« Juneteenth | Main | git yer joy on... »

June 20, 2021

Comments

Even more stringent rules came into effect in 2000 and again in 2007 as a result of some high profile accidents where children were killed.

Bit of a contrast with the US reaction when children were killed by gunfire. Oh wait, those weren't accidents, so it's totally different.

One additional factor in Japan is that there is a genetic divide between people who cannot tolerate much alcohol (similar to Native Americans) and those who have much higher tolerance. There's a geographic correlation to the genetic divide.

I personally have seen a big Japanese guy, over 6 feet tall, not heavyset but definitely 'solid' 200lbs or so, have ONE weak Japanese beer and be three sheets to the wind*

That's probably why BAC isn't the only rule for intoxication. And why a national strict limit makes sense.

(*yes, we were all part of the same group, but I kept up with my Russian colleagues and was only mildly tipsy)

Russell at 9:21: I'm with you. I don't care if the entire car runs on a single computer chip, but I want my user interface to be "real."

And any screen in the front where the driver can see it, much less a touchscreen, strikes me as absolute madness.

gimme knobs and buttons and levers, please. anything that lets me operate the car without taking my eyes off the road.

What? There's no knob app?

Heads up displays are a good thing, if not too crowded, because that allows to keep the eyes on the road (I always wished for that when I learned driving*) but controls need to be in a way that can be handled blindly (i.e. without the need to look at them).

*Actually, I have not steered a car since the day I got my driving licence many years ago.
It would be utterly reckless of me to go behind the wheel without going back to driving school again first. And I'd still be a lousy driver.

If asked to guess, I think they'll go all in on speech recognition before they put the knobs back.

I've been fooling around with speech recognition in an embedded application using free open-source software. With a restricted vocabulary and a close-in microphone it's surprisingly good. Less so as the necessary vocabulary expands and the microphone setup gets worse. There's an outfit in Vancouver that seems to be doing remarkable things. And specialized hardware to support AI models (like the Alexa back-end) is getting much more affordable -- see, eg, Apple's M1 chip with a dedicated neural network processor that performs 11 trillion operations per second. Just writing that makes my head hurt.

So, while taking the Japanese regime and simply moving it to the states might seem to provide clarity, if you don't know the context and cultural factors, it can often lead you astray.

First, thanks for the insight. I don't think the Japanese regime would do at all well in the US, although I'd be fine with something along those lines. For some reason, I found myself reading up on Norway's DUI regime and it is also pretty Draconian.

Bit of a contrast with the US reaction when children were killed by gunfire. Oh wait, those weren't accidents, so it's totally different.

You might be missing something: no one is arguing in favor of drunk driving, nor are any gun owners arguing in favor of school shootings. The closer analogy would be to regulate alcohol purchases and consumption as tightly as many here propose to regulate otherwise legal firearms ownership. We tried something like that before with alcohol and it didn't work out well.

If asked to guess, I think they'll go all in on speech recognition before they put the knobs back.

I use Dragon a lot. It is ok, but a long way away from performing with the level of certainty I would want in a car or truck.

If asked to guess, I think they'll go all in on speech recognition before they put the knobs back.

This is one of those situations where I ask "why?".

The driver of a car needs to be able to command the operation of the car. Steer, control the speed, turn lights on and off, signal intention to other drivers.

Knobs, buttons, and levers are really simple ways to communicate all of that.

Voice recognition is, in fact, really good at this point. But its quality degrades with the quality of the audible environment.

Road noise, wind noise, kids yelling or someone on the radio talking (including saying "stop" and/or other things that may or may not be interpreted as driving commands by whatever software is running things.

I'm sure they can sort all of that out with another several years of research.

But why? Why is it better? Or even necessary?

Software has become the universal hammer in search of nails to hit. I'm not sure it always adds that much value.

For driving, specifically, there is also the fact that so much of driving is automatic.

When I drive my car, I do 1,000 things that basically make the car an extension of myself, without thinking about it all that much. The physical and neurological habits are burned in at this point, and have been since about the second month that I ever drove a car.

I don't have to think about it, I just do it, automatically. It's like walking or riding a bike.

If I had to tell my car to do all of the things I just automatically do, I'm not sure I'd keep up with the additional cognitive load. Not the car, I'm sure the computer can keep up.

I'm not *I* would be able to keep up.

I can see it working for stuff like "turn the lights on" or "change the radio station". I can't see it as a good solution for anything critical. And it's just not that hard to turn the lights on or change the radio station manually, so I don't see the point.

I use Dragon a lot. It is ok, but a long way away from performing with the level of certainty I would want in a car or truck.

Right there with you if it's mission-critical. OTOH, it is (or soon will be) suitable for handling "Turn the volume down," or "Blow warm air on my feet" or "Remind me to stop and get milk."

I'm with Russell. Voice recognition is good for Blue Tooth and whatnot. Maybe reminders as MC says. Beyond that, the steering wheel can and should have a range of hand or thumb-operated commands that do not involve taking one's eye off the road.

This dovetails with the comments about bad driving. Houston was awful ten years ago. Now, I'd trade that 'awful' in a heart beat. I regularly clock drivers doing 90+ on freeways and 60+ on streets. The last thing I want is any kind of added distraction or uncertainty for those clowns.

Also, I stutter from time to time (not awesome for a trial lawyer, but there you go). How would that work with voice commands?

@russell,
An example of the problem I thought we were discussing is "Change the radio station at 60 mph using this touchscreen." There's no tactile feedback, so you have to take your eyes off the road. In addition to not watching the road, many people tend to steer slightly in the direction they are looking. Look to the right to see the touchscreen and some people will immediately start drifting to the right. (I have noticed an increased tendency to this myself as I have become an oldster.)

So, if the design decision is no physical knobs, and the touchscreen is dangerous, what alternatives are left?

any screen in the front where the driver can see it, much less a touchscreen, strikes me as absolute madness.

Totally agree. Madness. And it's going to get people killed. One has to wonder, if it's illegal to drive while using your cell phone (as it is in California -- "distracted driving" is, I believe, the term of art), how is it OK to drive while using a touch screen nowhere near your line of sight to traffic?

Now if they get to the point where you are running complete VR, with verbal modifications of the various buttons and knobs. That might be another story. But that's a completely different story.

The closer analogy would be to regulate alcohol purchases and consumption as tightly as many here propose to regulate otherwise legal firearms ownership. We tried something like that before with alcohol and it didn't work out well.

Prohibition was more like a total ban. Damn few people (albeit, admittedly, some very loud ones) are pushing for anything even nearly that extreme when it comes to guns.

but a long way away from performing with the level of certainty I would want in a car or truck.

Control with voice recognition would require a period of the computer learning your personal voice in order to work. In addition to stuttering, which McKinney mentioned, there are regional and foreign accents. And dialects. Could anything like out-of-the-box voice recognition handle both a Southern drawl and Cockney? (Not to mention things like rhyming slang.)

Plus, you'd need to reprogram every time you added a new driver.

Then there is the detail that cars sold internationally would also need to deal with other languages, of which there are hundreds (over 400 just using the Latin script, which I'm personally aware of). And their dialects as well. Anybody can learn to push a button. No language skill required.

Also, I stutter from time to time (not awesome for a trial lawyer, but there you go). How would that work with voice commands?

There's at least one PhD dissertation in that question :^)

There's a subfield of speech recognition that is sometimes called "speech to intent". For a limited intent "space", things can be much easier. For example, "T-t-t-turn the volume up" in the context of what can be done in the car is much simpler to handle correctly than transcribing "The b-b-beast was t-t-t-transmogrified into a handsome prince."

I've been thinking about adding to Cain's Laws™. Something about software, and that solving a new problem always takes longer than initially thought, but in the long run the solution is better than anyone imagined. It used to be conventional wisdom that computers would never learn to play Go. Given suitable scoring cues, AlphaZero can derive the rules of Go, play against itself for a few days, and reach the point that humans simply can't beat it.

how is it OK to drive while using a touch screen nowhere near your line of sight to traffic?

it's exactly the same as every other car's center console. no car has all of its controls in the driver's line of sight. the climate controls, radio, phone, etc. are always in the center console.

at best, some functions can be controlled from buttons on the steering wheel (which Tesla does) or from stalks in front of it (which Tesla does).

the one frequently-used thing Telsa puts on the screen that other cars put on the dash or a HUD is the speedometer. and they put it in the top left corner of the screen, so it's always just to the right of- and above your right hand, instead of to the left of your right hand as it is in cars with in-dash speedos.

that's the only thing i've had to get used to - control-wise.

wipers, lights, turn signals, gear, radio volume and channel, phone and cruise are all on stalks and wheel buttons.

it's exactly the same as every other car's center console. no car has all of its controls in the driver's line of sight.

No, it is not. Just for openers, you can find the button you want by feel. And feel whether you have rotated a dial. Touch screens, at least the ones I am familiar with, simply don't provide that kind of tactile feedback.

A different question about the touchscreen, cleek. My polarized sunglasses turn my smartphone screen completely black. (Not surprising, LCD technology depends heavily on polarized light.) When I have the phone in the dash mount to use for maps, I have to take the sunglasses off in order to read it. Is the Tesla touchscreen subject to the same problem?

wj,
just tried it... with the sunglasses i have, it start to go black if i turn my head 90deg. so, they got the polarization angle correct.

i've had cars where the LCD screens were always black with sunglasses on.

Just for openers, you can find the button you want by feel.

you can, but if it's a knob, you are still probably going to look at it to see where you want to to turn it to. or, you'll look at a display of some kind that tells you which station, temp or whatever you've selected. or to make sure the button picks the right seat warmer level or whatever. you're already looking at the console a lot.

i was wary of the screen. i thought it would be hard to do things. but it really hasn't been. the important things are right there and music, for example uses huge icons for channels - don't have to be precise, just punch the 1.5x1.5" icon that has your channel on it. and all the other things i really don't mess with much anyway (tire PSI, phone settings, etc.).

i thought it would be hard to do things. but it really hasn't been

Well, I'm getting ready to buy a new car, so I guess I'll be getting some first hand experience.

"what alternatives are left?"

https://www.investors.com/news/technology/crnc-stock-news-cerence-partners-with-sirius-xm-on-car-radio-voice-controls/

Maybe.

I wonder, when carpooling, when an argument breaks out regarding what station to listen to, which passenger's voice will Sirius listen to?

No doubt the insistent teenager's riding shotgun?

FOX, OAN, and the rest of the America-killing radio stations will recognize only Trump's voice to change the channel, and then summon law enforcement harassment to pull you over when you insist.

"Also, I stutter from time to time (not awesome for a trial lawyer, but there you go)."

I went thru a period in my 40's and maybe into my 50's when I stuttered from time to time. It was puzzling, because prior to and since I rarely displayed that.

All I know is when Porky Pig declaimed "bedea .. bediet .. t-th-th-that's all folks!" followed by the Looney Tunes fanfare, as kids, we knew it was time to grab a cookie.

When guns incorporate voice-activation, I don't know how I'm going to voice DE-activate them when a shooter points one at me, and I stutter out: "P..pa..please da-don't sch ..scha ... shoot!"

Just like now.

OAN's voice activated genocide:

https://talkingpointsmemo.com/news/oan-host-muses-about-mass-executions-of-radical-democrats-for-election-crimes

Plus, you'd need to reprogram every time you added a new driver.

All the information pertaining to a particular driver, from voice recognition to seat position, could be stuffed in the driver's smartphone or maybe the key fob.

and Ingraham wants to Defund The Military because they aren't Republicanly-correct.

I personally don’t have a big problem with voice commanding for things like “turn the radio down”, if that’s what floats folks’ boat. If it fails, the worst case is... your radio is still loud.

That said, I’m able to do all the radio stuff I need to do in my car (2012 Mazda 5) without taking my eyes off the road.

And all of *that* said, my confidence level in vox recognition for critical functions is pretty low. I worked for a few years in a company whose bread and butter was vox recognition. It’s very very good, but sometimes very very good is not enough.

Enough people drive enough miles that edge cases are going to be an everyday occurrence.

Mostly, I think people like shiny new gadgets. They’re cool. Nothing wrong with that per se, I just don’t want to absorb those kind of learning curves embodied in 1 or 2 ton metal objects traveling at high speeds.

All IMVHO.

Plus, you'd need to reprogram every time you added a new driver.

(Preface: I prefer buttons and knobs as much as anyone, and you can pry my steering wheel from my cold dead fingers. But if the spec says "speech recognition"...)

The large-vocabulary model I've been playing with comes with no specialization beyond the language model(s). I've done nothing to train it. Accuracy is about 95%. With a sharply restricted vocabulary, much better accuracy. In its case, the problem would be restricting it to specific speakers.

How large a vocabulary? Given this (spoken) input, "Still following reasonable precautions given uncertainty around variant and post-vaccination transmissibility," the only word it got wrong was the last one, which came out as "trans miscibility." Heck, the Firefox dictionary rejects transmissibility. Granted, I read it in Standard American, which most college professors eventually acquire. ("Mike, in lecture mode" according to my wife.)

Carnegie Mellon, where the software came from, has free models for English, Indian English, German, French, Spanish, Portuguese, Chinese, Russian, Turkish, Vietnamese, Italian, Dutch, Catalan, Arabic, Greek, Farsi, Filipino, and Ukrainian.

I have been pleasantly surprised by how good the free software is. I may take a month later this year and see about combining it with some free software I've seen that guesses about punctuation and paragraph breaks just to see what happens.

So, 17 languages out of, what, 400+ living languages currently? Consider me underwhelmed. (How did they miss Japanese???)

It's one piece of free software.

But are the others more inclusive?

Google's cloud speech-to-text service covers 125. Of course, "cloud" implies a bunch of limitations that exclude more people than the missing languages will.

Others in the existing universe of speech-cognition software, free or otherwise? Yes. (That's not even something I can claim to "know" in a specific way or would bother to look into. But, still, yes.)

takes a huge amount to work to build a decent speech interpreter for a given language. companies aren't going to bother going after small languages.

"The vodka is good, but the meat is rotten"

And we'll have fun fun fun 'til Alexa formats the Tesla's hard drive.

I worked for a few years in a company whose bread and butter was vox recognition. It’s very very good, but sometimes very very good is not enough.

to expand on this a bit...

one place I worked for a couple of years had proprietary vox recognition that they used to field interactive apps on the phone. like a robo-call, except with logical branching based on responses to walk through a scripted interaction.

for its time, it was very very good. I think a lot of newer stuff is better - Polly, Alexa, etc. but it was and still is pretty good.

part of the workflow for that stuff was having exit points, where if the software just couldn't figure out what you were saying, it would exit gracefully and route a recording of the call to humans for review. because humans hear humans better than machines do, in general.

this was all for health care stuff, so we needed to make sure no critical stuff fell through the cracks - hence, the human review for the edge cases. but it didn't have to be in hard real time.

driving a car is hard real time.

even Alexa gets confused, now and then.

They want to make car interiors more like the bridge of the USS Enterprise? I mean, I love starships, but...

Those modern houses that are so popular, the big square things with plain exteriors or paneled siding, and flat roofs. I hate them.

I understand they're wonderful inside, energy efficient, yadda yadda, but the outside just doesn't look like a house, a home, to me. At best, it looks like a little office building. At worst, it looks like an institution of some kind - there's a very modern mansion in a very ritzy neighborhood here in Seattle that I swear looks like a prison or a mental hospital.

Maybe I just want the inside of my car to look like a car, not a starship.

Maybe I just want the inside of my car to look like a car, not a starship.

Or a fighter jet?

"Though most car interiors stick to these relatively straightforward standards, there are a few of them out there that have taken interior design in a very distinct direction to give the driver a "cockpit" feel, in other words making the driver feel as though he or she is sitting at the controls of an F-22 fighter jet, rather than a used sports coupe that makes them look rich. These types of interiors usually have some variety of gauge cluster screen(s), a dashboard angled towards the driver, and at least a few buttons in places you might not expect to find them. "
15 Car Interiors That Make You Feel Like You’re In A Fighter Jet: While most cars share the same basic interior layout, a few have all the controls angled towards the driver, making them feel like a fighter jet. (McLaren F1)

driving a car is hard real time.

yeah. i didn't cough up the extra $10K for the self-annihilatingdriving package.

programming is hard enough when it's just drawing widgets on a web page. i'm not about to let someone's code drive me around.

well fuuuuudge

(but he didn't say fudge)

I've noticed something. In addition to in-state tourist destinations, my TV is being deluged with ads for Las Vegas and for New York City. But not, intetestingly, for Florida. Guess Gov DeSantis doesn't want to get any extra business while he's busy driving the cruise industry to relocate elsewhere.

some days I miss my Chevy panel van with three on the column.

my motto these days is "OK boomer".

One very important innovation in the traffic law enforcement is the license olate recognition technology. This makes it much easier for the police to find people driving cars with expired insurance or inspection.

For example, my own country, Finland, requires an annual vehicle tax payment, a vehicle insurance and yearly inspection. If your car is lacking any one of those, the cameras installed in the police cars driving on the road will recognise this, and unless the patrol has more urgent things to do, you will be stopped and fined. The same happens if you are a fugitive and driving your own car.

The actual consequences for those minor offences are mild: a small fine and the payment of tax or insurance with penal increase, but the likelihood of getting caught is really high.

After WW2 Willy Messerschmitt (probably the best-known German designer of fighter airplanes) went into car design and it shows at least from the outside. Not much concerning the dashboard though.

What I'd like in my car:

- manual control of essential things: steering, accelerator, brake, gears, clutch if appropriate, indicators.
- computer control of things it can work out for itself: engine management, anti-lock braking, windscreen wipers, headlights
- voice override of the things the computer controls, where appropriate
- a heads-up display of useful information: speed, navigation...
- voice control of what's shown on the heads-up display
- voice control of amenities: music, air conditioning
- computer monitoring of whether I'm doing something stupid, with either computer override, if the monitoring is reliable enough, or audible and visual warnings

why wife's 2-year old Acura has most of that, ProBono.

auto wipers and headlights have been somewhat standard for a while. HUDs are becoming widespread. voice control, check. lane departure warning (beeps) and lane departure mitigation (it will keep steer to keep up in your lane automatically unless you fight it), auto-braking (to prevent rear-enders), auto gearbox which you can override, adaptive cruise control where it senses traffic ahead of you and adjusts speed accordingly, etc..

there are a few cars that can auto-park themselves, too.

there are a few cars that can auto-park themselves, too.

So I'll be able to complain about how *kids these days* don't know how to parallel park the same way I do about how they can't make change in their heads?

(Back when I still used paper money, I was paying for something or other that came to $10 and change. Having no tens, I handed the young cashier a twenty and a one so I could get a ten and whatever coinage back instead of $9 and change. The young cashier was completely befuddled. I had to explain it twice, the second time saying, "Just enter $21 as payment in the register and see what comes up as the change due back. It'll make sense to you then.")

it will keep steer to keep up in your lane

parse that, AI muthafukka!

I, for one, look forward to our brave new world of feral AI cars roaming the city in search of parking spaces and cheap gas.

I can haz hunting license to keep the population in check, plz?

lane departure mitigation (it will keep steer to keep up in your lane automatically unless you fight it),

Just curious. Does it connect to the turn indicators? So it fights you if you don't signal a lane change (or even a turn), but not if the blinker is on.

That would dramatically increase the number of people actually signaling their lane changes.

Just curious. Does it connect to the turn indicators? So it fights you if you don't signal a lane change (or even a turn), but not if the blinker is on

right.

if you signal, it doesn't fight you. if you don't it puts a bit of resistance in the steering wheel and beeps.

the problem is when lanes split off the main road (ex. exit ramps). it will try to follow the exit ramp because that's where the solid line on the right is going, and then when you try to steer back, it complains because that would mean crossing the dotted line that now marks the right side of lane you.

good times.

it is handy when you're in the left lane, and don't have to worry about exit lanes, though.

...the right side of the lane you want to be in.

err. i'm uh, apparently, um, declining.

Going back to the start of the thread, the Oxford Nanopore tech which I noted has just been used to sequence the onion genome.

A prosaic endeavour described rather poetically:
https://www.wur.nl/en/Research-Results/Research-Institutes/plant-research/show-wpr/Onion-genome-finally-reveals-its-secrets.htm
...“Assembling the onion genome is comparable to completing a (jigsaw) puzzle with 100,000 pieces, of which 95,000 are just bits of blue sky,” says Finkers. “Only 5000 pieces really make the difference"...

Didn't help in that there was always another layer...

a (jigsaw) puzzle with 100,000 pieces, of which 95,000 are just bits of blue sky

Except it keeps turning out that those "bits of blue sky" actually (if looked at correctly) reveal critical bits about how the 5,000 pieces that "really make the difference" are able to work at all. So called "junk DNA" appears to be far more important than originally assumed.

Ghost in the machine.

The ingredients list is easy to decipher compared to deciphering the recipe.

Nature throws stuff against the wall until it finds something good enough. Then it doesn't bother with it anymore until such time it's no longer good enough.

There's research being done to develop tomatoes that have capsaicin. One advantage besides having literally hot tomatoes is that tomato plants are, if I remember correctly, 5-6 times more productive than are pepper plants.

my wife's company is figuring out how to get common watermelons to produce a sweetening agent that is currently only naturally produced in monkfruit - a little 3" thing that only grows in SE Asia.

In a similar vein...having your meat and eating it too.

"Future Meat Technologies announced last week that it can now produce 1,100 pounds of meat daily from animal cells grown in industrial-scale bioreactors at its facilities in Israel. The company is scouting several locations in the U.S to build large-scale plants to grow cultivated chicken, lamb, pork, and beef. It aims to get its cultivated meats into U.S. grocery stores in 2022. Sadly, this timeline may be too optimistic since getting lab-grown meats onto your plate will require approval from two notoriously sluggish federal regulatory agencies, the U.S. Department of Agriculture and the Food and Drug Administration."
Manufactured Meats Coming to a Grocery Store Near You Next Year?: The process uses 99 percent less land and 96 percent less freshwater than traditional meat production.

it can now produce 1,100 pounds of meat daily from animal cells grown in industrial-scale bioreactors

The critical question is: How does it taste? Well, that and texture. Just being "meat" isn't going to get people buying it. (Although I suppose they could sell it to McDonald's and Burger King -- their "meat" is tasteless anyway.)

The critical question is: How does it taste? Well, that and texture.

Have you had any of the Beyond/Impossible products? My understanding is that Future Meat Technologies are doing something similar, using vegetable protein for texture, but combining it with cultured animal protein cells to dial in the flavor.

I mean, your average factory burger patty is made up of enough different cows to not be all that different.

… Except it keeps turning out that those "bits of blue sky" actually (if looked at correctly) reveal critical bits about how the 5,000 pieces that "really make the difference" are able to work at all. …

Absolutely.
The problem until very recently, though, is that sequencing has only been able to ‘read’ pretty short stretches of DNA, and the bits then need to be assembled through statistical analysis.
That simply wasn’t possible for huge and relatively featureless genomes like that of the onion.
The Nanopore tech enables far longer individual reads, with pretty good fidelity.
It also reveals a lot of very small features which statistical methods discard when reassembling short reads.

DNA analysis is pretty clearly going to be at the heart of "the next big thing." Not just biotech, but something based around DNA analysis and understanding. By the end of this decade, I expect to see something analogous to the dot com boom.

DNA analysis is pretty clearly going to be at the heart of "the next big thing."

Beyond that is being able to reverse engineer and redesign an organism from scratch. But that may be decades away.

Crossing the streams, perhaps, but...:

https://futurism.com/neoscope/lab-leak-hall-mirrors

Epidemiologists like Wertheim — who’s made a career of tracking down how viruses like HIV emerge and spread during outbreaks — and other experts have been tasked with unraveling an endless list of possible timelines that range from unhinged conspiracy theories to plausible-yet-unlikely lab accidents.

“I guess a philosophical problem with the lab leak hypothesis is that it is unfalsifiable,” Wertheim said. “And what I mean by that is the specific scenario where a virus was isolated from an animal brought into the lab and then infected a technician or was accidentally released by somebody working there — there is no virus that we can find in nature that can disprove that scenario.”

Probably do things like "gene cleaning" long before building from scratch. (Denounced as "designer babies", no doubt.)

Beyond that is being able to reverse engineer and redesign an organism from scratch. But that may be decades away.

Take a look at (for example) Ginko Biotech.
It's already starting to happen for simple organisms.

They got a billion dollar loan from the US government to help Pfizer/Moderna industrialise their processes for mRNA vaccine production.
https://www.reuters.com/article/us-health-coronavirus-loan-ginkgo-idUSKBN2851TA

Probably do things like "gene cleaning" long before building from scratch. (Denounced as "designer babies", no doubt.)

if we eventually figure out how to safely edit-out congenital ailments, we'd be fools not to do it. (there will be lots of unsafe attempts first)

if we eventually figure out how to safely edit-out congenital ailments, we'd be fools not to do it

Similarly, having figured out how to safely vaccinate against childhood diseases, we'd be fools not to do it. And yet, and yet.

eventually we're going to find a bug or two that will solve the anti-vaxxer problem once and for all.

covid-19 looks to be providing a step in that direction. Currently virtually all of the fatal cases reported are for people who have not been vaccinated (for whatever reason).

Evolution in action.

To put it in a Rumsfeldian way: do they spawn more offspring than is offed by the disease? Do they outbreed the pandemic and manage to infect their kids with their antivaxx sentiments before they die of it (to the end denying it to be the cause)?
That's why a precondition for a Darwin Award is not having produced offspring.

That's why a precondition for a Darwin Award is not having produced offspring

Make that "...not having produced offspring who agree with your delusions."

I think Hartmut's right, because the crazy can skip a generation. To qualify, their genes have to be ended, taken out of the pool.

However,IRL, I have read reports of teenage children of extreme anti-vaxxers secretly going against the indoctrination and getting the vaccination. I don't know numbers, and am on phone so can't find link, but it's rather encouraging to hear.

The pioneers of atheism often came from over-religious households. They knew what they were rebelling against. That's a (small) glimmer of hope that the plans of fundamentalists to just outbreed us will backfire against them by creating a young generation filled with disgust for their parents' bigotry (more so than kids that grow up in a 'neutral' family).

Still idiots will outbreed those with common sense (as opposed to the 'common sense' among the leaders of the idiots).

the plans of fundamentalists to just outbreed us

Is there any data on whether fundamentalists** actually are having more kids?

** Well, other than Mormons, of course.

And (in Israel, where it matters) ultra orthodox haredi Jews.

It's at least a declared goal
https://en.wikipedia.org/wiki/Quiverfull

Quick googling yields results like this
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2723861/
(not very up to date, based on data nearly 20 years old)

A bit newer
https://www.pewforum.org/2015/05/12/chapter-3-demographic-profiles-of-religious-groups/

And the WaPo from 2015
https://www.washingtonpost.com/news/wonk/wp/2015/05/12/charted-the-religions-that-make-the-most-babies/

Looks like evanglicals are slightly above self-sustaining while most other groups are below that (iirc self-sustaining is about 2.1 per woman).
So, in theory they could outbreed the non-Mormons but all of this assumes that all the kids stay in the fold and don't become renegades (i.e. becoming mainline Christians or atheists).
Definitely far below the planned production targets.

all of this assumes that all the kids stay in the fold and don't become renegades (i.e. becoming mainline Christians or atheists).

Well, assuming that and that their kids don't die of something that they could have been vaccinated for. Before they, in turn, can reproduce.

The comments to this entry are closed.

Blog powered by Typepad