Yeah, I feel like they're sort of trying to take of Slack in terms of being a professional collaboration hub, but it's yet another experiment with some overlap with Google's other products, without a clear sense of how real people are supposed to use all of these semi-overlapping tools.
I understand the value of experimenting, but at a certain point, especially when you're targeting businesses, you need to make a stable and robust product instead of starting from scratch every couple of years. And I also understand the idea of creating versatile tools and letting users find ways to use them, but to some extent you also have to come up with some suggested and likely use-cases that are actually going to be useful. Google seems to flounder at both of those things.
For example, can they just come up with a coherent strategy on messaging/chat apps?
"Because these flights were consistently late, airlines have now baked delays..." is just a biased way of saying they realized they were wrong, and fixed it.
It seems like the truth might be somewhere in between. There was a schedule, and flights could have stuck to that schedule. However, the flights kept getting delayed for a variety of reasons-- some economic, some technical, some regulatory. Being unable to fix those problems right now, airlines have chosen to pad their schedules so that flights are more likely to arrive on time. It's a reasonable approach, but not necessarily the best outcome. It'd be great if we could fix the myriad problems with air travel, but our society has thrown up its hands and given up on trying to fix problems.
Sorry, that's not fair. If it's a problem that can be fixed by spending a few months creating a half-assed mobile app and then sell your company for a billion dollars, then we'll come up with 20 different solutions to that problem. Otherwise, nope.
Not being self-aware doesn't mean it couldn't be dangerous. There's just an inherent problem in that AI systems might come up with solutions that we don't like.
I remember a story a little while back where they were training an AI to solve a maze in the shortest time possible. It happened to do something that basically caused the program to crash. Since its parameters didn't distinguish between solving the maze and simply having the program end, it settled on causing the crash as the fastest way to "complete the maze".
Now in some ways, it can be helpful that AI would come up with a solution like that. If the AI finds a solution outside of our range of expected solutions because we didn't set adequate parameters, it could come up with a great solution we just hadn't considered. However, it might also come up with a terrible solution that we wouldn't consider because it doesn't actually accomplish the thing we want to accomplish, or because it has a side-effect that we consider unacceptable.
Therefore, for safety's sake, it make sense to have a person in the loop to review the AI's decisions to make sure they accomplish what we want to accomplish and have no unacceptable side-effects. That's true whether the AI is sentient or not.
Except the goal of the college then stops being to educate you, and becomes all about pushing you towards career goals that are lucrative for the college.
That might be fine for some strictly vocational programs. If the whole point is to teach you a trade and then place you in a job, then getting paid based on the earnings at the job they've places you in makes a certain amount of sense. However, if we're talking about college for the purpose of higher education, it makes a lot less sense.
There seems to be an implication that it was just a cover story to cover the illegal source of the information. I don't know if that's true, but that's what I think de Becker is suggesting.
Every time I've heard someone say it out loud, it's jay-peg. Never once have I heard someone say jay-feg. Yet the P comes from "photograph", which means that, by the logic of the parent post, the P should sound like an F. We all have to pronounce JPEG as jay-feg.
By that logic, if the vote count is so close that the electoral college makes a difference, then the electoral college isn't the solution.
Fundamental to your argument is that the electoral college can't stray far enough from the popular vote to make a meaningful difference. So if that's true, why bother having it?
There's also a very good argument that computerized voting is just inherently untrustworthy. You have to trust the developers both to make it work properly in the first place, and to make it unhackable. You need to make sure the code is audited, and that the software actually running is identical to the audited code. You need to make sure that nothing has been tampered with-- not the hardware or software of the machines themselves, nor any systems where the information is gathered or stored. And as you address all of those things, you're going to make the system more complex, which decreases any given person's ability to understand how the whole system is working, thereby increasing the likelihood that some security hole might be missed.
Paper? You can have verifiable physical security. We've been doing physical security for thousands of years, and generally have a pretty good ability to do that. People can lay eyes on the votes. There isn't much to understand or be fooled by.
I think if you do electronic voting, there should at least be a legal requirement that it prints a paper receipt that you can view and verify that it printed what you wanted it to print, and then that receipt should be handled with the same physical security as we would provide to a paper ballot. The electronic voting can provide quick and easy counting, but the paper receipt would give a definite audit trail.
It's also important to note that there's been a long history of the US trying to suppress minorities from voting. As soon as black people had the right to vote, there were laws specifically designed to make it hard for them to vote. So when someone suddenly proposes a new law that makes it harder to vote, and seems like it would disproportionately affect minorities, it should be understandable that people would be suspicious. They should be prepared to make a case why the change is needed, and how they're going to prevent any disparities in who faces hardship in voting.
Oh, he's done stuff. He's certainly committed crimes. This report was specifically addressing two questions:
1) Did Trump actively and knowingly conspire with the Russians to hack Hillary's emails in a way that constitutes a specific crime? The short answer here was, "We didn't find enough evidence to charge him with a crime."
2) Did Trump technically commit the crime of obstruction of justice? The short answer here was, "He at least kind of did, but it's a difficult legal question and I'll leave it to the Attorney General to decide."
So far, we only know those short answers, not the long answers. Either way, there's still work for Congress to do to investigate.
I don't think you read it in context. That quote is specific to the charge of obstruction of justice, and the report says that Mueller gathered up the facts and declined to evaluate whether the activity constituted a crime. That's the context of "While this report does not conclude that the President committed a crime, it also does not exonerate him."
The implication is that the President did things which are debatably criminal, but Mueller felt there were enough legal/constitutional issues that it was proper to have the AG decide whether those actions constituted a crime. We can't be sure why that is without having seen the report, but we could guess:
There's been an ongoing debate because the President explicitly stated that he fired Comey to stop the Russian investigation, which is, in non-legal terms, obstructing an ongoing criminal investigation. On the other side, there have been variations of the Nixonian argument that "when the President does it, it's not illegal." Attempting to charge the President in this case would almost certainly go to the Supreme Court and create a bit of a constitutional crisis, and Mueller seems to have decided that it was simply "above his pay grade" as the special counsel, and a decision the AG should make.
It's also noteworthy that the AG decided not to prosecute on the grounds that it was a question of corrupt intent, and the President's intent couldn't be proven beyond a reasonable doubt.
Also, Trump's campaign manager, son, and son-in-law met with an agent of the Russian government for the purpose of coordinating campaign assistance. I can believe that there isn't enough evidence of a specific crime to charge them with anything, but it's still collusion and it should still be an enormous scandal.
They simply determine whether the evidence indicates a person committed a crime or not.
Not quite. They determine whether there's enough evidence to punish someone for having committed a crime. The evidence might still indicate that they've committed a crime, but there might not be enough of it.
I feel like the solution is to prevent people from spoofing caller ID. If it's a physical phone, the phone company should be able to verify the physical path to some extent. If it's a VoIP number, put some PKI in place that requires connection be signed somehow-- something like DKIM for VoIP.
It's not a complete solution, but you're not going to solve the problem if you can't create some system of accountability.
Apple has stated repeatedly they want nothing like the singularity, that desktops are inherently different than tablets or mobile devices.
Well specifically what they said was, they think an attempt to have a singular mobile + desktop UI is doomed to fail, and they're not going to do it. However, I don't think that necessarily means that they wouldn't consider a unified OS.
This news is that they're likely to move some of their desktop machines to use the same processor line as their mobile devices. Meanwhile, they recently announced that they're making it easier in Xcode to make a single app that will run on both iOS and macOS with just a change in UI.
What they may be headed towards is an ecosystem where app developers have the ability to make a single app that presents a mobile UI on a mobile device and a desktop UI on a laptop/desktop. Then, they could make phones and tablets that display an entire mobile UI when operating as a mobile device, but turn into a fully-functioning desktop machine with a desktop UI (and all the apps presenting a desktop UI) when docked.
I don't think it's just "Unix-ish". macOS is technically actual Unix, and iOS is just a modified version of macOS. As far as I know, iOS is actual Unix, based off of BSD.
It's just hobbled by being in a walled garden, unfortunately.
To my mind, most Apple users can get away with a lot less CPU power than they've got in their laptops and desktops right now
In general, Apple's approach to the market is not really to put out commodity computers that compete on the metric of raw computing power. The anti-Apple crowd never seems to grasp that. It's more helpful to think of them as producing appliance computers as a whole product, aimed at specific use-cases. Granted, one of the main use-cases is general business and home productivity computing, but that's still a particular use case.
It may sound like I'm just spouting nonsense, but here's my point: They're selling appliances to let people check their email, browse the Internet, and edit Word documents. None of that requires much processing power, relative to how much power is in a current-generation CPU. Apple doesn't need to push the envelope in high-performance CPUs. Instead, they tend to focus on things like a convenient and attractive form factor, making things lightweight, making the battery last longer, having a pretty built-in screen, etc.
If they can deliver on those things somehow better with ARM CPUs, I wouldn't be surprised if they did that even if it meant less raw power, and I wouldn't be surprised if their fans were happy with the result.
I know, it's so stupid to think you'd travel around the universe in a planet. If you were a real pro, you'd launch the entire solar system in the direction you want to go. That way, you bring a nice power source with you.
Again, I think that's a misunderstanding of genetics and evolution. It's not like genes are simply advantageous and disadvantageous, and advantageous genes are spread and disadvantageous genes die out. Genes might have multiple effects which might be advantageous or disadvantageous based on the environment the organism finds itself in.
But also, being advantageous or disadvantageous isn't by itself the issue. It had to be advantageous enough to increase your chances of surviving long enough to reproduce, or disadvantageous enough to decrease you chances of surviving long enough to reproduce. If it's just sort of theoretically disadvantageous to the human race in general, but you can still live to maturity and have offspring, and that gene will still get passed on. Evolution does not refine people to the point of perfection, it merely to the point of enabling procreation.
Any person would be in jail, but when companies do things that would put a person in jail, the worst thing that happens to corporations is that they get fined. More often, lawmakers hold meaningless hearings and then say, "Let the invisible hand sort this out." One of the main features of corporations is to reduce liability for everyone involved, but that also tends to turn companies collectively into risk-taking psychopaths.
All it takes for reform is that we elect people interested in reform.
I mean, it still has a cut-out for the front-facing camera. It's smaller, which is nice, but it looks like it might just be a camera instead of the multiple sensors that Apple puts in the notch.
So it's not like they totally avoided the problem and didn't do anything like the notch. In both cases, the manufacturer was trying to find a way to have an edge-to-edge screen while also putting front-facing sensors. In both cases, they decided to resolve it by having a cut-out in the screen. Apple did a larger cut-out with a multiple sensors, centered at the very top of the screen. Samsung did a smaller cut-out with fewer sensors, not connected to the top of the screen, offset to one side. It's kind of just a different design choice of how to implement the same basic thing.
I don't think there's any doubt that one of the things that will happen is, iOS developers will make lazy macOS ports of their apps, creating junky apps that look and behave like mobile apps running incongruously on a desktop OS. I don't see that as a problem in and of itself. Give developers tools, and let them do what they want. Some will make junk, but that's fine.
Just so long as Apple doesn't use it as an excuse to further limit what developers can do on their OS. We need iOS to become more open like a desktop operating system, not for the macOS to become more closed off like a mobile operating system.
Slashdot Mirror
Yeah, I feel like they're sort of trying to take of Slack in terms of being a professional collaboration hub, but it's yet another experiment with some overlap with Google's other products, without a clear sense of how real people are supposed to use all of these semi-overlapping tools.
I understand the value of experimenting, but at a certain point, especially when you're targeting businesses, you need to make a stable and robust product instead of starting from scratch every couple of years. And I also understand the idea of creating versatile tools and letting users find ways to use them, but to some extent you also have to come up with some suggested and likely use-cases that are actually going to be useful. Google seems to flounder at both of those things.
For example, can they just come up with a coherent strategy on messaging/chat apps?
"Because these flights were consistently late, airlines have now baked delays..." is just a biased way of saying they realized they were wrong, and fixed it.
It seems like the truth might be somewhere in between. There was a schedule, and flights could have stuck to that schedule. However, the flights kept getting delayed for a variety of reasons-- some economic, some technical, some regulatory. Being unable to fix those problems right now, airlines have chosen to pad their schedules so that flights are more likely to arrive on time. It's a reasonable approach, but not necessarily the best outcome. It'd be great if we could fix the myriad problems with air travel, but our society has thrown up its hands and given up on trying to fix problems.
Sorry, that's not fair. If it's a problem that can be fixed by spending a few months creating a half-assed mobile app and then sell your company for a billion dollars, then we'll come up with 20 different solutions to that problem. Otherwise, nope.
By what standards would a superior intelligence judge us "worthy"? How do you judge whether bedbugs are worthy to live?
Not being self-aware doesn't mean it couldn't be dangerous. There's just an inherent problem in that AI systems might come up with solutions that we don't like.
I remember a story a little while back where they were training an AI to solve a maze in the shortest time possible. It happened to do something that basically caused the program to crash. Since its parameters didn't distinguish between solving the maze and simply having the program end, it settled on causing the crash as the fastest way to "complete the maze".
Now in some ways, it can be helpful that AI would come up with a solution like that. If the AI finds a solution outside of our range of expected solutions because we didn't set adequate parameters, it could come up with a great solution we just hadn't considered. However, it might also come up with a terrible solution that we wouldn't consider because it doesn't actually accomplish the thing we want to accomplish, or because it has a side-effect that we consider unacceptable.
Therefore, for safety's sake, it make sense to have a person in the loop to review the AI's decisions to make sure they accomplish what we want to accomplish and have no unacceptable side-effects. That's true whether the AI is sentient or not.
Except the goal of the college then stops being to educate you, and becomes all about pushing you towards career goals that are lucrative for the college.
That might be fine for some strictly vocational programs. If the whole point is to teach you a trade and then place you in a job, then getting paid based on the earnings at the job they've places you in makes a certain amount of sense. However, if we're talking about college for the purpose of higher education, it makes a lot less sense.
There seems to be an implication that it was just a cover story to cover the illegal source of the information. I don't know if that's true, but that's what I think de Becker is suggesting.
Or how about JPEG?
Every time I've heard someone say it out loud, it's jay-peg. Never once have I heard someone say jay-feg. Yet the P comes from "photograph", which means that, by the logic of the parent post, the P should sound like an F. We all have to pronounce JPEG as jay-feg.
Except that acronyms don't work that way.
By that logic, if the vote count is so close that the electoral college makes a difference, then the electoral college isn't the solution.
Fundamental to your argument is that the electoral college can't stray far enough from the popular vote to make a meaningful difference. So if that's true, why bother having it?
There's also a very good argument that computerized voting is just inherently untrustworthy. You have to trust the developers both to make it work properly in the first place, and to make it unhackable. You need to make sure the code is audited, and that the software actually running is identical to the audited code. You need to make sure that nothing has been tampered with-- not the hardware or software of the machines themselves, nor any systems where the information is gathered or stored. And as you address all of those things, you're going to make the system more complex, which decreases any given person's ability to understand how the whole system is working, thereby increasing the likelihood that some security hole might be missed.
Paper? You can have verifiable physical security. We've been doing physical security for thousands of years, and generally have a pretty good ability to do that. People can lay eyes on the votes. There isn't much to understand or be fooled by.
I think if you do electronic voting, there should at least be a legal requirement that it prints a paper receipt that you can view and verify that it printed what you wanted it to print, and then that receipt should be handled with the same physical security as we would provide to a paper ballot. The electronic voting can provide quick and easy counting, but the paper receipt would give a definite audit trail.
It's also important to note that there's been a long history of the US trying to suppress minorities from voting. As soon as black people had the right to vote, there were laws specifically designed to make it hard for them to vote. So when someone suddenly proposes a new law that makes it harder to vote, and seems like it would disproportionately affect minorities, it should be understandable that people would be suspicious. They should be prepared to make a case why the change is needed, and how they're going to prevent any disparities in who faces hardship in voting.
Oh, he's done stuff. He's certainly committed crimes. This report was specifically addressing two questions:
1) Did Trump actively and knowingly conspire with the Russians to hack Hillary's emails in a way that constitutes a specific crime? The short answer here was, "We didn't find enough evidence to charge him with a crime."
2) Did Trump technically commit the crime of obstruction of justice? The short answer here was, "He at least kind of did, but it's a difficult legal question and I'll leave it to the Attorney General to decide."
So far, we only know those short answers, not the long answers. Either way, there's still work for Congress to do to investigate.
"Sure, I killed my wife. Just like Jeffrey Dahmer has killed a lot of people. Nothing to see here... move on."
I don't think you read it in context. That quote is specific to the charge of obstruction of justice, and the report says that Mueller gathered up the facts and declined to evaluate whether the activity constituted a crime. That's the context of "While this report does not conclude that the President committed a crime, it also does not exonerate him."
The implication is that the President did things which are debatably criminal, but Mueller felt there were enough legal/constitutional issues that it was proper to have the AG decide whether those actions constituted a crime. We can't be sure why that is without having seen the report, but we could guess:
There's been an ongoing debate because the President explicitly stated that he fired Comey to stop the Russian investigation, which is, in non-legal terms, obstructing an ongoing criminal investigation. On the other side, there have been variations of the Nixonian argument that "when the President does it, it's not illegal." Attempting to charge the President in this case would almost certainly go to the Supreme Court and create a bit of a constitutional crisis, and Mueller seems to have decided that it was simply "above his pay grade" as the special counsel, and a decision the AG should make.
It's also noteworthy that the AG decided not to prosecute on the grounds that it was a question of corrupt intent, and the President's intent couldn't be proven beyond a reasonable doubt.
Also, Trump's campaign manager, son, and son-in-law met with an agent of the Russian government for the purpose of coordinating campaign assistance. I can believe that there isn't enough evidence of a specific crime to charge them with anything, but it's still collusion and it should still be an enormous scandal.
They simply determine whether the evidence indicates a person committed a crime or not.
Not quite. They determine whether there's enough evidence to punish someone for having committed a crime. The evidence might still indicate that they've committed a crime, but there might not be enough of it.
I feel like the solution is to prevent people from spoofing caller ID. If it's a physical phone, the phone company should be able to verify the physical path to some extent. If it's a VoIP number, put some PKI in place that requires connection be signed somehow-- something like DKIM for VoIP.
It's not a complete solution, but you're not going to solve the problem if you can't create some system of accountability.
Apple has stated repeatedly they want nothing like the singularity, that desktops are inherently different than tablets or mobile devices.
Well specifically what they said was, they think an attempt to have a singular mobile + desktop UI is doomed to fail, and they're not going to do it. However, I don't think that necessarily means that they wouldn't consider a unified OS.
This news is that they're likely to move some of their desktop machines to use the same processor line as their mobile devices. Meanwhile, they recently announced that they're making it easier in Xcode to make a single app that will run on both iOS and macOS with just a change in UI.
What they may be headed towards is an ecosystem where app developers have the ability to make a single app that presents a mobile UI on a mobile device and a desktop UI on a laptop/desktop. Then, they could make phones and tablets that display an entire mobile UI when operating as a mobile device, but turn into a fully-functioning desktop machine with a desktop UI (and all the apps presenting a desktop UI) when docked.
I don't think it's just "Unix-ish". macOS is technically actual Unix, and iOS is just a modified version of macOS. As far as I know, iOS is actual Unix, based off of BSD.
It's just hobbled by being in a walled garden, unfortunately.
To my mind, most Apple users can get away with a lot less CPU power than they've got in their laptops and desktops right now
In general, Apple's approach to the market is not really to put out commodity computers that compete on the metric of raw computing power. The anti-Apple crowd never seems to grasp that. It's more helpful to think of them as producing appliance computers as a whole product, aimed at specific use-cases. Granted, one of the main use-cases is general business and home productivity computing, but that's still a particular use case.
It may sound like I'm just spouting nonsense, but here's my point: They're selling appliances to let people check their email, browse the Internet, and edit Word documents. None of that requires much processing power, relative to how much power is in a current-generation CPU. Apple doesn't need to push the envelope in high-performance CPUs. Instead, they tend to focus on things like a convenient and attractive form factor, making things lightweight, making the battery last longer, having a pretty built-in screen, etc.
If they can deliver on those things somehow better with ARM CPUs, I wouldn't be surprised if they did that even if it meant less raw power, and I wouldn't be surprised if their fans were happy with the result.
I know, it's so stupid to think you'd travel around the universe in a planet. If you were a real pro, you'd launch the entire solar system in the direction you want to go. That way, you bring a nice power source with you.
Again, I think that's a misunderstanding of genetics and evolution. It's not like genes are simply advantageous and disadvantageous, and advantageous genes are spread and disadvantageous genes die out. Genes might have multiple effects which might be advantageous or disadvantageous based on the environment the organism finds itself in.
But also, being advantageous or disadvantageous isn't by itself the issue. It had to be advantageous enough to increase your chances of surviving long enough to reproduce, or disadvantageous enough to decrease you chances of surviving long enough to reproduce. If it's just sort of theoretically disadvantageous to the human race in general, but you can still live to maturity and have offspring, and that gene will still get passed on. Evolution does not refine people to the point of perfection, it merely to the point of enabling procreation.
Do you mean, if it's useful why haven't we all evolved to have it? You do know that evolution doesn't work that way, right?
Any person would be in jail, but when companies do things that would put a person in jail, the worst thing that happens to corporations is that they get fined. More often, lawmakers hold meaningless hearings and then say, "Let the invisible hand sort this out." One of the main features of corporations is to reduce liability for everyone involved, but that also tends to turn companies collectively into risk-taking psychopaths.
All it takes for reform is that we elect people interested in reform.
I mean, it still has a cut-out for the front-facing camera. It's smaller, which is nice, but it looks like it might just be a camera instead of the multiple sensors that Apple puts in the notch.
So it's not like they totally avoided the problem and didn't do anything like the notch. In both cases, the manufacturer was trying to find a way to have an edge-to-edge screen while also putting front-facing sensors. In both cases, they decided to resolve it by having a cut-out in the screen. Apple did a larger cut-out with a multiple sensors, centered at the very top of the screen. Samsung did a smaller cut-out with fewer sensors, not connected to the top of the screen, offset to one side. It's kind of just a different design choice of how to implement the same basic thing.
I don't think there's any doubt that one of the things that will happen is, iOS developers will make lazy macOS ports of their apps, creating junky apps that look and behave like mobile apps running incongruously on a desktop OS. I don't see that as a problem in and of itself. Give developers tools, and let them do what they want. Some will make junk, but that's fine.
Just so long as Apple doesn't use it as an excuse to further limit what developers can do on their OS. We need iOS to become more open like a desktop operating system, not for the macOS to become more closed off like a mobile operating system.