The blood alcohol level is a red herring. It correlates with impairment, but a number of other factors also affect it. The test should be for reactions and situational awareness. If you fail for any reason, then you should be prevented from driving. If you fail and also have been taking drugs that are known to cause this kind of impairment, then you might get some extra penalty.
I wouldn't find it too hard to believe. The last two vaguely-mainstream tech conferences I went to were FOSDEM and the LLVM DevMeeting. Both of those were at least 90% white male. Assuming that the talk submitters were from the same approximate demographic as the attendees, it's quite plausible that they made up all of the submissions. It's very plausible that they made up the best submissions: if 90% of submissions have any arbitrary characteristic that doesn't correlate with ability then it's very easy to end up with the top 10% all having that characteristic (or the bottom 10%, or any other 10%).
And if they top 10% of proposed speakers were all white males, would you want them to put some token minority people in, bumping a better speaker from their slot? It would most likely be counterproductive: if everyone sees that the worst talk is the only person with a given orthogonal characteristic (gender, skin colour, whatever) then they're likely to subconsciously associate that characteristic with technical inferiority.
Just take a look at the forum discussions of pretty much any open source project
Try the mailing lists. I participate in a few open source projects, and none of them use forums as their primary communication mechanism between developers. Forums tend to be for users or fans of the project, and that's a much wider audience.
and the default is to assume that the poster is a male. This creates a very unfriendly environment for women.
Really? The default is generally to use male pronouns, because that's the correct style for people of unknown gender in most dialects of English, but I don't think I remember reading any message where the gender of anyone was assumed in a post, because I can't think of a single post where the gender of the author would have come up as a topic.
That's pretty difficult to do. The people on the conference committee generally have a pretty good idea of who is working on what, so when someone presents an anonymised paper describing a certain bit of work, then you can generally have a pretty good idea of who the person is based on the work. The anonymity only really helps people who don't have an established track record from being identified with respect to others in the same position: it's still easy to identify the ones who don't have that record.
I think my proposal for the protest against the UK ID cards would work in this case: create a bright yellow holder for the badge in the shape of a Star of David and wear it on your arm. Get some local press to photograph you wearing it.
To be fair to Linux, glibc is not in the Linux kernel. That's why it's important to say GNU/Linux: because Drepper deserves the blame at least as much as Linus. Android, for example, is Linux and uses a FreeBSD libc derivative instead of glibc.
I've found that with all of the music that I've considered buying on iTunes. Most things seem to be 20-50% cheaper (including delivery) if you buy the shiny plastic disc than if you get them from iTunes. I guess that's the price that you pay for immediacy. Or, I suppose if listen to bands that suck and only produce one track per album that's worth listening to then it's cheaper to buy a single track on iTunes than an entire album anywhere...
The songs played on the radios were regarded by the bands as adverts (see: payola), and as such they didn't want to play the whole album because they wanted people to have to buy it to listen to the whole thing. The individual songs played on the radio were regarded as previews, not as complete works in themselves. In contrast, a downloaded track is regarded as a complete work by the band. No one complains that film previews contain scenes out of order, or that book previews only contain the first chapter, but the creators of both would strongly object to the idea of selling films by the scene[1] or books by the chapter.
[1] Certain Hollywood companies, however, would be very much in favour of this if they thought that they could get people to pay more that way.
I'd say the real reason Google asks the sort of questions it does is because it's the only way the company has found to get a handle on what it really wants to find out about candidates: Problem-solving ability.
There are two problems with this. The first is that even Google admits that it doesn't work. The most successful people inside Google are the ones that had one or more negative reports during the interview process: exactly the ones that would be rejected in the normal process. Second, Google isn't short of people who can solve problems, and being able to solve problems isn't even an especially rare skill. Google is short of people who can identify the problems that are worth solving, and the interview process does nothing to address that.
Treaties are not permanent. If all (or even most) parties involve decide that they are no longer relevant, then they are renegotiated. This happens all the time. The Berne convention is currently a problem, but if the USA were to decide to change policy on copyright law then negotiating a new revision would be relatively easy, especially as most other countries would benefit from shorter copyright.
Taxing copyright would have interesting effects on open source. If you're not commercially exploiting the copyright, would you be able to afford the tax? Or would you just let stuff lapse into the public domain and then have some company decide to incorporate it into proprietary products without even crediting you (in most of the US, copyright law does not include a notion of natural rights). How would this interact with the Berne Convention? You have to treat foreign-registered copyrights the same way as local ones, would this mean that you'd try to tax everyone? If not, then you'd just see all of the big companies transferring all of their copyrights to a wholly-owned subsidiary in a different jurisdiction and paying them for the use, giving big companies yet another way to turn local income into off-shore income and reduce the amount of tax that they paid. Would it only apply to registered copyright, or would you have to pay taxes on every email, every Slashdot post, and every other creative work that is automatically copyrighted?
If you don't pay you taxes I'm pretty sure you will lose your house and other property
And you only have a house and other property because society collectively decides to respect your property rights and not just take all of your stuff because it's convenient to do so. And when individual members of society decide that they disagree with this, it is government-funded institutions such as the police and prisons that are responsible for apprehending them. If you don't pay taxes, then you are choosing to opt out of this protection. You can always try heavily arming yourself to protect your property and see how well that works out.
Really do seem to know what they're doing, and are very proactive with their security.
The security team and cluster admin has also been working very hard over the past few months to partition the FreeBSD cluster a lot better. If this attack had happened in a month or two, you wouldn't be hearing about it because nothing of value would have been compromised. The attack was against the legacy package building infrastructure, which is due to be retired soon. It was able to get access to more systems because it had the developers' home directories mounted (this isn't be the case with the new system, which is completely isolated). They're also rolling out the new audit logging daemon, which should mean that this sort of thing would be detected much sooner in the future.
We were lucky. The attackers seemed not to know what they'd found. There was an (apparently) automated scan for ssh keys (which have now all been revoked) and that's about it. They seemed to be trying to add the machine to a botnet, rather than to attack it directly. We believe that they got access via a compromised developer VM which had an ssh key that connected to the cluster (for doing svn+ssh commits), so it's possible that it was an entirely automated attack. They attempted to run a load of Linux admin commands and apparently gave up when they didn't work. As far as we can tell, they didn't actually modify anything (although, of course, the compromised machines are offline pending imaging for forensic analysis and clean reinstalls or replacement, just in case). The announcement tells you not to trust any of the things that we know that they could have touched, but it currently looks like there's a very low probability that they did touch anything. For example, they might have modified ports / base cvs, but the top of the tree is identical to the svn tree (which had off-site backups, and has had every recent commit manually audited) and so they'd have had to insert something bad and then remove it in a subsequent CVS version, and that seems unlikely. Verifying cvs is pretty hard, which is part of the reason why we're encouraging everyone to move to svn now.
I had an account on a machine in the same rack as kernel.org, and that machine was taken away for forensic analysis and still isn't back. Apparently (I don't do security research, but I work on a team that does) kernel.org contained the world's best collection of rootkits found to date, which was incredibly useful to people doing work in this area.
The idea of a rental is problematic anyway for digital content. You loan me a physical object, you no longer have it. I use it and then give it back, and you have it again. With an electronic version, you give me a copy of an object, but still have the original. Later, I delete the copy and you still have the original. The problem is trying to use a metaphor that simply doesn't make sense as the basis for a business model.
The service of value that Netflix offers is not rental, it's access to an ever-increasing library. From their perspective, it should be great if someone downloads something and watches it locally the second time, because they've reduced Netfilx's bandwidth costs. You can't reasonably download their entire catalog (even limiting people to downloading 30 or 60 hours a month would satisfy most people and prevent this easily), the thing you're paying for is a service that means that when you want to watch something you can easily get something that you want to watch. The more technical obstacles that they put in the way of this, the less valuable their service becomes.
I started ripping my DVDs a while ago. I bought 3 2TB disks just before the floods and put them in a RAID-Z array, giving me 4GB of usable space. That's enough for full backups of my laptop (with periodic snapshots) and for my DVD collection (probably about 1TB when I've ripped them all). The NAS box runs FreeBSD and is connected to my projector and speakers. Because none of the streaming services use open standards for their services, I can't use them from this machine, so they don't get my money even though I'm right in the middle of the early-adopter demographic for video stuff (I bought a DVD player when they were still expensive, I have a projector and surround sound system for watching them, I subscribed to the first DVD-in-the-mail service to appear in this country as soon as it appeared).
The more ethical pizzeria already has enough employees to meet its current demand, but if people switch from Papa Johns to them, then they will need more employees (possibly even a new branch) to meet the increased demand.
If you actually did pay for yours, then no one would complain. The problem is when you dump all of your externalities on everyone else and expect them to pay for yours too.
Secondly, when one looks at the amounts of malware available for each platform it does become clear that the 'walled garden' does seem to have an affect on device security.
Until you factor in the fact that iOS uses the MAC framework from FreeBSD to enforce a pretty restrictive access, and has a solid centralised update system, while Android uses a fragile, hacky chroot()-based system and only gets updates for core system libraries on a few devices / carriers. Then it becomes a lot less clear.
These are people who have already opted in to receive messages from you. A fair comparison is people who have subscribed to your mailing list, RSS feed, or whatever. If it's costing you three cents per subscriber to your mailing list, then you're probably doing something wrong.
It's not that uncommon. You have lots of blood vessels close to the surface there, so drugs get into your blood stream quickly. There was a story recently about people dying from vodka suppositories - so much alcohol goes into their blood so quickly that it's fatal.
Actually, in all seriousness, my gay friends tell me that the biggest down side is more or less the opposite of what I said: there's a significant chance that anyone that they're attracted to won't even be interested in people the same gender as them, let alone them specifically. If you think that gay people make up around 5% of the population, there's a one in twenty chance that someone they find attractive will also be gay. And if they are, the probability of them being gay, single, and attracted to you is even lower. At least for heterosexuals start with a nineteen in twenty chance before adding in the other factors...
Slashdot Mirror
The blood alcohol level is a red herring. It correlates with impairment, but a number of other factors also affect it. The test should be for reactions and situational awareness. If you fail for any reason, then you should be prevented from driving. If you fail and also have been taking drugs that are known to cause this kind of impairment, then you might get some extra penalty.
And if they top 10% of proposed speakers were all white males, would you want them to put some token minority people in, bumping a better speaker from their slot? It would most likely be counterproductive: if everyone sees that the worst talk is the only person with a given orthogonal characteristic (gender, skin colour, whatever) then they're likely to subconsciously associate that characteristic with technical inferiority.
Just take a look at the forum discussions of pretty much any open source project
Try the mailing lists. I participate in a few open source projects, and none of them use forums as their primary communication mechanism between developers. Forums tend to be for users or fans of the project, and that's a much wider audience.
and the default is to assume that the poster is a male. This creates a very unfriendly environment for women.
Really? The default is generally to use male pronouns, because that's the correct style for people of unknown gender in most dialects of English, but I don't think I remember reading any message where the gender of anyone was assumed in a post, because I can't think of a single post where the gender of the author would have come up as a topic.
That's pretty difficult to do. The people on the conference committee generally have a pretty good idea of who is working on what, so when someone presents an anonymised paper describing a certain bit of work, then you can generally have a pretty good idea of who the person is based on the work. The anonymity only really helps people who don't have an established track record from being identified with respect to others in the same position: it's still easy to identify the ones who don't have that record.
I think my proposal for the protest against the UK ID cards would work in this case: create a bright yellow holder for the badge in the shape of a Star of David and wear it on your arm. Get some local press to photograph you wearing it.
To be fair to Linux, glibc is not in the Linux kernel. That's why it's important to say GNU/Linux: because Drepper deserves the blame at least as much as Linus. Android, for example, is Linux and uses a FreeBSD libc derivative instead of glibc.
I've found that with all of the music that I've considered buying on iTunes. Most things seem to be 20-50% cheaper (including delivery) if you buy the shiny plastic disc than if you get them from iTunes. I guess that's the price that you pay for immediacy. Or, I suppose if listen to bands that suck and only produce one track per album that's worth listening to then it's cheaper to buy a single track on iTunes than an entire album anywhere...
The songs played on the radios were regarded by the bands as adverts (see: payola), and as such they didn't want to play the whole album because they wanted people to have to buy it to listen to the whole thing. The individual songs played on the radio were regarded as previews, not as complete works in themselves. In contrast, a downloaded track is regarded as a complete work by the band. No one complains that film previews contain scenes out of order, or that book previews only contain the first chapter, but the creators of both would strongly object to the idea of selling films by the scene[1] or books by the chapter.
[1] Certain Hollywood companies, however, would be very much in favour of this if they thought that they could get people to pay more that way.
I'd say the real reason Google asks the sort of questions it does is because it's the only way the company has found to get a handle on what it really wants to find out about candidates: Problem-solving ability.
There are two problems with this. The first is that even Google admits that it doesn't work. The most successful people inside Google are the ones that had one or more negative reports during the interview process: exactly the ones that would be rejected in the normal process. Second, Google isn't short of people who can solve problems, and being able to solve problems isn't even an especially rare skill. Google is short of people who can identify the problems that are worth solving, and the interview process does nothing to address that.
Treaties are not permanent. If all (or even most) parties involve decide that they are no longer relevant, then they are renegotiated. This happens all the time. The Berne convention is currently a problem, but if the USA were to decide to change policy on copyright law then negotiating a new revision would be relatively easy, especially as most other countries would benefit from shorter copyright.
Taxing copyright would have interesting effects on open source. If you're not commercially exploiting the copyright, would you be able to afford the tax? Or would you just let stuff lapse into the public domain and then have some company decide to incorporate it into proprietary products without even crediting you (in most of the US, copyright law does not include a notion of natural rights). How would this interact with the Berne Convention? You have to treat foreign-registered copyrights the same way as local ones, would this mean that you'd try to tax everyone? If not, then you'd just see all of the big companies transferring all of their copyrights to a wholly-owned subsidiary in a different jurisdiction and paying them for the use, giving big companies yet another way to turn local income into off-shore income and reduce the amount of tax that they paid. Would it only apply to registered copyright, or would you have to pay taxes on every email, every Slashdot post, and every other creative work that is automatically copyrighted?
If you don't pay you taxes I'm pretty sure you will lose your house and other property
And you only have a house and other property because society collectively decides to respect your property rights and not just take all of your stuff because it's convenient to do so. And when individual members of society decide that they disagree with this, it is government-funded institutions such as the police and prisons that are responsible for apprehending them. If you don't pay taxes, then you are choosing to opt out of this protection. You can always try heavily arming yourself to protect your property and see how well that works out.
Really do seem to know what they're doing, and are very proactive with their security.
The security team and cluster admin has also been working very hard over the past few months to partition the FreeBSD cluster a lot better. If this attack had happened in a month or two, you wouldn't be hearing about it because nothing of value would have been compromised. The attack was against the legacy package building infrastructure, which is due to be retired soon. It was able to get access to more systems because it had the developers' home directories mounted (this isn't be the case with the new system, which is completely isolated). They're also rolling out the new audit logging daemon, which should mean that this sort of thing would be detected much sooner in the future.
We were lucky. The attackers seemed not to know what they'd found. There was an (apparently) automated scan for ssh keys (which have now all been revoked) and that's about it. They seemed to be trying to add the machine to a botnet, rather than to attack it directly. We believe that they got access via a compromised developer VM which had an ssh key that connected to the cluster (for doing svn+ssh commits), so it's possible that it was an entirely automated attack. They attempted to run a load of Linux admin commands and apparently gave up when they didn't work. As far as we can tell, they didn't actually modify anything (although, of course, the compromised machines are offline pending imaging for forensic analysis and clean reinstalls or replacement, just in case). The announcement tells you not to trust any of the things that we know that they could have touched, but it currently looks like there's a very low probability that they did touch anything. For example, they might have modified ports / base cvs, but the top of the tree is identical to the svn tree (which had off-site backups, and has had every recent commit manually audited) and so they'd have had to insert something bad and then remove it in a subsequent CVS version, and that seems unlikely. Verifying cvs is pretty hard, which is part of the reason why we're encouraging everyone to move to svn now.
I had an account on a machine in the same rack as kernel.org, and that machine was taken away for forensic analysis and still isn't back. Apparently (I don't do security research, but I work on a team that does) kernel.org contained the world's best collection of rootkits found to date, which was incredibly useful to people doing work in this area.
The idea of a rental is problematic anyway for digital content. You loan me a physical object, you no longer have it. I use it and then give it back, and you have it again. With an electronic version, you give me a copy of an object, but still have the original. Later, I delete the copy and you still have the original. The problem is trying to use a metaphor that simply doesn't make sense as the basis for a business model.
The service of value that Netflix offers is not rental, it's access to an ever-increasing library. From their perspective, it should be great if someone downloads something and watches it locally the second time, because they've reduced Netfilx's bandwidth costs. You can't reasonably download their entire catalog (even limiting people to downloading 30 or 60 hours a month would satisfy most people and prevent this easily), the thing you're paying for is a service that means that when you want to watch something you can easily get something that you want to watch. The more technical obstacles that they put in the way of this, the less valuable their service becomes.
I started ripping my DVDs a while ago. I bought 3 2TB disks just before the floods and put them in a RAID-Z array, giving me 4GB of usable space. That's enough for full backups of my laptop (with periodic snapshots) and for my DVD collection (probably about 1TB when I've ripped them all). The NAS box runs FreeBSD and is connected to my projector and speakers. Because none of the streaming services use open standards for their services, I can't use them from this machine, so they don't get my money even though I'm right in the middle of the early-adopter demographic for video stuff (I bought a DVD player when they were still expensive, I have a projector and surround sound system for watching them, I subscribed to the first DVD-in-the-mail service to appear in this country as soon as it appeared).
The more ethical pizzeria already has enough employees to meet its current demand, but if people switch from Papa Johns to them, then they will need more employees (possibly even a new branch) to meet the increased demand.
You might want to go back and watch the Monty Python sketch before you're so quick to say 'whoosh' to people who got the joke that you didn't...
If you actually did pay for yours, then no one would complain. The problem is when you dump all of your externalities on everyone else and expect them to pay for yours too.
Secondly, when one looks at the amounts of malware available for each platform it does become clear that the 'walled garden' does seem to have an affect on device security.
Until you factor in the fact that iOS uses the MAC framework from FreeBSD to enforce a pretty restrictive access, and has a solid centralised update system, while Android uses a fragile, hacky chroot()-based system and only gets updates for core system libraries on a few devices / carriers. Then it becomes a lot less clear.
And Cuban (stating the obvious) summed it up nicely : "I Wouldn't Buy Facebook Stock"
Well, he wouldn't now: he bought a load at the IPO and lost money on it...
These are people who have already opted in to receive messages from you. A fair comparison is people who have subscribed to your mailing list, RSS feed, or whatever. If it's costing you three cents per subscriber to your mailing list, then you're probably doing something wrong.
It's not that uncommon. You have lots of blood vessels close to the surface there, so drugs get into your blood stream quickly. There was a story recently about people dying from vodka suppositories - so much alcohol goes into their blood so quickly that it's fatal.
Uh, I meant Itanium. Freud got me again - I'm still bitter about it killing a superior architecture through employing better sales drones.
Actually, in all seriousness, my gay friends tell me that the biggest down side is more or less the opposite of what I said: there's a significant chance that anyone that they're attracted to won't even be interested in people the same gender as them, let alone them specifically. If you think that gay people make up around 5% of the population, there's a one in twenty chance that someone they find attractive will also be gay. And if they are, the probability of them being gay, single, and attracted to you is even lower. At least for heterosexuals start with a nineteen in twenty chance before adding in the other factors...