This is a stupid and naive point of view. Law enforcement will never eliminate black markets, so we need practical ways to address loss.
The assets controlled by that key have a market value. Theft of the key easily translates to theft of the assets. You can recover the value of lost assets either from the thief or from a party who was responsible for securing them. This is why most parking garages explicitly disclaim responsibility on the tickets---they do not want to be legally responsible for securing your vehicle and its contents.
The question here is whether TMobile is legally responsible for ensuring the integrity of his account and SMS communications. I'm not a lawyer so I'm not going to guess at the outcome of a trial. If I were TMobile, I'd probably just pay the ~$20K to avoid court and bad publicity.
They could make the Pro version free for personal use. Or reduced cost. A lot of developers/companies only charge businesses and government for their products.
If they decide to charge for personal use, then we'll have to look at the software to see if the free version is worthwhile on its own merits.
I don't begrudge them a revenue stream. If they're doing this more as a job than a hobby, good for them.
A really programmer will have to learn all of the complex "geeky" concepts---eventually.
A good starter language should abstract away a lot of those things. It is very difficult and potentially discouraging to tackle flow of control, obscure IO interfaces, and UI quirks all at once.
I don't know if this is marketing fluff (never personally written Swift code), but a good starter language that runs on a popular platform should be a basic elective course for middle/high schools.
Out of curiosity, would anyone have suggestions for the best starter languages for Windows and Linux systems? Preferably something with native libraries for implementing a GUI?
Checkmarx found that they still leaked enough information to tell encrypted commands apart, allowing a hacker on the same network to watch every swipe left, swipe right, or match on the target's phone nearly as easily as if they were looking over the target's shoulder.
When different user actions result in widely different application behavior, it will always be easy to infer the user action. E.g., if matching is the only action that does not result in a new profile being presented, then observation of the smaller data exchange will lead to that inference.
The only way to avoid this is to make the network traffic identical for all cases, which is extremely wasteful of bandwidth and, presumably, battery life.
That said, encryption of all data should be standard now. There is some overhead, but it's not the 1990s---crypto is not that burdensome.
We have had proxy activities for competition, aggression, and violence since the dawn of history. Everything from boxing, rugby, and polo to swimming and track to Go and chess.
Video games are just a new spin on an ancient habit. We substitute relatively harmless activities as outlets for our less-than-friendly instincts. In this respect, we have reached a new zenith with the variety, ubiquity, and flexibility of computer games. Participation in previous sports or hobbies has never been as safe and widely appealing as video games.
I'm not surprised there is no link between video games and violent behavior; the games themselves are the outlet for the urges that lead to violence.
Disney owns so many franchises now that a Netflix without Disney will kill it.
Not really. Netflix's original content ranges from decent to excellent.
They could probably coexist with Disney and a few other streaming content producers. E.g., Netflix + HBO Go + a hypothetical Disney service would total up to less than the average cable bill---unless Disney tries to charge way more than everyone else. That is already more content than most people need. Plus, there is still broadcast TV for news, some sports, and other local content.
Netflix doesn't need to be everything to everyone; it only needs to justify its ~$10/month price tag. More streaming services may actually help, insofar as they encourage cord-cutting, and many cord-cutters will pick up Netflix plus one or two other services.
You mean the warranty would be enforceable even using 3rd party parts for maintenance?
Third-party parts are usually associated with aftermarket mods, but it works the same for standard maintenance and consumable parts (oil filters, tires, etc).
In these cases, the third-party parts aren't covered by the OEM warranty, but the original parts remain covered. They cannot void your warranty for using third-party parts in the US; the warranty continues to cover the OEM parts regardless.
There is a difference between positivity and the foolish optimism born from ignorance. OP was right in saying that general-purpose "molecular printing" is very, very far from being available.
This isn't even a prototype of such a device---it was a one-off proof of concept with a specific target. If you can't even criticize the reasons for his claim, then maybe you should keep your mouth shut about his opinion.
If they come to fruition, the first generation of molecular printers will likely have a limited use case, much like 3D printers. The headline is misleading at best, and that's not even accounting for the patent and regulatory issues related to drug production.
Researchers have done work in turning the immune system against cancer, and it is remarkably effective. The FDA has approved two CAR-T therapies to date.
CRISPR is just an easier means of tinkering with cellular DNA. This makes it likely that we can develop more effective treatments for more types of cancer.
Is Intel developing new chips that don't have this problem?
Of course, but the lead time on CPU development is long. It will be at least a year or two before consumers can buy anything with builtin resistance to Spectre and Meltdown.
Are they going to be slower, too?
The generational improvements will probably offset the losses, so I strongly doubt that will be the case.
Plus, fixed CPUs will not require KPTI anymore, so they will be secure with the "normal" OS-level performance optimizations. Windows and Linux can go back to doing things the pre-Meltdown way if Intel fully addresses their problems.
If humans are no better than a machine then how can a machine programmed by humans do any better
This is a stupid, Luddite sentiment. Computers are already far better than humans at a wide range of tasks. This increases dramatically if your usage of "machine" include robots performing physical tasks.
Deep learning is opening the doors to new machine skills. Quantum computing will likely open a few more in the next few decades. It's anyone's guess what comes after that.
Humans were smart enough to make a machine that could do some things better than they could. We used those new capabilities to make better machines---again and again. The only thing that stops this process is the possibility that a final generation of machines will be incapable of developing meaningful advances for a new generation.
It suggests that the additional variables are either unrelated to the outcome
Or their application was slapped together haphazardly in order to start bringing in those contract dollars as soon as possible. The other 135 variables could provide useful information, if their analysis of that information were better. Until this study, there was no incentive to perform well because the cost of poor performance is borne by convicts.
I am far more willing to believe that crapware is being shoveled into government computers at taxpayer expense.
Bear in mind that this system performs as well as untrained humans. If humans with correctional or psychological expertise perform better than untrained humans (which I assume is true), then this system is embarrassingly bad. It's doing a worse job than the people it's supposed to help.
US national security in the IT space is shot to hell, and most of it is the NSA's fault, with Intel a close second.
Way too short-sighted there.
1. Heartbleed was remotely exploitable. The Spectre vulnerabilities require local code or else a second exploit to allow remote execution. Heartbleed posed a far greater risk to internet as a whole.
2. The NSA didn't create any of the vulnerabilities they exploited. The fault lies with the respective developers such as Cisco, Microsoft, etc. The NSA essentially provided a full disclosure with POC for those vulnerabilities, but all of those problems existed regardless.
3. These zero-days wouldn't be such a huge problem if people did defense-in-depth properly. I.e., at all.
From the often-blamed-on-the-NSA pile:
WannaCry spread by exploiting a vulnerability in SMB. FFS. Who is exposing SMB to the internet in the first place? There are much better choices when it comes to secure file transfer protocols; and there is hardly anything worse.
This was not even the first SMB exploit. Sometimes you can disclaim responsibility by saying "no has ever thought of that before". But in this case, we have not only warned about it---we have seen it firsthand.
This has always been an option, and the industry chose Door #1 years ago because it is by far the most productive and economical.
It's not like the machinery is getting any simpler either. How much data do they send and how often do they need to report? Do they integrate with inventory systems to track usage of raw materials and other consumables? How good are the predictive wear/failure alerts?
More precision and more complex automation are going to push those 1970s-era control systems from outdated to unworkable---if it hasn't happened already.
Granted, all industries are not affected the same way. But you're still fundamentally stuck with buying a mass-produced configurable SCADA supervisor for a reasonable price vs a custom system at an insane price. And the mass-produced system is going to have a bit of that "everything and the kitchen sink" flavor.
So, you installed just the web protection and then disabled the SSL traffic scanner? Even though 50% of traffic is encrypted now?
Financial and health data is among the SSL-encrypted traffic. If you don't want something seeing those things, then you either need to exempt your bank/healthcare sites or disable the SSL scanner entirely. Enterprise proxies usually offer this out of the box---most US organizations will not decrypt traffic to these destinations.
And besides, it's up to him as to whether he wants SSL decryption at all. The feature should be configurable.
Odds are good that you disabling the SSL middleware means the uninstaller didn't realize it was there and didn't uninstall it.
Shit application, shit installer. There is no reason an application cannot keep track of which modules are installed regardless of whether a user disables them.
Meanwhile, Comodo wants to keep resetting Chrome to use Yahoo "for web protection" and repeatedly reinstalls a Yahoo Search extension if you delete it.
They have a deal with Yahoo, and they value Yahoo's money over your express wishes. That should tell you all you need to know about your security vendor.
Antivirus programs are getting to be a bit pointless these days. If you have good security measures, you won't get hit by the kind of crap they can find in the first place.
TLA's are here and are "controlling the narrative" - but failing. We know there's no other reason to call a legit request for "how you know what you claim" as trolling.
I'm not particularly surprised by the lack of public evidence. Classified computer systems aren't going to be passed around for inspection. Do you really expect to see logs or forensic results from a classified system? You're demanding something that will never be forthcoming, and, in fact, may be illegal to release. From day one, the entire Kaspersky investigation was destined to land in "trust us" territory. There is no other way it could play out.
You think the Russians are doing all the badware on earth?
Since the article is about leaked NSA malware, I don't even know why you're asking this question. Major non-sequitur right there.
Peek-a-boo - I see you, paid "intelligence community trolls with mod points"
I see you've been forgetting your meds. Put on an extra layer of tin foil and consult your psychiatrist on the best way to handle missed doses.
If the attack was successful, then it's the American company's fault for not better protecting.
You sound certain of that. But there is no such thing as perfect security. Without technical details, there is no way to distinguish between a brilliant attack and negligence.
If you deal with government secrets
Be careful with that idea...
leaving them out in a bar for anyone to see is considered treasonous
...because this only applies to classified material.
Also, it's not treason. Things don't automatically become treason because the government is especially inconvenienced.
Finally, I doubt that voting machine designs could even be classified in the first place. There are specific types of information which are eligible for classification, and voting machines simply don't fall into any of them. Go read Executive Order 12958 or 13526 if you're inclined to disagree---it doesn't mater which because Section 1.4 is consistent between them.
If I had an algorithm that occasionally misidentified people in a way that can cause public outrage, I would filter the outputs to avoid controversy too.
Wake me up when they release a fixed version. Hell, a paper describing the issue in detail would be interesting---even fascinating, if I were any sort of expert.
Googles themselves admitted that their algorithms still make the same mistake. This article boils down to "hard problem takes longer than 3 year to solve"---with excessive puffery.
At most, Intel will adjust prices downward to compensate for the performance delta.
This single event isn't enough to turn AMD's fortunes around. AMD's financial woes cut far deeper than a single quarter or even an entire year can fix.
A 10% reduction in that is significant, but not catastrophic for most users.
That 10% is also an average, with some loads seeing a greater hit. I imagine the people on the tail end of the distribution are going to be very unhappy. Also, Intel is only considering processors from the last 3 years, whereas Microsoft has stated publicly that older processors see an even greater hit.
Most enterprises replace their equipment on a 2-4 year cycle, depending on the business. For them, 3-year-old processors are either trash or on the chopping block for the next tech refresh. They will grumble, but they won't really care once the frenzy of patching is done. Home users---especially those on a budgets---are more likely to run into the larger performance hits that Intel isn't discussing.
First, there is no such thing as a "non-profit" college. They ALL profit.
This is a fallacy called equivocation.
The standard use of "non-profit" means that there are no dividends or revenues being paid out to investors or owners. Your use of "profit" in the second sentence is clearly intended to mean something else.
Second, Not relying on tests means relying on transcripts.
This is a false dilemma fallacy. There are more choices than tests and transcripts, including: recommendations, extracurricular/volunteer work, professional experience, essays or personal statements, interviews, and project/portfolio submissions.
I had all of those things except a portfolio when I applied to university, although the professional experience was obviously very limited due to age.
excluding any kind of objective or semi-objective measurements leaves only
Another false dilemma: you imply that making tests optional renders them irrelevant. Making standardized tests optional does not eliminate them as a source of information. A test-optional school can still look favorably on applicants with high scores on relevant tests.
What we have here is the gradual degradation of the US higher Education system
The article indicates that Wake Forest, UGA, and De Paul reviewed their records, and they get around a 1% improvement in their predictions of college success from considering standardized test scores. Why would anyone bother with an ineffective tool?
Slashdot Mirror
This is a stupid and naive point of view. Law enforcement will never eliminate black markets, so we need practical ways to address loss.
The assets controlled by that key have a market value. Theft of the key easily translates to theft of the assets. You can recover the value of lost assets either from the thief or from a party who was responsible for securing them. This is why most parking garages explicitly disclaim responsibility on the tickets---they do not want to be legally responsible for securing your vehicle and its contents.
The question here is whether TMobile is legally responsible for ensuring the integrity of his account and SMS communications. I'm not a lawyer so I'm not going to guess at the outcome of a trial. If I were TMobile, I'd probably just pay the ~$20K to avoid court and bad publicity.
They could make the Pro version free for personal use. Or reduced cost. A lot of developers/companies only charge businesses and government for their products.
If they decide to charge for personal use, then we'll have to look at the software to see if the free version is worthwhile on its own merits.
I don't begrudge them a revenue stream. If they're doing this more as a job than a hobby, good for them.
A really programmer will have to learn all of the complex "geeky" concepts---eventually.
A good starter language should abstract away a lot of those things. It is very difficult and potentially discouraging to tackle flow of control, obscure IO interfaces, and UI quirks all at once.
I don't know if this is marketing fluff (never personally written Swift code), but a good starter language that runs on a popular platform should be a basic elective course for middle/high schools.
Out of curiosity, would anyone have suggestions for the best starter languages for Windows and Linux systems? Preferably something with native libraries for implementing a GUI?
Oregon is already electing competent Senators. Well, at least this one---I don't know about their other guy.
We need more tech-literate voters and advocates in other places. It takes 50-60 Senators to pass a good law.
Checkmarx found that they still leaked enough information to tell encrypted commands apart, allowing a hacker on the same network to watch every swipe left, swipe right, or match on the target's phone nearly as easily as if they were looking over the target's shoulder.
When different user actions result in widely different application behavior, it will always be easy to infer the user action. E.g., if matching is the only action that does not result in a new profile being presented, then observation of the smaller data exchange will lead to that inference.
The only way to avoid this is to make the network traffic identical for all cases, which is extremely wasteful of bandwidth and, presumably, battery life.
That said, encryption of all data should be standard now. There is some overhead, but it's not the 1990s---crypto is not that burdensome.
We have had proxy activities for competition, aggression, and violence since the dawn of history. Everything from boxing, rugby, and polo to swimming and track to Go and chess.
Video games are just a new spin on an ancient habit. We substitute relatively harmless activities as outlets for our less-than-friendly instincts. In this respect, we have reached a new zenith with the variety, ubiquity, and flexibility of computer games. Participation in previous sports or hobbies has never been as safe and widely appealing as video games.
I'm not surprised there is no link between video games and violent behavior; the games themselves are the outlet for the urges that lead to violence.
Disney owns so many franchises now that a Netflix without Disney will kill it.
Not really. Netflix's original content ranges from decent to excellent.
They could probably coexist with Disney and a few other streaming content producers. E.g., Netflix + HBO Go + a hypothetical Disney service would total up to less than the average cable bill---unless Disney tries to charge way more than everyone else. That is already more content than most people need. Plus, there is still broadcast TV for news, some sports, and other local content.
Netflix doesn't need to be everything to everyone; it only needs to justify its ~$10/month price tag. More streaming services may actually help, insofar as they encourage cord-cutting, and many cord-cutters will pick up Netflix plus one or two other services.
Most cancers are already far easier to treat if they're caught early.
I bet that comment sounded a lot more clever before you typed it.
You mean the warranty would be enforceable even using 3rd party parts for maintenance?
Third-party parts are usually associated with aftermarket mods, but it works the same for standard maintenance and consumable parts (oil filters, tires, etc).
In these cases, the third-party parts aren't covered by the OEM warranty, but the original parts remain covered. They cannot void your warranty for using third-party parts in the US; the warranty continues to cover the OEM parts regardless.
Maybe you should just stay positive instead.
There is a difference between positivity and the foolish optimism born from ignorance. OP was right in saying that general-purpose "molecular printing" is very, very far from being available.
This isn't even a prototype of such a device---it was a one-off proof of concept with a specific target. If you can't even criticize the reasons for his claim, then maybe you should keep your mouth shut about his opinion.
If they come to fruition, the first generation of molecular printers will likely have a limited use case, much like 3D printers. The headline is misleading at best, and that's not even accounting for the patent and regulatory issues related to drug production.
Lots of hype.
No science.
Wrong.
Researchers have done work in turning the immune system against cancer, and it is remarkably effective. The FDA has approved two CAR-T therapies to date.
CRISPR is just an easier means of tinkering with cellular DNA. This makes it likely that we can develop more effective treatments for more types of cancer.
I believe by default they do the throttling, you can shut it off by sending a specific text message to a specific number.
The feature is called BingeOn, and it can be enabled on a per-line basis for every phone on the account. The default is enabled.
It can be configured by the account owner on the TMobile web site, so the text/app toggles might not work for everyone.
Is Intel developing new chips that don't have this problem?
Of course, but the lead time on CPU development is long. It will be at least a year or two before consumers can buy anything with builtin resistance to Spectre and Meltdown.
Are they going to be slower, too?
The generational improvements will probably offset the losses, so I strongly doubt that will be the case.
Plus, fixed CPUs will not require KPTI anymore, so they will be secure with the "normal" OS-level performance optimizations. Windows and Linux can go back to doing things the pre-Meltdown way if Intel fully addresses their problems.
If humans are no better than a machine then how can a machine programmed by humans do any better
This is a stupid, Luddite sentiment. Computers are already far better than humans at a wide range of tasks. This increases dramatically if your usage of "machine" include robots performing physical tasks.
Deep learning is opening the doors to new machine skills. Quantum computing will likely open a few more in the next few decades. It's anyone's guess what comes after that.
Humans were smart enough to make a machine that could do some things better than they could. We used those new capabilities to make better machines---again and again. The only thing that stops this process is the possibility that a final generation of machines will be incapable of developing meaningful advances for a new generation.
It suggests that the additional variables are either unrelated to the outcome
Or their application was slapped together haphazardly in order to start bringing in those contract dollars as soon as possible. The other 135 variables could provide useful information, if their analysis of that information were better. Until this study, there was no incentive to perform well because the cost of poor performance is borne by convicts.
I am far more willing to believe that crapware is being shoveled into government computers at taxpayer expense.
Bear in mind that this system performs as well as untrained humans. If humans with correctional or psychological expertise perform better than untrained humans (which I assume is true), then this system is embarrassingly bad. It's doing a worse job than the people it's supposed to help.
US national security in the IT space is shot to hell, and most of it is the NSA's fault, with Intel a close second.
Way too short-sighted there.
1. Heartbleed was remotely exploitable. The Spectre vulnerabilities require local code or else a second exploit to allow remote execution. Heartbleed posed a far greater risk to internet as a whole.
2. The NSA didn't create any of the vulnerabilities they exploited. The fault lies with the respective developers such as Cisco, Microsoft, etc. The NSA essentially provided a full disclosure with POC for those vulnerabilities, but all of those problems existed regardless.
3. These zero-days wouldn't be such a huge problem if people did defense-in-depth properly. I.e., at all.
From the often-blamed-on-the-NSA pile:
WannaCry spread by exploiting a vulnerability in SMB. FFS. Who is exposing SMB to the internet in the first place? There are much better choices when it comes to secure file transfer protocols; and there is hardly anything worse.
This was not even the first SMB exploit. Sometimes you can disclaim responsibility by saying "no has ever thought of that before". But in this case, we have not only warned about it---we have seen it firsthand.
This has always been an option, and the industry chose Door #1 years ago because it is by far the most productive and economical.
It's not like the machinery is getting any simpler either. How much data do they send and how often do they need to report? Do they integrate with inventory systems to track usage of raw materials and other consumables? How good are the predictive wear/failure alerts?
More precision and more complex automation are going to push those 1970s-era control systems from outdated to unworkable---if it hasn't happened already.
Granted, all industries are not affected the same way. But you're still fundamentally stuck with buying a mass-produced configurable SCADA supervisor for a reasonable price vs a custom system at an insane price. And the mass-produced system is going to have a bit of that "everything and the kitchen sink" flavor.
Pho is facing 10 years for copying that information. Yes, there are rules and procedures---and he broke them. No sympathy, really.
So how can Russia use software that isn't supposed to be exposed to secret information, to steal secret information?
Did you miss the part where a dumbass contractor copied the files and then put them on his computer at home? It was a courier delivery by Air Retard.
So, you installed just the web protection and then disabled the SSL traffic scanner? Even though 50% of traffic is encrypted now?
Financial and health data is among the SSL-encrypted traffic. If you don't want something seeing those things, then you either need to exempt your bank/healthcare sites or disable the SSL scanner entirely. Enterprise proxies usually offer this out of the box---most US organizations will not decrypt traffic to these destinations.
And besides, it's up to him as to whether he wants SSL decryption at all. The feature should be configurable.
Odds are good that you disabling the SSL middleware means the uninstaller didn't realize it was there and didn't uninstall it.
Shit application, shit installer. There is no reason an application cannot keep track of which modules are installed regardless of whether a user disables them.
Meanwhile, Comodo wants to keep resetting Chrome to use Yahoo "for web protection" and repeatedly reinstalls a Yahoo Search extension if you delete it.
They have a deal with Yahoo, and they value Yahoo's money over your express wishes. That should tell you all you need to know about your security vendor.
Antivirus programs are getting to be a bit pointless these days. If you have good security measures, you won't get hit by the kind of crap they can find in the first place.
TLA's are here and are "controlling the narrative" - but failing. We know there's no other reason to call a legit request for "how you know what you claim" as trolling.
I'm not particularly surprised by the lack of public evidence. Classified computer systems aren't going to be passed around for inspection. Do you really expect to see logs or forensic results from a classified system? You're demanding something that will never be forthcoming, and, in fact, may be illegal to release. From day one, the entire Kaspersky investigation was destined to land in "trust us" territory. There is no other way it could play out.
You think the Russians are doing all the badware on earth?
Since the article is about leaked NSA malware, I don't even know why you're asking this question. Major non-sequitur right there.
Peek-a-boo - I see you, paid "intelligence community trolls with mod points"
I see you've been forgetting your meds. Put on an extra layer of tin foil and consult your psychiatrist on the best way to handle missed doses.
If the attack was successful, then it's the American company's fault for not better protecting.
You sound certain of that. But there is no such thing as perfect security. Without technical details, there is no way to distinguish between a brilliant attack and negligence.
If you deal with government secrets
Be careful with that idea...
leaving them out in a bar for anyone to see is considered treasonous
...because this only applies to classified material.
Also, it's not treason. Things don't automatically become treason because the government is especially inconvenienced.
Finally, I doubt that voting machine designs could even be classified in the first place. There are specific types of information which are eligible for classification, and voting machines simply don't fall into any of them. Go read Executive Order 12958 or 13526 if you're inclined to disagree---it doesn't mater which because Section 1.4 is consistent between them.
If I had an algorithm that occasionally misidentified people in a way that can cause public outrage, I would filter the outputs to avoid controversy too.
Wake me up when they release a fixed version. Hell, a paper describing the issue in detail would be interesting---even fascinating, if I were any sort of expert.
Googles themselves admitted that their algorithms still make the same mistake. This article boils down to "hard problem takes longer than 3 year to solve"---with excessive puffery.
At most, Intel will adjust prices downward to compensate for the performance delta.
This single event isn't enough to turn AMD's fortunes around. AMD's financial woes cut far deeper than a single quarter or even an entire year can fix.
A 10% reduction in that is significant, but not catastrophic for most users.
That 10% is also an average, with some loads seeing a greater hit. I imagine the people on the tail end of the distribution are going to be very unhappy. Also, Intel is only considering processors from the last 3 years, whereas Microsoft has stated publicly that older processors see an even greater hit.
Most enterprises replace their equipment on a 2-4 year cycle, depending on the business. For them, 3-year-old processors are either trash or on the chopping block for the next tech refresh. They will grumble, but they won't really care once the frenzy of patching is done. Home users---especially those on a budgets---are more likely to run into the larger performance hits that Intel isn't discussing.
First, there is no such thing as a "non-profit" college. They ALL profit.
This is a fallacy called equivocation.
The standard use of "non-profit" means that there are no dividends or revenues being paid out to investors or owners. Your use of "profit" in the second sentence is clearly intended to mean something else.
Second, Not relying on tests means relying on transcripts.
This is a false dilemma fallacy. There are more choices than tests and transcripts, including: recommendations, extracurricular/volunteer work, professional experience, essays or personal statements, interviews, and project/portfolio submissions.
I had all of those things except a portfolio when I applied to university, although the professional experience was obviously very limited due to age.
excluding any kind of objective or semi-objective measurements leaves only
Another false dilemma: you imply that making tests optional renders them irrelevant. Making standardized tests optional does not eliminate them as a source of information. A test-optional school can still look favorably on applicants with high scores on relevant tests.
What we have here is the gradual degradation of the US higher Education system
The article indicates that Wake Forest, UGA, and De Paul reviewed their records, and they get around a 1% improvement in their predictions of college success from considering standardized test scores. Why would anyone bother with an ineffective tool?