I'm down with that, as long as you flip things up so everything is completely whitelisted and unregulated and the FDA greylists things found to need controls on manufacturing/production and blacklists known dangerous stuff.
Actually, dangerous stuff will take care of itself. Without FDA evaulation the drug companies lose automatic legal immunity for anything they disclosed to the FDA and those 700 dead people would come with heavy lawsuits for drug makers which make skimping on testing way more expensive than doing the best testing they can voluntarily.
Trump has yet to be accused of anything anywhere near as bad as being caught red handed warrantless wiretapping the entire country and passing a bill to "fix" the abuse that actually just gave it legitimacy.
I know right. Crazy freak, I suppose he's going to try to claim that government clandestine organizations are illegally monitoring our communications and tracking our movements without warrants next! Or let me guess, the CIA has sold drugs to raise slush funds for interfering with other governments.
HAHA Or maybe he'll try to claim some power company would skimp out on lining wells, knowingly contaminating the groundwater with deadly hexavalent chromium and then engage in a conspiracy to "disclose" it to the local residents by telling them it's good for them and paying for bogus health checkups from doctors.
Or maybe next he'll try to claim some oil company would hire nearly every marine biology expert after an oil spill for a short analysis in order to ethically and legally bind them up from independently assessing and commenting on other consequences of that spill and its effects on the gulf.
There are two extremes, on one you have ancient aliens, on the other you have the assumption that corporations and government are generally honest and just going about their day and all this conspiracy crap is just the stuff of movies. Reality is definitely somewhere between the two and frankly as information has become more readily available and shared the evidence suggests that grey mark is actually closer to the crazy ancient alien guy than Joe sucker who trusts authority.
It isn't particularly difficult to get meth in the United States but you can buy pseudo ephedrine with a photo id in every pharmacy with an id and signature and pharmacies are generally 2 to a four block square. There are places where meth dealers might be that dense and as visible as a pharamcy but that is hardly the norm.
Meth works just fine as a nasal decongestant. Better actually. You just need a smaller dose whereas you have to double what it says on the box of sudafed to get any noticeable result. It is also more effective in ADHD treatment than adderall with far more room to safely increase the dose for adult adhd.
There are no shortage of things like this that have fallen victim to the war on drugs. Cocaine in very light doses (natural coca leaf tea vs concentrated powder extract) is much more like mild coffee with far fewer side effects.
"It's kind of crap, but it's better than putting the drug into the "normal drug with scary warning labels like every other drug for this condition" class by flatly approving it for normal dispensation in normal pharmacies."
Yeah, but if you do this you might as well not approve it all. Insurance companies won't pay for what will always in a case like this be a brand new and extremely expensive medication.
The issue is UNDER hiring. The perception that it is difficult to find qualified talent and the needing someone instantly who is an exact fit for a tech fingerprint are all symptoms of what should be overlapping and split duties across dozens of people being collapsed and folded under one "supergeek" over and over and over again... round after round. It should be no big deal if anyone in IT leaves and it takes 2 years to teach a high schooler to replace them.
You can lay off hundreds of thousands of tech workers a year, claiming there is a shortage of "talent" to fill the handful of heavily specialized job you replace them with. But those positions you just invented aren't hard to fill because of lack of experts in the world... they are hard to fill because they didn't exist before you just made them up. This isn't some new field or type of education people are missing, it is technology fingerprint unique the company and the handful of people they collapsed those jobs down to... who naturally don't stick around long. At that point you should either not do that in the first place or start paying them seven figures to stick around because you've boxed yourself in a corner. Worse, you've done it in a way that probably will take years to crash your fortune 500 ship.
Actually hiring correctly with modern technology means armies of tech workers who frankly fill most of their days learning and playing with tech or even watching a movie. You have hot spares and redundant data in your raid arrays... what idiot thought it was a good long term strategy to pull them all and sell them on ebay? That's what has been done across the board with the "hot spares" and redundancy in tech knowledge and the "talent shortage" is nothing more than complaining about the remaining drives not being able to cope as well under the extra wear and tear.
"No more ignorant then thinking things we've deliberately built with a targeted purpose are innately superior."
Good luck supporting the concept that all else being equal something which has been intentionally designed for a purpose isn't innately superior to a random result. All else unknown i'll try a prototype motorcycle to transport to work, without knowing anything else, before a vat of acid.
A particular thing we've tried to make might be inferior to a particular result that came from nature but yes, having been intentionally designed by us to serve a specific purpose is a point toward the probability something is superior for that purpose.
"GMO, including CRISPR is a tool, and like most tools, can be used for good but it is not automatic. History shows that even the best engineers can screw up using their tools, especially when there are bean counters involved. For example, a recent pedestrian bridge collapse in Florida."
I wouldn't disagree with that at all. In fact, I said in my post that unintended consequences do occur. But the probability of an engineered bridge collapsing when you try to walk across it is lower than a tree trunk. Even the use of a tree trunk as an obvious example is essentially a less sophisticated engineering effort... man has tested walking across tree trunks and found them to be a more reliable tool for crossing things than a vine. Nothing we eat was designed by nature to be our food, no natural remedy was made to be such, we use the things we find and make as tools for what amount to artificial and man made goals. Just because we sometimes screw up doesn't mean we should start from the assumption we are wrong when we've given our best effort to intentionally building such a tool vs some random evolved process.
The better question is what tool is Monsanto actually trying to make... is it bigger tomatoes or a tight grip on the food supply via sterile plants. Paranoia about GMOs expresses a great deal of ignorance... people aren't paranoid enough about the control we are letting Monsanto gain over the global food supply and the power it gives them over everything else once they have it.
Fair enough. Although in the noble dandelions defense grass related allergies are also fairly common. I don't have numbers to compare but in their absence I'm calling it a wash.
Generally no, fearing the result would require picking and choosing particular edits to oppose. The arguments fall more along the "all things natural are safe, nature is better than man" lines.
While there can always be unintended consequences the nearly random editing process of nature produces all sorts of things which are deadly to us, thinking the things we've deliberately built with a targeted purpose are innately inferior is ignorance to the extreme.
Patently false. That's like saying only governments can unjustly execute someone. I think you are confusing the matter with the legal restriction preventing government from censoring free speech. That is only one very limited example of censorship.
Just because businesses can legally censor some things or engage in any particular behavior today doesn't mean they have to be allowed to do so tomorrow. Further, there is nothing a business is legally allowed to do which grants them any protection from people not liking that decision, raising awareness, and proposing someone else prop up a competing solution minus that behavior to wipe the floor with them in the free market.
If the behavior is morally reprehensible, like attempting to shove moral or religious ideals down the throats of customers using their service, awareness is raised, and somehow that business isn't losing in the "free" market, the most likely explanation is an illegal monopoly in which case the rules very much change with regard to what a business is permitted to do by us..
Legally Amazon can censor their selections (although with some content served through prime this may be an illegal bait and switch since Amazon does so silently) and I a healthy young man can take the last flu shot when there is a pregnant woman in line behind me but people have every right to call either of us out on this behavior and people have even more right to call out Amazon abusing them since they are paying for the privilege.
"How was an indie author making it before Amazon? And where does it say Amazon has to ensure the business vitality of all writers?"
This has nothing to do with ensuring the business vitality of anyone and everything to with amazon censorship practices... people choose amazon because of convenience and technical capabilities almost nobody voting for Amazon with their dollar realizes that Amazon applies censorship in broad strokes.
They are growing massive and most people using their services don't know anything about their massive and widespread censorship practices. People are choosing them based on convenience and technical capabilities with no idea they are engaging in these practices.
It depends on how to define "country." If you look at the economic and employment status of the individual the US and China are hurt the most by AI. AI provides little threat to manual labor and physical tasks, the robotics of 80's, 90's, and 00's threatens them more. AI will mostly put knowledge workers out of work. If your interface to what you do is a computer, voice, video, network link, etc and/or redundancies like management you are at serious risk. If your work heavily depends on actual physical interaction in non-trivial ways you are safe... at least from AI itself, though it's friend robotics might still come for you.
The US will do better, because overall more money will flow in but the people, the real people, it will be harmful not helpful.
"Generally speaking though, given the security flaws in the previous standards it really is time to force the upgrade."
How exactly do you propose to do that? An upgrade has been forced with every "mode" you just listed and yet magically no upgrades have occurred. After all, there is a reason you know you have the option to adjust and select those modes manually and it isn't because everything magically updates to the latest and greatest TLS.
"If you're still using these 10+ year old ciphers for security, you aren't secure to begin with - your TLS client may as well tell you so outright."
Nobody is using 10+ year old ciphers for security... they are stuck with what someone was using for security 10+ years ago. Your TLS client doesn't tell you so outright, it drops support for them or forces you to turn on lower security settings that compromise more than just the cipher.
Yes, there is more to TLS 1.3 than the cipher suite. There is no reason you can make TLS 1.3 able to fall back and support older ciphers and dial down only the degree necessary to negotiate a resolution. In that way, claiming to support TLS 1.3 also means continuing to be able to support those older ciphersets.
Honestly, even a self-signed cert serves it's purpose if it is has been saved as an exception in the browser of the admin... that point you not only have encryption but personally known identity verification with fingerprint which blows a third party issue out of the water. At least it does if it's using a decent cipher.
TLS 1.3 needs to continue to include the ciphers so there is something in TLS 1.3 that forces the browsers to continue supporting older (incidently weaker) algorithms. It is the responsibility of TLS 1.x to include backward compatibility for TLS 1.x-1 otherwise it stops becoming practical to support TLS as the standard at all. Previous versions have cost millions of dollars because of this nonsense and resulted in thousands of old systems running IE 6 with poorly chosen settings so they can interact with the hardware that doesn't get caught or have to be replaced on the latest PCI audit or that got tossed on the exception list if someone was bright enough to realize that you notate pretty much anything onerous to implement on an exception list and still be PCI compliant.
"What you have to understand is that there is a good chance that an enemy is on your LAN."
Highly improbable. The enemy might be on the users lan but it's very doubtful they are anywhere useful. In most all cases detection software will find all sorts of threats... somehow magically none of these threats resulted in a compromise before the system was added and that is because these attacks are dumb poorly tested and automated bots that don't work with a damn in practice. They work by hamfisted attacks on thousands of thousands of systems to slowly pick up the ones they get lucky on. Admins following commonly employed admin best practices won't be vulnerable to most attacks even if the software on their hosts is. For instance, an apache vulnerability on a web server sounds critical and will require patching fast... unless of course your web servers don't host any content and are several layers of caching and proxy server behind where someone who doesn't already have root could access.
You are aware this is the situation with about 80% of ILO/IPMI/Lights out, etc interfaces on the management networks of essentially every fortune 500 company, ESPECIALLY the ones selling security solutions (maybe not especially but certainly including).
There is nobody upgrading a critical system, anywhere, because the ILO management interface can't run the latest cipher suite. There is also nobody upgrading such an interface after racking the thing even if one is available. Further these are all running self-signed certs in the first place. Yet another ignorant practice. The possibility a MITM attack can happen doesn't mean there is no benefit to encrypting the content on the wire. There is a dramatically higher probability that someone is sniffing the white than has managed to successfully employ a man in the middle attack against an admin. Especially when that admin will have all those self-signed certs cached.
The enemy is not likely inside your lan as suggested in another one of these sub-threads. Hell the enemy is most likely nowhere except some malware downloaded by an end user. A high profile compromise here and there doesn't change that 99.9999% of "threats" detected are utter bullshit and with proper firewalls, zoning, and minimal lazy security in place the same number of vulnerabilities detected on your platform are practically exploitable on the systems as deployed whether present in the underlying software or not.
No, it won't because everyone is going to start requiring the latest and greatest as part of their compliance policy and you'll force people to enable the lowest common denominator and lose the other benefits of the newer protocol as well. Hell, the fallback probably won't even work.
Thanks to forced obsolescence in these standards you have thousands of systems running IE 6 with security settings cranked down because it is the only way for people to do their job and interact with older systems that can't or won't be upgraded until their actual operational function can't be performed. Continuing this poor practice extends this massive risk and probably expands it.
Sounds like the argument made with TLS 1.2... which proved to be incorrect and now thousands of enterprises have crappy old browsers that still allow them to turn on "insecure" old settings for all kinds of old hardware they have to use and administer with embedded or unsupported appliances running old webservers and standards. People don't replace things because the oob interface uses a crappy version old HTTPS to load it's self-signed cert driven interface and there are systems like this in EVERY fortune 500 company.
By the time the systems you are forcing to "not have garbage default settings" become an issue, the "not garbage" settings will be old and garbage settings already. They won't use the crappy old settings out of the box on mainstream servers and browsers regardless and clueless admins don't even know to change anything less alone reduce the security.
"However fast you get at ray tracing, you can instead use that power for rasterization and do far far more."
Not when you are real time rendering the graphics to be integrated on the fly along side real time light vectors flying at the retina. Think pokemon go, without the phone, or the screen, and with the pokemon actually sitting on your counter with it's legs hanging down.
Slashdot Mirror
I'm down with that, as long as you flip things up so everything is completely whitelisted and unregulated and the FDA greylists things found to need controls on manufacturing/production and blacklists known dangerous stuff.
Actually, dangerous stuff will take care of itself. Without FDA evaulation the drug companies lose automatic legal immunity for anything they disclosed to the FDA and those 700 dead people would come with heavy lawsuits for drug makers which make skimping on testing way more expensive than doing the best testing they can voluntarily.
Trump has yet to be accused of anything anywhere near as bad as being caught red handed warrantless wiretapping the entire country and passing a bill to "fix" the abuse that actually just gave it legitimacy.
I know right. Crazy freak, I suppose he's going to try to claim that government clandestine organizations are illegally monitoring our communications and tracking our movements without warrants next! Or let me guess, the CIA has sold drugs to raise slush funds for interfering with other governments.
HAHA Or maybe he'll try to claim some power company would skimp out on lining wells, knowingly contaminating the groundwater with deadly hexavalent chromium and then engage in a conspiracy to "disclose" it to the local residents by telling them it's good for them and paying for bogus health checkups from doctors.
Or maybe next he'll try to claim some oil company would hire nearly every marine biology expert after an oil spill for a short analysis in order to ethically and legally bind them up from independently assessing and commenting on other consequences of that spill and its effects on the gulf.
There are two extremes, on one you have ancient aliens, on the other you have the assumption that corporations and government are generally honest and just going about their day and all this conspiracy crap is just the stuff of movies. Reality is definitely somewhere between the two and frankly as information has become more readily available and shared the evidence suggests that grey mark is actually closer to the crazy ancient alien guy than Joe sucker who trusts authority.
It isn't particularly difficult to get meth in the United States but you can buy pseudo ephedrine with a photo id in every pharmacy with an id and signature and pharmacies are generally 2 to a four block square. There are places where meth dealers might be that dense and as visible as a pharamcy but that is hardly the norm.
Meth works just fine as a nasal decongestant. Better actually. You just need a smaller dose whereas you have to double what it says on the box of sudafed to get any noticeable result. It is also more effective in ADHD treatment than adderall with far more room to safely increase the dose for adult adhd.
There are no shortage of things like this that have fallen victim to the war on drugs. Cocaine in very light doses (natural coca leaf tea vs concentrated powder extract) is much more like mild coffee with far fewer side effects.
Why is this marked flamebait? This is too accurate and runs really counter to your politics?
"It's kind of crap, but it's better than putting the drug into the "normal drug with scary warning labels like every other drug for this condition" class by flatly approving it for normal dispensation in normal pharmacies."
Yeah, but if you do this you might as well not approve it all. Insurance companies won't pay for what will always in a case like this be a brand new and extremely expensive medication.
The issue is UNDER hiring. The perception that it is difficult to find qualified talent and the needing someone instantly who is an exact fit for a tech fingerprint are all symptoms of what should be overlapping and split duties across dozens of people being collapsed and folded under one "supergeek" over and over and over again... round after round. It should be no big deal if anyone in IT leaves and it takes 2 years to teach a high schooler to replace them.
You can lay off hundreds of thousands of tech workers a year, claiming there is a shortage of "talent" to fill the handful of heavily specialized job you replace them with. But those positions you just invented aren't hard to fill because of lack of experts in the world... they are hard to fill because they didn't exist before you just made them up. This isn't some new field or type of education people are missing, it is technology fingerprint unique the company and the handful of people they collapsed those jobs down to... who naturally don't stick around long. At that point you should either not do that in the first place or start paying them seven figures to stick around because you've boxed yourself in a corner. Worse, you've done it in a way that probably will take years to crash your fortune 500 ship.
Actually hiring correctly with modern technology means armies of tech workers who frankly fill most of their days learning and playing with tech or even watching a movie. You have hot spares and redundant data in your raid arrays... what idiot thought it was a good long term strategy to pull them all and sell them on ebay? That's what has been done across the board with the "hot spares" and redundancy in tech knowledge and the "talent shortage" is nothing more than complaining about the remaining drives not being able to cope as well under the extra wear and tear.
"No more ignorant then thinking things we've deliberately built with a targeted purpose are innately superior."
Good luck supporting the concept that all else being equal something which has been intentionally designed for a purpose isn't innately superior to a random result. All else unknown i'll try a prototype motorcycle to transport to work, without knowing anything else, before a vat of acid.
A particular thing we've tried to make might be inferior to a particular result that came from nature but yes, having been intentionally designed by us to serve a specific purpose is a point toward the probability something is superior for that purpose.
"GMO, including CRISPR is a tool, and like most tools, can be used for good but it is not automatic. History shows that even the best engineers can screw up using their tools, especially when there are bean counters involved. For example, a recent pedestrian bridge collapse in Florida."
I wouldn't disagree with that at all. In fact, I said in my post that unintended consequences do occur. But the probability of an engineered bridge collapsing when you try to walk across it is lower than a tree trunk. Even the use of a tree trunk as an obvious example is essentially a less sophisticated engineering effort... man has tested walking across tree trunks and found them to be a more reliable tool for crossing things than a vine. Nothing we eat was designed by nature to be our food, no natural remedy was made to be such, we use the things we find and make as tools for what amount to artificial and man made goals. Just because we sometimes screw up doesn't mean we should start from the assumption we are wrong when we've given our best effort to intentionally building such a tool vs some random evolved process.
The better question is what tool is Monsanto actually trying to make... is it bigger tomatoes or a tight grip on the food supply via sterile plants. Paranoia about GMOs expresses a great deal of ignorance... people aren't paranoid enough about the control we are letting Monsanto gain over the global food supply and the power it gives them over everything else once they have it.
Fair enough. Although in the noble dandelions defense grass related allergies are also fairly common. I don't have numbers to compare but in their absence I'm calling it a wash.
Generally no, fearing the result would require picking and choosing particular edits to oppose. The arguments fall more along the "all things natural are safe, nature is better than man" lines.
While there can always be unintended consequences the nearly random editing process of nature produces all sorts of things which are deadly to us, thinking the things we've deliberately built with a targeted purpose are innately inferior is ignorance to the extreme.
Dandelions are harmless, edible, and more attractive than grass... Grass is definitely the weed in this comparison.
"only governments censor."
Patently false. That's like saying only governments can unjustly execute someone. I think you are confusing the matter with the legal restriction preventing government from censoring free speech. That is only one very limited example of censorship.
Just because businesses can legally censor some things or engage in any particular behavior today doesn't mean they have to be allowed to do so tomorrow. Further, there is nothing a business is legally allowed to do which grants them any protection from people not liking that decision, raising awareness, and proposing someone else prop up a competing solution minus that behavior to wipe the floor with them in the free market.
If the behavior is morally reprehensible, like attempting to shove moral or religious ideals down the throats of customers using their service, awareness is raised, and somehow that business isn't losing in the "free" market, the most likely explanation is an illegal monopoly in which case the rules very much change with regard to what a business is permitted to do by us..
Legally Amazon can censor their selections (although with some content served through prime this may be an illegal bait and switch since Amazon does so silently) and I a healthy young man can take the last flu shot when there is a pregnant woman in line behind me but people have every right to call either of us out on this behavior and people have even more right to call out Amazon abusing them since they are paying for the privilege.
"How was an indie author making it before Amazon? And where does it say Amazon has to ensure the business vitality of all writers?"
This has nothing to do with ensuring the business vitality of anyone and everything to with amazon censorship practices... people choose amazon because of convenience and technical capabilities almost nobody voting for Amazon with their dollar realizes that Amazon applies censorship in broad strokes.
They are growing massive and most people using their services don't know anything about their massive and widespread censorship practices. People are choosing them based on convenience and technical capabilities with no idea they are engaging in these practices.
It depends on how to define "country." If you look at the economic and employment status of the individual the US and China are hurt the most by AI. AI provides little threat to manual labor and physical tasks, the robotics of 80's, 90's, and 00's threatens them more. AI will mostly put knowledge workers out of work. If your interface to what you do is a computer, voice, video, network link, etc and/or redundancies like management you are at serious risk. If your work heavily depends on actual physical interaction in non-trivial ways you are safe... at least from AI itself, though it's friend robotics might still come for you.
The US will do better, because overall more money will flow in but the people, the real people, it will be harmful not helpful.
"Generally speaking though, given the security flaws in the previous standards it really is time to force the upgrade."
How exactly do you propose to do that? An upgrade has been forced with every "mode" you just listed and yet magically no upgrades have occurred. After all, there is a reason you know you have the option to adjust and select those modes manually and it isn't because everything magically updates to the latest and greatest TLS.
"If you're still using these 10+ year old ciphers for security, you aren't secure to begin with - your TLS client may as well tell you so outright."
Nobody is using 10+ year old ciphers for security... they are stuck with what someone was using for security 10+ years ago. Your TLS client doesn't tell you so outright, it drops support for them or forces you to turn on lower security settings that compromise more than just the cipher.
Yes, there is more to TLS 1.3 than the cipher suite. There is no reason you can make TLS 1.3 able to fall back and support older ciphers and dial down only the degree necessary to negotiate a resolution. In that way, claiming to support TLS 1.3 also means continuing to be able to support those older ciphersets.
Honestly, even a self-signed cert serves it's purpose if it is has been saved as an exception in the browser of the admin... that point you not only have encryption but personally known identity verification with fingerprint which blows a third party issue out of the water. At least it does if it's using a decent cipher.
TLS 1.3 needs to continue to include the ciphers so there is something in TLS 1.3 that forces the browsers to continue supporting older (incidently weaker) algorithms. It is the responsibility of TLS 1.x to include backward compatibility for TLS 1.x-1 otherwise it stops becoming practical to support TLS as the standard at all. Previous versions have cost millions of dollars because of this nonsense and resulted in thousands of old systems running IE 6 with poorly chosen settings so they can interact with the hardware that doesn't get caught or have to be replaced on the latest PCI audit or that got tossed on the exception list if someone was bright enough to realize that you notate pretty much anything onerous to implement on an exception list and still be PCI compliant.
"What you have to understand is that there is a good chance that an enemy is on your LAN."
Highly improbable. The enemy might be on the users lan but it's very doubtful they are anywhere useful. In most all cases detection software will find all sorts of threats... somehow magically none of these threats resulted in a compromise before the system was added and that is because these attacks are dumb poorly tested and automated bots that don't work with a damn in practice. They work by hamfisted attacks on thousands of thousands of systems to slowly pick up the ones they get lucky on. Admins following commonly employed admin best practices won't be vulnerable to most attacks even if the software on their hosts is. For instance, an apache vulnerability on a web server sounds critical and will require patching fast... unless of course your web servers don't host any content and are several layers of caching and proxy server behind where someone who doesn't already have root could access.
You are aware this is the situation with about 80% of ILO/IPMI/Lights out, etc interfaces on the management networks of essentially every fortune 500 company, ESPECIALLY the ones selling security solutions (maybe not especially but certainly including).
There is nobody upgrading a critical system, anywhere, because the ILO management interface can't run the latest cipher suite. There is also nobody upgrading such an interface after racking the thing even if one is available. Further these are all running self-signed certs in the first place. Yet another ignorant practice. The possibility a MITM attack can happen doesn't mean there is no benefit to encrypting the content on the wire. There is a dramatically higher probability that someone is sniffing the white than has managed to successfully employ a man in the middle attack against an admin. Especially when that admin will have all those self-signed certs cached.
The enemy is not likely inside your lan as suggested in another one of these sub-threads. Hell the enemy is most likely nowhere except some malware downloaded by an end user. A high profile compromise here and there doesn't change that 99.9999% of "threats" detected are utter bullshit and with proper firewalls, zoning, and minimal lazy security in place the same number of vulnerabilities detected on your platform are practically exploitable on the systems as deployed whether present in the underlying software or not.
No, it won't because everyone is going to start requiring the latest and greatest as part of their compliance policy and you'll force people to enable the lowest common denominator and lose the other benefits of the newer protocol as well. Hell, the fallback probably won't even work.
Thanks to forced obsolescence in these standards you have thousands of systems running IE 6 with security settings cranked down because it is the only way for people to do their job and interact with older systems that can't or won't be upgraded until their actual operational function can't be performed. Continuing this poor practice extends this massive risk and probably expands it.
Sounds like the argument made with TLS 1.2... which proved to be incorrect and now thousands of enterprises have crappy old browsers that still allow them to turn on "insecure" old settings for all kinds of old hardware they have to use and administer with embedded or unsupported appliances running old webservers and standards. People don't replace things because the oob interface uses a crappy version old HTTPS to load it's self-signed cert driven interface and there are systems like this in EVERY fortune 500 company.
By the time the systems you are forcing to "not have garbage default settings" become an issue, the "not garbage" settings will be old and garbage settings already. They won't use the crappy old settings out of the box on mainstream servers and browsers regardless and clueless admins don't even know to change anything less alone reduce the security.
"However fast you get at ray tracing, you can instead use that power for rasterization and do far far more."
Not when you are real time rendering the graphics to be integrated on the fly along side real time light vectors flying at the retina. Think pokemon go, without the phone, or the screen, and with the pokemon actually sitting on your counter with it's legs hanging down.