No. Keywords were allocated. Finding someone on facebook doesn't imply you're using an allocated keyword to get there -- you probably use a search, which is completely different than a keyword. If you typed in "http://www.facebook.com/" to get there, then perhaps. But that's only because URLs typically behave the same way as keywords. The fact remains: these new TLDs serve no beneficial purpose. They're completely redundant. A TLD was meant to imply a category for a website, but that disappeared long ago. Most people are used to typing ".com", ".net", etc. In fact, UIs have been built around these assumptions -- check your iPhone. Sure, we have "smart" URL bars these days in chrome and the likes where a URL that doesn't resolve throws you into a search result for that URL which will probably get you to the right place, but that's an error-prone approach and it's vulnerable to a trust attack. It's a horrible idea to make it even more difficult to guess or remember a company's web address. Liken this to it being announced that everyone can pick a second, third, fourth, and Nth phone number so long as the phone number is 15 digits in length instead of 10, just sign up and pay $N dollars and now everyone can call you from a different, longer number!
This is very clearly just a cash grab. ICANN wants more money. It's time to dissolve them. They cooperate with governments in illegal takedowns and they clearly do not operate in the best interest of the internet. No more.
I am not a psychiatrist nor a psychologist. I do, however, have an explanation I find logical for why both of these questions would get wrong answers.
A bat and ball cost a dollar and ten cents. The bat costs a dollar more than the ball. How much does the ball cost?
The reason this "problem" will yield a common answer of 1 dollar is because so many of us have seen the same thing over and over in school. It has been over the course of 5+ years engraved into our thought process to separate pieces of the sentence into logical portions and stop as soon as we have enough information (ie: to assume most of it is useless information). So as soon as the reader sees the intentionally deceptively worded sentence, it's effectively an expected response from this programmed behavior: most people stop where I'm about to show you:
A bat and a ball cost a dollar and ten cents. The bat costs a dollar --
Immediately, we have a situation: a + b = 110, a = 100. We immediately deduce that b = 10, and have a solution instantaneously without completing the thought. This is what standardized testing and predictable word problems with extraneous information teaches people. This isn't a result of their intelligence, this is a result of cognitive process sculpted by years of stupid, pointless exercises. You'd have to be outrageously stupid to think this is somehow unexpected. The people who we classify as "smart" are people who perform well at these tasks (high score on standardized test, breezed through courses with similar problems). This is causation -- people who make this mental leap are considered "smart." So you ask "why are all these smart people making this stupid mistake!?" The answer is clear -- your fundamental measure of intelligence is wrong. The solution is that these so-called "smart" people aren't very smart at all. They're just good at solving tricky word problems as quickly as possible, primarily by ignoring information. In my experience, this methodology is often the inverse of an intelligent process.
Now for the second problem:
In a lake, there is a patch of lily pads. Every day, the patch doubles in size. If it takes 48 days for the patch to cover the entire lake, how long would it take for the patch to cover half of the lake?
What most people will do, because this is how they've been taught, is to read sentence one. Note it as an interesting fact, then proceed. Upon finishing the second sentence, we realize we didn't come up with an answer yet, so we refer to only the information in the latter part of the question. What most people just read is:
If it takes 48 days for the patch to cover the entire lake, how long would it take for the patch to cover half of the lake?
We aren't used to thinking in terms of exponentiation, so it's natural to assume a linear growth rate when you completely discard the first sentence.
While I agree, these are both absurd questions, they have something in common: people tend to ignore part of the question and answer the question with incomplete information. This is not something I do very often, intentionally. This is something, though, that I recall being the fundamental "trick" to answering 99.99999999999% of questions on standardized tests. They gave you extraneous information. When literally every problem exposed to you has extraneous information, of 2 forms: A, B or B, A, where B = worthless information, it becomes habitual to process information in this manner, especially when the problem is worded like a problem you'd find on a high-school level standardized test (you know, you never really forget how to ride a bike, like you never forget how to solve very badly designed problems that don't test intelligence in any way).
I don't know, maybe I'm too smart for this researcher. But the answer seems obvious: years and yea
One more bug in the footsteps of Larry Page? This guy is out of control. Somebody at Google needs to stop this maniac before he does something really bad.
I'm quite confident it's Larry Page. A lot of really bad decisions coming from Google lately started after he was named CEO. I don't think that's a coincidence.
Availability, convenience, you name it. For instance, imagine being in an area where your have no cell phone service and you don't have access to anything but a dumb terminal but you need to communicate with someone. There's always e-mail, perhaps, but being able to jump on a messaging service by visiting a website to talk to someone is an incredibly awesome feature.
Sure, there's facebook chat and gtalk, but we're talking about a conversation we might want to be private. Neither facebook nor google have a good track record there. What's more saddening: Google's buying one of the better services. Time to make a new one or move to one someone else has made for the same reason. I've thought about making a Meebo clone on my vpserver for my personal use (and a guarantee of privacy, etc), but it's not the simplest task in the world. I am quite sad to see this service get extinguished by Google. They're really lining up the horrible headlines lately, bad decision after bad decision. We really need Eric back in the saddle. Larry seems to be a horrible CEO.
Because I did this once. You can save your company ten times your annual salary and not get paid a dime more. Companies tend to think if you can do what we want without spending the money, you don't need the money, and so they throw it at departments that will spend it instead of trying to save the company's money. The real problem: that money is dedicated as "expendable" but marked as "not for salaries." Therefore, you can hire a consultant at $100/hr but you can't give any of your employees even $.01 of it. You probably can't even use it to fund "perks" like free lunches and sodas, etc. Why bother saving massive amounts of time and money when you get ABSOLUTELY NOTHING for it?
The skillset required to do this demands significantly more than the salary companies are willing to offer for it. It's that simple. Therefore it's impossible for positions like this to exist.
The problem is that "UX" has lost its original meaning. It now refers to artists. A person who designs the horribly unusable "user interfaces" you see in movies would pass as a "UX" designer these days. Art and aesthetic design is NOT in any way related to user experience. Google recently demonstrated this to us in a really huge way (gmail). Microsoft is doing it in many ways in W8 and did so with Ribbon.
It's about time to fire all of these artists ruining software and get some real UX engineers.
I have a suggestion as to how to fix patents and copyrights. It's really simple, IMO.
Patents:
Infringement can now only occur if a person or company is directly competing in the same realm as the patent owner for the same product. This is intended to do what patents were meant to do originally: prevent competitors from reproducing your patented work and selling it more cheaply because they don't have the same underlying R&D cost. It would not, however, give you the ability to sue any company producing your product but in a manner that does not compete directly with your means of income.
As a side effect of this new rule, not actually producing the thing your patent protects voids your rights to collect royalties on products created that do use your patent. This is the anti-patent-troll effect.
Copyrights:
1. Better define "fair use" and expand it dramatically.
2. Redefine infringement to not include the use of a work in a manner that is not generating revenue in direct competition with the copyright owner. (similar to the patent idea above)
3. All copyright owners must make a good faith effort to make available to all persons a copy at a reasonable price and by reasonable means. Failure to adhere to this requirement forfeits all exclusive rights granted by said copyright until such time as this requirement is met.
"Strike 'persons' and insert 'any human being that is also a citizen of the United States'."
Just wanted to clarify there. Remember, corporations are persons, and that provision didn't prevent them from amassing huge quantities of copyrights/patents and becoming more powerful than the government.
And not because it's technically too challenging. The reason you don't teach them ANYTHING at all about programming is because of a fact that transcends computer science and indeed is present in all facets of life. It is a simple and well known truth that the more you know about a subject, the more you know of your own limitations within that subject. This comes with a caveat, though. When first introduced to a subject, a person is more ignorant of that subject than they were when they didn't even know it existed, because they know something about it and think that knowledge extends far beyond where it actually does. Essentially, they gain a new perspective, but they don't know how to use that perspective correctly.
So to show someone that they can do the same thing those really smart engineers can do and give them the know-how to turn basic ideas into actual working code, they immediately internalize that process and look at problems in a new way. They begin looking at problems as engineers, not as salespeople. But they ARE NOT ENGINEERS. To give them that perspective without the full understanding it requires to be used properly is a really, really bad idea, in my opinion. You know how those really novice engineers will say "sure, I can do that" to any problem, without knowing about and therefore not considering the > 9000 edge cases they need to deal with because the basic idea is so simple you can explain it to a third grader? Now replace "really novice engineer" with "salesperson who is also a really novice engineer." I can't imagine a more horrific situation.
When will the developers at Microsoft tell their bosses "no, no we can't put Internet Explorer in that, because it fucking sucks. That's right, it sucks. Internet Explorer is FUCKING TERRIBLE." Just be sure to record it and put it up on YouTube so we can all witness such pure win.
Yep, because they took away file sharing. It has nothing to do with the fact that they took an awesome fan-centric event created by and for the community and turned it into a profit-oriented commercial showcase (with a little LAN attached to it, too). Gone is the sense of meritocracy and individuality it conveyed. Gone is the enjoyment and the pride of attending and that which gave people the drive to help out and contribute. What I'm saying here is effectively gone is Quakecon. Enjoy your ZeniMaxCon, though.
Structure isn't really the important part. Knowing what everything does, both primarily and any side effects is what's actually hard about a code base, and why documentation is useful. Code dumps don't convey this, reading the code might, but then you're just generating documentation for them. And while reading the source is a fun thing to do in itself, for businesses and individuals/groups that want to use it to make something, it's important that information is available at a glance rather than after hours of analyzing source code. I would bet if you got a commercial license you'd get a really huge helping of documentation + direct access for support. I don't see any of that with the GPL releases, so they're effectively just code dumps. Interesting, but not terribly awesome.
It has nothing to do with id. It has everything to do with ZeniMax. Just take a look at what they've done to QuakeCon. Never again will I attend a QuakeCon event nor will I ever purchase a title from any company owned by or affiliated with ZeniMax. Never.
I say we postpone any decisions like this until after they figure out how to send information faster than the speed of light. Greed is a pretty powerful tool, it appears.
To set this up, install it, buy all of the necessary equipment, and hire the people to maintain it you're going to spend a considerable amount effort and money for a net gain of absolutely zero value over the current system. Unless you can get a phenomenal deal on a guaranteed Gbps+ line (which is highly unlikely), and your tenants unanimously prefer WAP and ethernet direct connections over their own choice, just let the local ISPs handle it. There's just no real plus side here. Unless you're wiring fiber connections to each condo from a backbone with the hardware capable of handling that now and moving forward (not cheap), your solution won't scale as well as coax to each room and will come with the added cost of needing equipment to terminate the fiber in each condo, which is also not commonplace (read: cheap) in consumer networking equipment yet (last I checked). You can implement something that will work with today's speeds but in 10-20 years, you'll have to retrofit the entire building to handle then-modern speeds, and that's just a massive red number you don't ever want to see.
Round-trippers (one full page per GET with some sprinkling of JS effects here and there) are going out of style. It seems though that some companies, notably Twitter are heading towards doing more work on the back-end for users who have less capability to render the page, for performance reasons. Client-side heavy apps scale better by default, but if UX is the primary goal, some of the stuff they're doing is pretty cool to consider.
GMail, and much of Google's stuff is written with GWT, which is a platform for developing the client-side of web applications that compiles Java code into JS. You can't really make a modern web app with Java and run it as java (you know, through a JRE, unless someone makes a JRE in the browser, and lol @ that), but you can cross-compile. GWT is like Coffeescript on steroids with a framework behind it as well. These meta-compilation schemes are becoming more and more popular. Look at Facebook, they write their website in PHP and compile it to C++ for efficiency with Hiphop. Google's Traceur compiles code from ES5+ down to ES3 so you can write code with advanced ES5 features on modern browsers that still only have ES3 support.
It is still pretty common to write Java back-ends (primarily on Tomcat and a few other major players), but that's becoming less and less common in newer more modern web apps. I have no citation for this outside of my own observations, and I surely haven't seen everything so I might just be horribly wrong.
After reading the PDF, the conclusion is absolutely not that "geezers pick stronger passwords," rather that in a snapshot of data, accounts with ages under 25 had significantly less strong passwords than those over 55. This doesn't take a LOT of information into account, it's just a passing observation in a paper not really pointed towards this analysis. For instance, there are a lot more young people than old people, unless you account for this, you can easily argue that there are a lot more weak passwords from "younguns" than "geezers." There's also the issue of bot vs real person, active account vs inactive account (which he does address, but which is not mentioned in either this summary nor TFA, when he talks about password updates implying an increase in strength, which would imply "geezers" who still use Yahoo are likely to have updated their passwords more than "younguns" that haven't logged in in over 5 years who would have relatively weak passwords as a result).
Overall, the paper is interesting, but this summary and TFA are completely wrong in their conclusions.
Slashdot Mirror
No. Keywords were allocated. Finding someone on facebook doesn't imply you're using an allocated keyword to get there -- you probably use a search, which is completely different than a keyword. If you typed in "http://www.facebook.com/" to get there, then perhaps. But that's only because URLs typically behave the same way as keywords. The fact remains: these new TLDs serve no beneficial purpose. They're completely redundant. A TLD was meant to imply a category for a website, but that disappeared long ago. Most people are used to typing ".com", ".net", etc. In fact, UIs have been built around these assumptions -- check your iPhone. Sure, we have "smart" URL bars these days in chrome and the likes where a URL that doesn't resolve throws you into a search result for that URL which will probably get you to the right place, but that's an error-prone approach and it's vulnerable to a trust attack. It's a horrible idea to make it even more difficult to guess or remember a company's web address. Liken this to it being announced that everyone can pick a second, third, fourth, and Nth phone number so long as the phone number is 15 digits in length instead of 10, just sign up and pay $N dollars and now everyone can call you from a different, longer number!
This is very clearly just a cash grab. ICANN wants more money. It's time to dissolve them. They cooperate with governments in illegal takedowns and they clearly do not operate in the best interest of the internet. No more.
The reason this "problem" will yield a common answer of 1 dollar is because so many of us have seen the same thing over and over in school. It has been over the course of 5+ years engraved into our thought process to separate pieces of the sentence into logical portions and stop as soon as we have enough information (ie: to assume most of it is useless information). So as soon as the reader sees the intentionally deceptively worded sentence, it's effectively an expected response from this programmed behavior: most people stop where I'm about to show you:
Immediately, we have a situation: a + b = 110, a = 100. We immediately deduce that b = 10, and have a solution instantaneously without completing the thought. This is what standardized testing and predictable word problems with extraneous information teaches people. This isn't a result of their intelligence, this is a result of cognitive process sculpted by years of stupid, pointless exercises. You'd have to be outrageously stupid to think this is somehow unexpected. The people who we classify as "smart" are people who perform well at these tasks (high score on standardized test, breezed through courses with similar problems). This is causation -- people who make this mental leap are considered "smart." So you ask "why are all these smart people making this stupid mistake!?" The answer is clear -- your fundamental measure of intelligence is wrong. The solution is that these so-called "smart" people aren't very smart at all. They're just good at solving tricky word problems as quickly as possible, primarily by ignoring information. In my experience, this methodology is often the inverse of an intelligent process.
Now for the second problem:
What most people will do, because this is how they've been taught, is to read sentence one. Note it as an interesting fact, then proceed. Upon finishing the second sentence, we realize we didn't come up with an answer yet, so we refer to only the information in the latter part of the question. What most people just read is:
We aren't used to thinking in terms of exponentiation, so it's natural to assume a linear growth rate when you completely discard the first sentence.
While I agree, these are both absurd questions, they have something in common: people tend to ignore part of the question and answer the question with incomplete information. This is not something I do very often, intentionally. This is something, though, that I recall being the fundamental "trick" to answering 99.99999999999% of questions on standardized tests. They gave you extraneous information. When literally every problem exposed to you has extraneous information, of 2 forms: A, B or B, A, where B = worthless information, it becomes habitual to process information in this manner, especially when the problem is worded like a problem you'd find on a high-school level standardized test (you know, you never really forget how to ride a bike, like you never forget how to solve very badly designed problems that don't test intelligence in any way).
I don't know, maybe I'm too smart for this researcher. But the answer seems obvious: years and yea
One more bug in the footsteps of Larry Page? This guy is out of control. Somebody at Google needs to stop this maniac before he does something really bad.
I'm quite confident it's Larry Page. A lot of really bad decisions coming from Google lately started after he was named CEO. I don't think that's a coincidence.
Availability, convenience, you name it. For instance, imagine being in an area where your have no cell phone service and you don't have access to anything but a dumb terminal but you need to communicate with someone. There's always e-mail, perhaps, but being able to jump on a messaging service by visiting a website to talk to someone is an incredibly awesome feature.
Sure, there's facebook chat and gtalk, but we're talking about a conversation we might want to be private. Neither facebook nor google have a good track record there. What's more saddening: Google's buying one of the better services. Time to make a new one or move to one someone else has made for the same reason. I've thought about making a Meebo clone on my vpserver for my personal use (and a guarantee of privacy, etc), but it's not the simplest task in the world. I am quite sad to see this service get extinguished by Google. They're really lining up the horrible headlines lately, bad decision after bad decision. We really need Eric back in the saddle. Larry seems to be a horrible CEO.
Because I did this once. You can save your company ten times your annual salary and not get paid a dime more. Companies tend to think if you can do what we want without spending the money, you don't need the money, and so they throw it at departments that will spend it instead of trying to save the company's money. The real problem: that money is dedicated as "expendable" but marked as "not for salaries." Therefore, you can hire a consultant at $100/hr but you can't give any of your employees even $.01 of it. You probably can't even use it to fund "perks" like free lunches and sodas, etc. Why bother saving massive amounts of time and money when you get ABSOLUTELY NOTHING for it?
The skillset required to do this demands significantly more than the salary companies are willing to offer for it. It's that simple. Therefore it's impossible for positions like this to exist.
The problem is that "UX" has lost its original meaning. It now refers to artists. A person who designs the horribly unusable "user interfaces" you see in movies would pass as a "UX" designer these days. Art and aesthetic design is NOT in any way related to user experience. Google recently demonstrated this to us in a really huge way (gmail). Microsoft is doing it in many ways in W8 and did so with Ribbon.
It's about time to fire all of these artists ruining software and get some real UX engineers.
Reading while pooping. A much more concise example, in my opinion.
I have a suggestion as to how to fix patents and copyrights. It's really simple, IMO.
Patents:
Infringement can now only occur if a person or company is directly competing in the same realm as the patent owner for the same product. This is intended to do what patents were meant to do originally: prevent competitors from reproducing your patented work and selling it more cheaply because they don't have the same underlying R&D cost. It would not, however, give you the ability to sue any company producing your product but in a manner that does not compete directly with your means of income.
As a side effect of this new rule, not actually producing the thing your patent protects voids your rights to collect royalties on products created that do use your patent. This is the anti-patent-troll effect.
Copyrights:
1. Better define "fair use" and expand it dramatically.
2. Redefine infringement to not include the use of a work in a manner that is not generating revenue in direct competition with the copyright owner. (similar to the patent idea above)
3. All copyright owners must make a good faith effort to make available to all persons a copy at a reasonable price and by reasonable means. Failure to adhere to this requirement forfeits all exclusive rights granted by said copyright until such time as this requirement is met.
I move that we amend that amendment:
"Strike 'persons' and insert 'any human being that is also a citizen of the United States'."
Just wanted to clarify there. Remember, corporations are persons, and that provision didn't prevent them from amassing huge quantities of copyrights/patents and becoming more powerful than the government.
And not because it's technically too challenging. The reason you don't teach them ANYTHING at all about programming is because of a fact that transcends computer science and indeed is present in all facets of life. It is a simple and well known truth that the more you know about a subject, the more you know of your own limitations within that subject. This comes with a caveat, though. When first introduced to a subject, a person is more ignorant of that subject than they were when they didn't even know it existed, because they know something about it and think that knowledge extends far beyond where it actually does. Essentially, they gain a new perspective, but they don't know how to use that perspective correctly.
So to show someone that they can do the same thing those really smart engineers can do and give them the know-how to turn basic ideas into actual working code, they immediately internalize that process and look at problems in a new way. They begin looking at problems as engineers, not as salespeople. But they ARE NOT ENGINEERS. To give them that perspective without the full understanding it requires to be used properly is a really, really bad idea, in my opinion. You know how those really novice engineers will say "sure, I can do that" to any problem, without knowing about and therefore not considering the > 9000 edge cases they need to deal with because the basic idea is so simple you can explain it to a third grader? Now replace "really novice engineer" with "salesperson who is also a really novice engineer." I can't imagine a more horrific situation.
Almost all of those are using commercial licenses. You know, the one that comes with documentation + support. The non-code-dump version.
When will the developers at Microsoft tell their bosses "no, no we can't put Internet Explorer in that, because it fucking sucks. That's right, it sucks. Internet Explorer is FUCKING TERRIBLE." Just be sure to record it and put it up on YouTube so we can all witness such pure win.
Yep, because they took away file sharing. It has nothing to do with the fact that they took an awesome fan-centric event created by and for the community and turned it into a profit-oriented commercial showcase (with a little LAN attached to it, too). Gone is the sense of meritocracy and individuality it conveyed. Gone is the enjoyment and the pride of attending and that which gave people the drive to help out and contribute. What I'm saying here is effectively gone is Quakecon. Enjoy your ZeniMaxCon, though.
Structure isn't really the important part. Knowing what everything does, both primarily and any side effects is what's actually hard about a code base, and why documentation is useful. Code dumps don't convey this, reading the code might, but then you're just generating documentation for them. And while reading the source is a fun thing to do in itself, for businesses and individuals/groups that want to use it to make something, it's important that information is available at a glance rather than after hours of analyzing source code. I would bet if you got a commercial license you'd get a really huge helping of documentation + direct access for support. I don't see any of that with the GPL releases, so they're effectively just code dumps. Interesting, but not terribly awesome.
Code dumps. How useful.
It has nothing to do with id. It has everything to do with ZeniMax. Just take a look at what they've done to QuakeCon. Never again will I attend a QuakeCon event nor will I ever purchase a title from any company owned by or affiliated with ZeniMax. Never.
Scaleform?
http://gameware.autodesk.com/usage/games
The one thing I think all of these games has in common: a terrible UI. Coincidence? I think not.
I say we postpone any decisions like this until after they figure out how to send information faster than the speed of light. Greed is a pretty powerful tool, it appears.
To set this up, install it, buy all of the necessary equipment, and hire the people to maintain it you're going to spend a considerable amount effort and money for a net gain of absolutely zero value over the current system. Unless you can get a phenomenal deal on a guaranteed Gbps+ line (which is highly unlikely), and your tenants unanimously prefer WAP and ethernet direct connections over their own choice, just let the local ISPs handle it. There's just no real plus side here. Unless you're wiring fiber connections to each condo from a backbone with the hardware capable of handling that now and moving forward (not cheap), your solution won't scale as well as coax to each room and will come with the added cost of needing equipment to terminate the fiber in each condo, which is also not commonplace (read: cheap) in consumer networking equipment yet (last I checked). You can implement something that will work with today's speeds but in 10-20 years, you'll have to retrofit the entire building to handle then-modern speeds, and that's just a massive red number you don't ever want to see.
Yeah but he's describing Meetup, seems like that battle is already lost.
Round-trippers (one full page per GET with some sprinkling of JS effects here and there) are going out of style. It seems though that some companies, notably Twitter are heading towards doing more work on the back-end for users who have less capability to render the page, for performance reasons. Client-side heavy apps scale better by default, but if UX is the primary goal, some of the stuff they're doing is pretty cool to consider.
Source maps fix this issue.
GMail, and much of Google's stuff is written with GWT, which is a platform for developing the client-side of web applications that compiles Java code into JS. You can't really make a modern web app with Java and run it as java (you know, through a JRE, unless someone makes a JRE in the browser, and lol @ that), but you can cross-compile. GWT is like Coffeescript on steroids with a framework behind it as well. These meta-compilation schemes are becoming more and more popular. Look at Facebook, they write their website in PHP and compile it to C++ for efficiency with Hiphop. Google's Traceur compiles code from ES5+ down to ES3 so you can write code with advanced ES5 features on modern browsers that still only have ES3 support.
It is still pretty common to write Java back-ends (primarily on Tomcat and a few other major players), but that's becoming less and less common in newer more modern web apps. I have no citation for this outside of my own observations, and I surely haven't seen everything so I might just be horribly wrong.
After reading the PDF, the conclusion is absolutely not that "geezers pick stronger passwords," rather that in a snapshot of data, accounts with ages under 25 had significantly less strong passwords than those over 55. This doesn't take a LOT of information into account, it's just a passing observation in a paper not really pointed towards this analysis. For instance, there are a lot more young people than old people, unless you account for this, you can easily argue that there are a lot more weak passwords from "younguns" than "geezers." There's also the issue of bot vs real person, active account vs inactive account (which he does address, but which is not mentioned in either this summary nor TFA, when he talks about password updates implying an increase in strength, which would imply "geezers" who still use Yahoo are likely to have updated their passwords more than "younguns" that haven't logged in in over 5 years who would have relatively weak passwords as a result).
Overall, the paper is interesting, but this summary and TFA are completely wrong in their conclusions.