for me, is that self-running installers can create a mess that's hard to keep track of.

I've run Linux without meaningful package management, as that was kind of the style of the time 30 years ago with Slackware. It can quickly become untenable.

There's no real difference between an uninspected script that gets piped straight from the URL into the shell, or a similarly-uninspected make&&sudo make install routine from a tarball. They can both execute code that does bad things (whether unintentionally or deliberately), and they can both leave a mess that is hard to cleaned up.

I've found that it is better to just avoid going down that road to begin with. Whether distro-specific packages, Docker containers, flatpaks, or whatever: All of these make housekeeping easier.

I would to order two soft taco supremes, a beef chalupa, and some cinnamon crispas. I'm sorry, what? Cinnamon crispas are discontinued? Since when? 1988! Unbelievable!