That doesn't necessarily line up with my understanding of reality.

In the US, a PRI can handle up to exactly 23 concurrent, native g.711u phone calls. That's it's capacity: No more, and no less. It's always 23, with each concurrent call using exactly 64kbps of symmetric bandwidth....just because that's the number of B channels provided.

But if we take that same PRI and make it do IP packets instead using MLPPP, then our capacity is actually reduced. By adding the magic of packet switching, we also add overhead. And with that added overhead, we can only get only get ~19 g.711u calls through that same circuit.

(Now, sure: In a bigger picture, that PRI may be better utilized as an IP pipeline than as a dedicated telephony circuit. It's certainly more flexible that way.

But packetization is not something that automatically improves capacity. It often does the opposite.)

When the presentation of the website has the appearance of being deliberately terrible, then I (for one) am lead to presume that the corresponding book will be just as disdainful.

The OG Raspberry Pi was $50 in today's money -- for the bare board and nothing else at all.

There is nothing to discuss whilst real factors like inflation are willfully ignored.

2012 was a long time ago, and these boards were not as inexpensive as rose-tinted glasses may suggest.

Knowing just a tiny bit about the history of the Chicago River, my first thought when I saw the headline was "Cholera! It's for everybody!"

But learning that they've actually -- finally -- solved that issue really outweighs whatever snark I may have had in mind.

(And for those who don't know: The Chicago River has a really interesting history that definitely includes cholera, and also includes reversing its direction of flow (!) to help solve problems like that.)

13 years ago, we only had the Raspberry Pi Model B at $35 (about $50 in today's money -- same as a 2GB Pi 5), and just as today we still needed to get all of the accoutrements in-place to make it work: A power supply that it actually works with, a case, an SD card, perhaps a wifi adapter...

My time machine is a little rusty, but all that and a breadboard was about a hundred bucks around that time: https://www.linuxjournal.com/node/1206862

Or, you know: About $150 in today's money.

$150 is plenty to buy a used PC system here in 2025 that still works, just as $100 was plenty to buy a working used system in 2012.

As a point of reference: The last used system I bought was a little Lenovo M600. It was $50, delivered, a couple of years ago.

As another point of reference: My daily-driver laptop is a Thinkpad T530 that was ~$200 (I paid a little extra for a disturbingly-clean example that included a discrete GPU and the fanciest of the screens that could be equipped).

Anyway: I saw these same discussions about pricing back when the first Pi was still new -- just on Slashdot instead of HN. People have been comparing the prices of used PC hardware to the prices of new Raspberry Pis for as long as we've had Raspberry Pis.

(And to be clear, I'm not trying to fanboy anything. This isn't Highlander: There can be more than one. I've got Raspberry Pis that do stuff, and I also have PC hardware that does stuff, and I'm OK with this.)

"For that price, you can just use an old laptop" has been true ever since the OG Raspberry Pi showed up ~13 years ago.

And that's great, and stuff, if what a person wants is the most compute they can get for the fewest dollars possible.

But when someone instead wants a quite small computer that is actually friendly to hardware tinkering, and they want to buy it new, then a used Thinkpad will not scratch that itch -- but a new Raspberry Pi will.

(It's a bad comparison. It always has been a bad comparison.)

It's just a summary — of a summary ⸺ of a summary ⸻ of a summary.

Why would anyone tire of something like that?

(/s)

So put a smart plug on a dumb Speed Queen, preset the knob and the buttons, and have it turn on at the time you wish.

Are there any other automation scenarios that you wish to address?

It is with immense displeasure that I find that my playful descriptions have been interpreted as "gaslighting" by someone on the internet who suffers from selection bias.

If your intent is to demonstrate that there is nothing here that can be discussed, then: Congratulations. You've accomplished that.

But tt's not just the buttons -- it's also the motorized knob, sensors, and control logic.

A board with model-specific software and some garden-variety relays is cheaper to copy than model-specific motorized knobs are.

And it's not the singular key to profit, and I don't think that anyone ever said that it is. It's just a part that we can see, and touch.

There's other things that modernization helps with, too.

For instance: Variable-speed motor drives, with a computer brain to drive them (which we already have in the BOM once we abandon the motorized knob).

These can improve electrical efficiency (reducing motor size and cost), and deliver power more smoothly (reducing transmission size and cost).

That's real copper and real iron that is saved by using electronic controls.

(I can do washing machines all week.)

It's not just the buttons themselves (in a capacitive-vs-mechanical sense) -- it's also warranty services.

When the electronicals are bottled up behind a sheet of glass or Perspex or whatever, then: They tend to last longer because their operating environment becomes less hazardous to them.

Exact hourly rates vary too much to write about specifics, but whatever they are: Sending a tech out to look at a thing costs real money (in the ballpark of hundreds of dollars, not dozens) that really bites into profit margin of any individual unit sale.

It bites into the margin even if the root problem is that the owner's roommate's friend pissed into the control panel with a head full of acid. They'll still be paying someone to physically go out and make that determination.

So if mush-buttons generate fewer service calls than push-buttons do, then: It's a big advantage to a manufacturer.

So... I think it's quite easy to understand how we got to where we are: Fewer moving parts + better environmental isolation for those parts = less after-sale risk.

(I don't necessarily like it, but there's lots of other things in the world that make good financial sense at the manufacturing level that I'm also not fond of. I can accept this reality without also pretending that it can't make sense for someone, somewhere.

Good answers? Speed Queen, for one, still makes good washers with real knobs and real buttons, for the consumer who favors these features.

Just add a smart plug or current monitoring and an iteration of Home Assistant or whatever running on a sleepy little Raspberry Pi or a VM/container or something to detect and notify soon after the wash is done. End-of-cycle detection is really all that is ever needed for smarts anyway. And that may sound convoluted, but these are smarts that you control yourself and are about as open-source as anyone may wish them to be.

(I don't want an appliance that I hope to last for 20 years or more to be connected to any networks at all: "Wake up, babe; new rootkit just dropped and our clothes washer is fucked" isn't a meme that I want to live through, even if it does have a nice API running on a stack that was last updated in [checks calendar] 2005.))

I have personally installed equipment that included non-optional, always-on cellular data connectivity that allowed it to be configured and monitored from The Clown.

And Amazon rather famously included cellular connectivity with their early Kindle e-ink book-reading devices -- back in '07.

It's been done before. It can be done again.

Why not buy the fridge that doesn't have wifi smarts to begin with?

If I want to monitor my fridge's temperature, I can buy a widget that does that for a dozen or so dollars and have that sensor talk to the home automation system of my choice. And when the fridge dies or otherwise gets replaced, I can move the sensor to the new fridge. (And when a new sensor comes out that I like better, I can spend another McDonald's Value Meal worth of money to use that instead.)

Besides: We here on HN should all have a certain amount of distrust for devices that self-report problems.

This distrust is part of the reason why ZFS doesn't trust hard drives to self-report issues and does its own checksums instead.

---

But that's a general rant. To answer your question more-directly, if somewhat-tangentially: One of the popular open-source-oriented YouTube dudes (Jeff Geerling?) recently bought a dishwasher that had functional modes that could not be accessed without a wifi connection to The Clown.

And that's... that's not good: In order to be able to use the functions that the thing natively includes, one must always allow it to call home to mother.

The problem with kerchunk-a-wheels and real pushbuttons (and the not-smarts they imply) is that they're expensive.

They're more expensive (and more failure-prone) than the rotary encoders, mush-buttons, and brain-boxes that replaced them.

But one cool part about things like motorized kerchunk-a-wheels is that, upon failure, a motivated person on Gilligan's Island can often mend them back into service with a screwdriver and a sharp rock.

Indeed.

It is important to try to avoid letting perfection be the enemy of good.

Firefox is at least something that is distinct from WebKit or Chromium (which is itself based on a fork of WebKit). That's good.

It's not perfect, in part because deals with Google pay for most of it, but it is still good despite its imperfect status.