There was a case on Slashdot a few years ago where someone had a camera and microphone in their porch, which they used to record the police. They were charged under wiretapping legislation. I don't know what happened to the case in the end, but if they were successfully prosecuted then I can imagine that Bezos and Pichai would be liable for a few million counts of the same.
The problem is signal to noise ratio. Consider email spam: the cost of sending spam to a million people is approximately zero. The cost of sending a thoughtful email to one person is, at the very minimum, a few minutes of real-human time. If you lower the cost of speaking sufficiently then the signal to noise ratio tends towards zero unless you actively filter.
If you're starting with -Os, then your baseline is optimising for size, so presumably that's what you care about. To beat it you need to either remove instructions entirely, replace multiple instructions with a shorter sequence, or pick instructions with shorter encodings. All of these are things that are really easy to automate. Most of the bloat comes from ABI constraints, and if you pass the right flags the compiler will ignore these (e.g. omitting the frame pointer).
If you're optimising for speed, then start with -O3. At this point the compiler has already done things like loop rotation, common subexpression elimination, and autovectorisation. You might be able to beat it, but even understanding what the code is doing for a nontrivial example is quite hard.
The craziness that is C++ locales, for example, need not concern you because you won't use it.
Are you sure? I have a version of libc++ that works in the FreeBSD kernel, but I needed to do some quite drastic surgery to remove all of the locale stuff (which I really don't want in the kernel - anything that needs localisation should be done in userspace). A lot of the standard library depends on it indirectly and so I needed a lot of stubs to even let the standard library build.
The C++ standard library could really benefit from some better modularity and layering. The core ADTs, the threading library, simple string handling, localised Unicode string handling, and so on should all be separated out and the dependencies between them made explicit. Unfortunately, the C++ standards committee is philosophically opposed to subsetting, which means that instead of a handful of standard subsets we have hundreds of per-project subsets.
Not the GP, but often airlines turn off most of the in-flight entertainment system on approach and leave it off after landing. If you'd planned on watching a film, weren't able to watch the last 30 minutes, but still had to sit in uncomfortable plane seat for that time, I can imagine that you'd be cranky.
At Cambridge, most of the systems use a single sign on system and provide tokens for the services, so no one sees your password except the authentication system. They've now integrated that with Office365, so Microsoft doesn't see the password when you log in (when they first set it up, they accidentally sent the entire password database to Microsoft, in plain text. Ooops). It ought to be easy to tell people 'only ever enter your password into raven.cam.ac.uk'. Unfortunately, they also:
Set the flag in the password field that prevents password managers from caching it (I think most browsers now ignore this), which prevented the obvious clue of 'Hmm, why is this not autofilled, maybe something bad is happening here'.
Use the same password for email, so every single mail client also contains a copy of the master password for that user's account, rather than something like an OAuth token generated for that device and granting access only to email.
Have a bunch of new systems written by muppets (such as the new payroll system) that ask for the password and don't integrate with the SSO system, so they require people to enter the password into that site (giving that password to Payroll gives Payroll access to everything, including the student information database - I'm astonished that this is allowed under the GDPR).
I filed numerous bugs against these systems while I was there. None of them were fixed.
SSH security leaves a lot to be desired. Do your users all use ssh-agent? If not, they're probably using ssh keys with no passphrase, which can be stolen by anyone who gets read access to their local filesystem. At that point, the attacker can gain access to your system. If they do use ssh-agent, then the attacker needs to gain debug privilege on their local machine, but that's also not too hard. ssh-agent has no protection against a compromised host OS, for example, unless you set up PAM on your systems to require a second factor such as a U2F key (there's no SGX version of ssh-agent, for example).
If their private key is compromised, ssh doesn't have a global revocation mechanism, so you need to go and find all of the places where an authorized_keys file contains their public key. What is your revocation policy? Do you have a simple way for people to submit a compromised public key and automatically revoke it across your entire system?
By default (though, thankfully, now not the only option) the known_hosts file contains a good list of all systems that an attacker should look at next. Do you require that your users turn on the feature that stores hashes of the machines, or does any compromise of one of your users' systems lead immediately to the attacker knowing that they have compromised a key that gains access to your system.
but similarly if my purchased disc fails I have nothing too.
Unless you back it up. A typical DVD is 6-7GB. You can fit a couple of hundred of them (using dvdbackup, just strip the CSS and store the VOBs, no reencoding) on a 1TB hard disk. 4TB NAS disks cost about £100, add a second one for mirroring and that's about 60p for the space to back up one DVD. As an added bonus, you can then watch it without ever getting it out of the box and so you can store the shiny disks somewhere safe and use them as the backups.
Exactly. Netflix never offered their DVD-by-mail service in the UK and Amazon bought and killed the largest company that did, but Cinema Paradiso still exists and has a wide library. I also subscribe to Netflix, but we use Cinema Paradiso to get recent films and TV shows that aren't on Netflix. This seems to be getting worse as content producers are worried about Amazon and Netflix's increasing domination of the distribution channel. Meanwhile, pretty much every film and every TV series that has a moderately large audience ends up on DVD.
That's a terrible idea in a multi-user environment, because when the Apache process dies any other user can open that port (they may even open it accidentally) and now they get all of your web server traffic.
On modern UNIX systems; however, it is possible to grant the permission to open specific low ports. For example, on FreeBSD the portacl MAC framework policy can control this. On Linux SELinux can do the same thing.
Last time I looked, about 10 years ago, Psi had all of those features and many more. I tried Conversation a couple of years ago and found it somewhat lacking in comparison to the features that desktop X11 XMPP clients had had a decade earlier.
Your claim, precisely as stated, appears to be true but, per your link, that doesn't mean that the watermarking hasn't been broken in other ways. In fact, citation 16 regarding DVD-Ranger CinEx appears to do precisely that: detect the signal and then remove it.
The Amazon technique sounds like exactly the same crap that you get from a lot of machine-learning researchers doing security work: they don't think about an adaptive adversary. There's an entire field of adversarial machine learning that works by training a machine-learning system on the inputs and outputs of another: if you can train a neural network to insert and recognise these watermarks, can you train another one to recognise and remove them? If you haven't even tried that, it's likely that an attacker will be able to.
I realise you're trying to be sarcastic, but you're correct. In the per-capita GDP rankings, the UK is 25th at $43,620, China is 78th at $16,624 (2017 IMF numbers). The USA is 11th at $59,495.
It's similar in the UK. Wales and Scotland made the switch almost a decade ago, England followed a few years later. You can still buy a disposable plastic bag for 5p, but you don't get one for free. Plastic bag usage dropped 85% since that law was introduced. Here, most people take reuseable bags (not the low-quality ones that shops sell at the checkout, something a bit more sturdy). A lot of companies have realised that this is a good marketing opportunity and now hand out sturdy canvas bags at recruiting events and similar.
Since we're playing that game, I live in a country that banned free single-use bags. You can still buy them, but they're 5p each. I occasionally see someone buy one, but it's very rare and plastic bag usage has dropped 85% since this law was introduced, after decades of usage increasing year-on-year.
Oh, and while most shops do sell thicker plastic bags that you can trade in for a replacement when they wear out, most people here carry their shopping in something a bit more sturdy (fabric, canvas or higher-quality plastic bags).
The USA started phasing out incandescent lightbulbs about 10 years after I replaced all of the ones in my house with brighter (and significantly lower power) CFLs, which saved me about as much money in electricity during their first two months of operation as they cost to buy. If the ones you could buy were worse, then that says a lot more about your local supply chain and access to modern technology than it does about the regulation.
As those bulbs die, I'm replacing them with LEDs, which are a bit brighter for around a quarter to a third of the power (around 10% of an equivalent incandescent). It's much less of an electricity saving - going from 60W to 12W makes more of a difference than going from 12W to 4W - but it's still probably a cost saving over the course of 1-2 years and they're expected to last at least 5-10.
That sounds like it's only a problem if you're putting unwrapped food in them.
Off topic: I didn't get a message that you're replied to my post and can no longer find the Slashdot message settings. Have these gone away? Are the new owners intentionally trying to prevent meaningful conversation on this site? That would explain why the standard of comments has dropped a lot recently...
And, to make that worse, the stuff that's supposed to be there is quite big, but a lot of the plastic has broken down to be particles of a similar size to grains of sand. Designing a net that will catch small things but let large things pass is very hard!
A natural fear, since casinos RELY on those rates, and the magic of statistics, to always be profitable, even when making payouts.
It's worse than that. In a lot of jurisdictions, the payout rates are mandated by law and there can be serious legal consequences if the advertised payout rates are not the real ones.
You don't know the target in advance. You know where the target is going to start from and where it's going to end up, and you probably know when it's going to start. You don't know what the atmospheric conditions at the time are going to be and how they're going to affect speed and trajectory.
At the speeds that these things travel, there's no such thing as a near miss. If your interceptor explodes a couple of metres away, then by the time the explosion reaches the target's position at the time of the explosion, the target will be long gone. You'll sometimes hear this kind of interceptor (including air-to-air missiles) referred to as 'hitiles' (which is a horrible word and, thankfully, seems to be going out of fashion) for this reason.
Once you can actually hit a fast-moving target, you've solved one of the two difficult problems in ICBM interception. The second one, calculating the trajectory for the interceptor fast enough, boils down to available computational resources and those are relatively easy to improve.
Of course, this is assuming that the target isn't actively trying to evade the interceptor, and that's why they call it an arms race...
This may not help with the other bits of Teams, but SharePoint shares (including the files tab in Teams) can also be used as a OneDrive for Business share. If you hit the 'Sync' button in the SharePoint share, it will sync the entire share in the same way that OneDrive does (things are loaded lazily, but then they're local and they're sync'd in the background). This also works directly from Office, so files open as fast as if they were local, but you get all of the collaborative editing stuff via the desktop version of Office.
At least on Windows, a Miracast display looks just like a wired display to the rest of the system. PowerPoint automatically puts you into the mode where Windows 10 notifications are silenced, but a lot of other applications have their own ad-hoc notification mechanisms (Thunderbird, I'm looking at you!) and don't respect the silence command.
At work, you can often see a few meeting rooms' Miracast targets from one room. This isn't a problem except in one room that you would expect to support Miracast, but doesn't, and is right next to the lab director's office - it's very easy to accidentally project onto his wall display and wonder why nothing is showing up in the room that you're in...
Does it go into general revenue? In the UK, shops are required to charge 5p for plastic bags (with a few exceptions) but that money isn't levied as taxation, instead shops are required to donate it to a registered charity of their choice. This removes any profit incentive from both the shops and the government. It took a little while to get used to, but now I carry a reusable bag, which is a lot more robust than a plastic carrier and still going strong after hundreds of uses.
Not Oracle, but Sun has a lot of history in closely related spaces. The original Java platform (back when Java was called Green) was the 7*, a handheld computer that ran a modified Solaris that supported execute in place and ran happily with a 32-bit SPARC and 1MB of RAM. The vast majority of pre-iPhone smartphones and featurephones included J2ME, which (unlike J2SE) required a license fee from each phone maker.
This is the main reason that Sun was unhappy with Android. They'd been receiving royalties from pretty much every phone to be able to use Java and then suddenly Google came along with a Java implementation that didn't require anyone to pay Sun. Worse, as with Microsoft's J++, it wasn't a fully conformant implementation of Java - it did both subsetting and supersetting, so arbitrary Java code doesn't work on Android and arbitrary Android Java code doesn't work on other JVMs.
Slashdot Mirror
There was a case on Slashdot a few years ago where someone had a camera and microphone in their porch, which they used to record the police. They were charged under wiretapping legislation. I don't know what happened to the case in the end, but if they were successfully prosecuted then I can imagine that Bezos and Pichai would be liable for a few million counts of the same.
The problem is signal to noise ratio. Consider email spam: the cost of sending spam to a million people is approximately zero. The cost of sending a thoughtful email to one person is, at the very minimum, a few minutes of real-human time. If you lower the cost of speaking sufficiently then the signal to noise ratio tends towards zero unless you actively filter.
If you're starting with -Os, then your baseline is optimising for size, so presumably that's what you care about. To beat it you need to either remove instructions entirely, replace multiple instructions with a shorter sequence, or pick instructions with shorter encodings. All of these are things that are really easy to automate. Most of the bloat comes from ABI constraints, and if you pass the right flags the compiler will ignore these (e.g. omitting the frame pointer).
If you're optimising for speed, then start with -O3. At this point the compiler has already done things like loop rotation, common subexpression elimination, and autovectorisation. You might be able to beat it, but even understanding what the code is doing for a nontrivial example is quite hard.
The craziness that is C++ locales, for example, need not concern you because you won't use it.
Are you sure? I have a version of libc++ that works in the FreeBSD kernel, but I needed to do some quite drastic surgery to remove all of the locale stuff (which I really don't want in the kernel - anything that needs localisation should be done in userspace). A lot of the standard library depends on it indirectly and so I needed a lot of stubs to even let the standard library build.
The C++ standard library could really benefit from some better modularity and layering. The core ADTs, the threading library, simple string handling, localised Unicode string handling, and so on should all be separated out and the dependencies between them made explicit. Unfortunately, the C++ standards committee is philosophically opposed to subsetting, which means that instead of a handful of standard subsets we have hundreds of per-project subsets.
Not the GP, but often airlines turn off most of the in-flight entertainment system on approach and leave it off after landing. If you'd planned on watching a film, weren't able to watch the last 30 minutes, but still had to sit in uncomfortable plane seat for that time, I can imagine that you'd be cranky.
I filed numerous bugs against these systems while I was there. None of them were fixed.
SSH security leaves a lot to be desired. Do your users all use ssh-agent? If not, they're probably using ssh keys with no passphrase, which can be stolen by anyone who gets read access to their local filesystem. At that point, the attacker can gain access to your system. If they do use ssh-agent, then the attacker needs to gain debug privilege on their local machine, but that's also not too hard. ssh-agent has no protection against a compromised host OS, for example, unless you set up PAM on your systems to require a second factor such as a U2F key (there's no SGX version of ssh-agent, for example).
If their private key is compromised, ssh doesn't have a global revocation mechanism, so you need to go and find all of the places where an authorized_keys file contains their public key. What is your revocation policy? Do you have a simple way for people to submit a compromised public key and automatically revoke it across your entire system?
By default (though, thankfully, now not the only option) the known_hosts file contains a good list of all systems that an attacker should look at next. Do you require that your users turn on the feature that stores hashes of the machines, or does any compromise of one of your users' systems lead immediately to the attacker knowing that they have compromised a key that gains access to your system.
but similarly if my purchased disc fails I have nothing too.
Unless you back it up. A typical DVD is 6-7GB. You can fit a couple of hundred of them (using dvdbackup, just strip the CSS and store the VOBs, no reencoding) on a 1TB hard disk. 4TB NAS disks cost about £100, add a second one for mirroring and that's about 60p for the space to back up one DVD. As an added bonus, you can then watch it without ever getting it out of the box and so you can store the shiny disks somewhere safe and use them as the backups.
Exactly. Netflix never offered their DVD-by-mail service in the UK and Amazon bought and killed the largest company that did, but Cinema Paradiso still exists and has a wide library. I also subscribe to Netflix, but we use Cinema Paradiso to get recent films and TV shows that aren't on Netflix. This seems to be getting worse as content producers are worried about Amazon and Netflix's increasing domination of the distribution channel. Meanwhile, pretty much every film and every TV series that has a moderately large audience ends up on DVD.
If you're using php-fpm, why bother with Apache? Nginx configuration for that scenario is a lot simpler.
That's a terrible idea in a multi-user environment, because when the Apache process dies any other user can open that port (they may even open it accidentally) and now they get all of your web server traffic.
On modern UNIX systems; however, it is possible to grant the permission to open specific low ports. For example, on FreeBSD the portacl MAC framework policy can control this. On Linux SELinux can do the same thing.
Last time I looked, about 10 years ago, Psi had all of those features and many more. I tried Conversation a couple of years ago and found it somewhat lacking in comparison to the features that desktop X11 XMPP clients had had a decade earlier.
Your claim, precisely as stated, appears to be true but, per your link, that doesn't mean that the watermarking hasn't been broken in other ways. In fact, citation 16 regarding DVD-Ranger CinEx appears to do precisely that: detect the signal and then remove it.
The Amazon technique sounds like exactly the same crap that you get from a lot of machine-learning researchers doing security work: they don't think about an adaptive adversary. There's an entire field of adversarial machine learning that works by training a machine-learning system on the inputs and outputs of another: if you can train a neural network to insert and recognise these watermarks, can you train another one to recognise and remove them? If you haven't even tried that, it's likely that an attacker will be able to.
I realise you're trying to be sarcastic, but you're correct. In the per-capita GDP rankings, the UK is 25th at $43,620, China is 78th at $16,624 (2017 IMF numbers). The USA is 11th at $59,495.
It's similar in the UK. Wales and Scotland made the switch almost a decade ago, England followed a few years later. You can still buy a disposable plastic bag for 5p, but you don't get one for free. Plastic bag usage dropped 85% since that law was introduced. Here, most people take reuseable bags (not the low-quality ones that shops sell at the checkout, something a bit more sturdy). A lot of companies have realised that this is a good marketing opportunity and now hand out sturdy canvas bags at recruiting events and similar.
Oh, and while most shops do sell thicker plastic bags that you can trade in for a replacement when they wear out, most people here carry their shopping in something a bit more sturdy (fabric, canvas or higher-quality plastic bags).
The USA started phasing out incandescent lightbulbs about 10 years after I replaced all of the ones in my house with brighter (and significantly lower power) CFLs, which saved me about as much money in electricity during their first two months of operation as they cost to buy. If the ones you could buy were worse, then that says a lot more about your local supply chain and access to modern technology than it does about the regulation.
As those bulbs die, I'm replacing them with LEDs, which are a bit brighter for around a quarter to a third of the power (around 10% of an equivalent incandescent). It's much less of an electricity saving - going from 60W to 12W makes more of a difference than going from 12W to 4W - but it's still probably a cost saving over the course of 1-2 years and they're expected to last at least 5-10.
Off topic: I didn't get a message that you're replied to my post and can no longer find the Slashdot message settings. Have these gone away? Are the new owners intentionally trying to prevent meaningful conversation on this site? That would explain why the standard of comments has dropped a lot recently...
And, to make that worse, the stuff that's supposed to be there is quite big, but a lot of the plastic has broken down to be particles of a similar size to grains of sand. Designing a net that will catch small things but let large things pass is very hard!
A natural fear, since casinos RELY on those rates, and the magic of statistics, to always be profitable, even when making payouts.
It's worse than that. In a lot of jurisdictions, the payout rates are mandated by law and there can be serious legal consequences if the advertised payout rates are not the real ones.
You don't know the target in advance. You know where the target is going to start from and where it's going to end up, and you probably know when it's going to start. You don't know what the atmospheric conditions at the time are going to be and how they're going to affect speed and trajectory.
At the speeds that these things travel, there's no such thing as a near miss. If your interceptor explodes a couple of metres away, then by the time the explosion reaches the target's position at the time of the explosion, the target will be long gone. You'll sometimes hear this kind of interceptor (including air-to-air missiles) referred to as 'hitiles' (which is a horrible word and, thankfully, seems to be going out of fashion) for this reason.
Once you can actually hit a fast-moving target, you've solved one of the two difficult problems in ICBM interception. The second one, calculating the trajectory for the interceptor fast enough, boils down to available computational resources and those are relatively easy to improve.
Of course, this is assuming that the target isn't actively trying to evade the interceptor, and that's why they call it an arms race...
This may not help with the other bits of Teams, but SharePoint shares (including the files tab in Teams) can also be used as a OneDrive for Business share. If you hit the 'Sync' button in the SharePoint share, it will sync the entire share in the same way that OneDrive does (things are loaded lazily, but then they're local and they're sync'd in the background). This also works directly from Office, so files open as fast as if they were local, but you get all of the collaborative editing stuff via the desktop version of Office.
At least on Windows, a Miracast display looks just like a wired display to the rest of the system. PowerPoint automatically puts you into the mode where Windows 10 notifications are silenced, but a lot of other applications have their own ad-hoc notification mechanisms (Thunderbird, I'm looking at you!) and don't respect the silence command.
At work, you can often see a few meeting rooms' Miracast targets from one room. This isn't a problem except in one room that you would expect to support Miracast, but doesn't, and is right next to the lab director's office - it's very easy to accidentally project onto his wall display and wonder why nothing is showing up in the room that you're in...
Does it go into general revenue? In the UK, shops are required to charge 5p for plastic bags (with a few exceptions) but that money isn't levied as taxation, instead shops are required to donate it to a registered charity of their choice. This removes any profit incentive from both the shops and the government. It took a little while to get used to, but now I carry a reusable bag, which is a lot more robust than a plastic carrier and still going strong after hundreds of uses.
Not Oracle, but Sun has a lot of history in closely related spaces. The original Java platform (back when Java was called Green) was the 7*, a handheld computer that ran a modified Solaris that supported execute in place and ran happily with a 32-bit SPARC and 1MB of RAM. The vast majority of pre-iPhone smartphones and featurephones included J2ME, which (unlike J2SE) required a license fee from each phone maker.
This is the main reason that Sun was unhappy with Android. They'd been receiving royalties from pretty much every phone to be able to use Java and then suddenly Google came along with a Java implementation that didn't require anyone to pay Sun. Worse, as with Microsoft's J++, it wasn't a fully conformant implementation of Java - it did both subsetting and supersetting, so arbitrary Java code doesn't work on Android and arbitrary Android Java code doesn't work on other JVMs.