Google — you could have one central authority organize everything into a single hierarchical structure that's organized very well... or, you can do as good of a guess as possible with lots of help from machines.
GMail — you could spend lots of man-hours neatly organizing your mail into folders... or, you could just search it.
Biology — using humans to figure out causation is best, but using machines to find correlation will work for now.
When you find yourself spending your whole day organizing data into trees, but the amount of data is constantly growing, you begin to realize that it's not useful to perfectly organize everything anymore.
If license plate numbers didn't exist, you'd have WAAAY more people running from the cops.
The absolute first thing a police officer does when they decide to pull you over is to take your license plate. THEN they turn their lightbar on. If you try to run after they take your license plate number, you're an idiot, because an "evading a police officer" charge will show up on your record, possibly including an arrest warrant, etc.
However, if you run before they get your number (usually by turning into a residential area... sadly, I know a guy who likes to do this), then all they have to go on is a vehicle of make X and color Y. If they were going to pull you over for a more minor infraction, it often isn't worth it for them to spend more than 10-15 minutes looking for you.
Yeah, all tools that do tcpdump/Wireshark-style packet inspection require root (you don't want normal user programs sniffing everything). It's true that it's alpha quality code that does TCP communications, so it's a good idea to not leave it running all the time, and/or wait until a beta version has been released.
A bigger issue is that some of your sniffed packets are sent in the clear to EFF, so 1) it's possible that a third party could sniff those few packets (but it's only a handful of packets, but it could still cause problems, and 2) if you use EFF's server, you have to trust EFF with the handful of sniffed packets you send them (but you can run your own server). It's too complicated to summarize in a few sentences, see the README.txt in the package.
They do say they'll fix the issue that third parties could sniff your packets though (by doing the obvious thing and encrypting them between endpoints), so again, waiting for a later version might be a good idea.
Actually, the main HTTP RFC (RFC2616) covers caching proxy servers. By putting standard-copyright content on a webserver, you're almost guaranteeing that a third-party caching proxy is going to redistribute your content, but the law (so far) has glossed over this detail. Even if courts do eventually address that issue, they're unlikely to conclude that this technicality somehow means that copyright owners intend their works to be redistributed in any additional ways.
The "whole world" part doesn't matter. When news organizations put photographs up on their website, for free, anybody in the world is allowed to view them, but they're not allowed to redistribute them.
Technically, once the tracker is taken down, no additional people can start downloading unless it's a DHT. It's true that existing users can finish their download. *shrug* I think this is an area of law that courts haven't considered yet, but I'd bet they'd take it to be analogous to non-P2P distribution (there have been companies that use P2P merely as the distribution mechanism that's an alternative to webservers, and they intend to be able to stop distributing their copy of the file at some point)
Bzzzt, nope. At least in the U.S., the law says that the copyright owner is the only one who has permission to make copies of their work. So yes, you are allowed to download copies (since the copyright owner is the one who's facilitating that), but no, you're not allowed to redistribute it.
You don't use the term "license" unless there's an actual legal blurb that modifies standard copyright. If there's no such blurb, it's legally covered under standard copyright. Whatever the judge thinks, that doesn't change the law.
(yes, bit torrent does, at a technical level, involve redistribution of the file by peers... from a legal standpoint, either 1) copyright law would gloss over this as it does in-RAM copies, or 2) it would say that end users are still prohibited from redistributing it, unless the copyright owner explicitly distributes their work under something other than standard copyright)
And the best way to protect a computer is to remove its network and power cables. In the real world though, security isn't the exclusive goal.
I agree that open source apps give a stronger guarantee of security, but going from "we want things to be more secure" to "we want absolute security" to "closed-source apps must necessarily be removed" seems like a stretch, even if it makes for good open source advocacy.
Wikipedia needs your penis movies. You would think that with anonymity and with the rare opportunity to flaunt their packages in front of the world, people would be tripping all over themselves to upload their junk, but sadly it just isn't so. </sarcasm>
No, burning the RPM DB to CD isn't sufficient. Any manual that says this is wrong.
The local rpm binary (or any important file it uses) can be trojaned. So, before you run `rpm`, you first have to compare checksums of those files to checksums you've written to CD. The 'md5sum' (or its replacement) program also has to be stored on CD to make sure it hasn't been trojaned.
However, even that isn't sufficient. Even if you make sure you're running a pristine copy of md5sum off the CD, the kernel itself could have been subverted to patch the md5sum binary during loading.
Rule #2 of intrusion detection: Never ever run the intrusion detection program from the same system you're trying to protect. (unless the kernel and everything else are running from read-only media, i.e. a live CD)
rpm -V checks against the LOCAL database of rpm md5 checksums. It uses the LOCAL md5 binary. And the LOCAL rpm binary. All of which are very easy to modify.
The first rule of intrusion detection is: the intrusion detector should NEVER use any files that are writable from the system it's protecting.
Q: Can I use slydial to connect to the voicemail of a landline phone number?
A: No. slydial only works to connect to the voicemail of a U.S. mobile phone user.
Which is weird, since landlines that have provider electronic voicemail, those have remote numbers too, no?
Anyway, there are some publicly-posted lists of these numbers. [1][2]
Well, SlyDial has figured out a way to do it across providers (some sort of fancy PBX setup?), so it should be possible for mainstream providers to do the same thing. But I agree that they probably don't.
Yup. There's exactly one place this rabbit hole ends up — exposing the fact that companies with more than a few hundred employees are almost guaranteed to have one of them doing unlawful things.
So, it'll be interesting if Google can get the judge to understand the difference between two or three employees misbehaving, and having an entire Google team tasked with uploading illegal videos.
What's the chance that Google did the latter? It would be unbelievably stupid for Google to do, not merely because they'd risk having a whistleblower, but because so many other people were willing to upload copyrighted videos for them. So, this is really just lawyers trying to gain whatever leverage they can to force an early settlement and increase the settlement payout, nothing more.
I'm wondering, first of all, how much it's really changed
At the time, Darwin didn't know about any of the actual mechanisms that enabled the transmission of genes, he just inferred that they must exist via statistics. Since then, we've discovered DNA, and it confirmed most of his findings. We've been able to use population genetics to figure out what route humans took to initially expand to all the continents, and everything else that the actual mitochondrial/nucleic DNA mechanisms taught us.
Whereas reading the top 20 comments on Digg or Reddit tell you that conspiracy theories are actually believable, that it's the people who pay their taxes who are the silly ones, and that cats are really cute. (okay, maybe the last one is true)
55 comments so far, and nobody has mentioned welcoming our robot overlords? It's true that the meme is getting a bit old (even Fox news has picked it up — quite the death-knell), but that's never stopped Slashdotters before.
Also, the ISPs have yet to realized that this is a two-way street.... the natural response is going to be content providers "unionizing" to increase their negotiating clout
Not really. The internet is great because YouTube spinoffs are a dime a dozen.The internet provides more competition than we've ever had before. But that's almost the mirror opposite of ISPs who own the last mile. Most customers have either one or two choices of ISP. When faced with such a disparity in competitive environments, the ISPs have tons of leverage, and content providers have almost none (even when they group up, they're still < 50% of the suppliers).
AOL+Yahoo doesn't strike me as being able to produce better services than Yahoo alone could. Or MS+Yahoo. Or any other combination.
The bigger a company is, the more cultural inertia it has, the less willing it is to try something new. Would strapping AOL's "never change anything" mentality to any company make it better? At least Microsoft has occasionally given one of its subdivisions such free-reign that it's been able to innovate (Microsoft mice, xbox360's networking features). Still, MS is mostly extra baggage.
Yahoo by itself is already producing tons of different services, on the off-chance that a handful will be successful. Combining with someone larger will certainly slow that down. Would that slowdown be offset by making some more likely to be successful? I doubt it.
My premise is that it's impossible (from a city-wide perspective) to have a short commute without an increase in density.
If you're saying that it's possible for an individual to optimize for commute distance at the expense of higher rent, less floor space, and/or a less desirable neighborhood, then I agree. However, if you advocate that everyone should try to optimize for the same variable, then all the other variables would get dramatically worse for everyone (much higher rent, much less floor space). (competition for scarce resources works best when people (as a whole) don't reject some of the available solutions — ie. sprawl as a substitute for density)
(Yes, in cities that primarily use trains in place of expressways, and buses and/or walking in place of local streets, it takes ~15 minutes to do the walking+bus part, but (for example, in Tokyo) average commutes are ~1 hour, so that's acceptable for those who aren't strongly optimizing for commute time)
That's not a scalable solution (unless you consider living in an ever-shrinking apartment because all of the suburbanites took your advice and moved into the city with you to be scalable). Trains avoid that problem, and they're better than expressways because they transport more people per route.
Go to Tokyo at rush hour, and observe. The only slowdown that occurs is when one train becomes too full, and people have to wait ~3 minutes for the next one. I never saw a situation where people had to wait for more than one additional train, because the trains can hold a lot of people because they're packed like cattle-cars. On the other hand, Japanese seem to be much better at being fairly quiet and avoiding talking on their cellphones when in such dense quarters, while Americans seem to think that the subway is the best place for talking really loudly on the phone.
Slashdot Mirror
When you find yourself spending your whole day organizing data into trees, but the amount of data is constantly growing, you begin to realize that it's not useful to perfectly organize everything anymore.
If license plate numbers didn't exist, you'd have WAAAY more people running from the cops.
The absolute first thing a police officer does when they decide to pull you over is to take your license plate. THEN they turn their lightbar on. If you try to run after they take your license plate number, you're an idiot, because an "evading a police officer" charge will show up on your record, possibly including an arrest warrant, etc.
However, if you run before they get your number (usually by turning into a residential area... sadly, I know a guy who likes to do this), then all they have to go on is a vehicle of make X and color Y. If they were going to pull you over for a more minor infraction, it often isn't worth it for them to spend more than 10-15 minutes looking for you.
Yeah, all tools that do tcpdump/Wireshark-style packet inspection require root (you don't want normal user programs sniffing everything). It's true that it's alpha quality code that does TCP communications, so it's a good idea to not leave it running all the time, and/or wait until a beta version has been released.
A bigger issue is that some of your sniffed packets are sent in the clear to EFF, so 1) it's possible that a third party could sniff those few packets (but it's only a handful of packets, but it could still cause problems, and 2) if you use EFF's server, you have to trust EFF with the handful of sniffed packets you send them (but you can run your own server). It's too complicated to summarize in a few sentences, see the README.txt in the package.
They do say they'll fix the issue that third parties could sniff your packets though (by doing the obvious thing and encrypting them between endpoints), so again, waiting for a later version might be a good idea.
Actually, the main HTTP RFC (RFC2616) covers caching proxy servers. By putting standard-copyright content on a webserver, you're almost guaranteeing that a third-party caching proxy is going to redistribute your content, but the law (so far) has glossed over this detail. Even if courts do eventually address that issue, they're unlikely to conclude that this technicality somehow means that copyright owners intend their works to be redistributed in any additional ways.
The "whole world" part doesn't matter. When news organizations put photographs up on their website, for free, anybody in the world is allowed to view them, but they're not allowed to redistribute them.
Technically, once the tracker is taken down, no additional people can start downloading unless it's a DHT. It's true that existing users can finish their download. *shrug* I think this is an area of law that courts haven't considered yet, but I'd bet they'd take it to be analogous to non-P2P distribution (there have been companies that use P2P merely as the distribution mechanism that's an alternative to webservers, and they intend to be able to stop distributing their copy of the file at some point)
Bzzzt, nope. At least in the U.S., the law says that the copyright owner is the only one who has permission to make copies of their work. So yes, you are allowed to download copies (since the copyright owner is the one who's facilitating that), but no, you're not allowed to redistribute it.
You don't use the term "license" unless there's an actual legal blurb that modifies standard copyright. If there's no such blurb, it's legally covered under standard copyright. Whatever the judge thinks, that doesn't change the law.
(yes, bit torrent does, at a technical level, involve redistribution of the file by peers... from a legal standpoint, either 1) copyright law would gloss over this as it does in-RAM copies, or 2) it would say that end users are still prohibited from redistributing it, unless the copyright owner explicitly distributes their work under something other than standard copyright)
And the best way to protect a computer is to remove its network and power cables. In the real world though, security isn't the exclusive goal.
I agree that open source apps give a stronger guarantee of security, but going from "we want things to be more secure" to "we want absolute security" to "closed-source apps must necessarily be removed" seems like a stretch, even if it makes for good open source advocacy.
Damn, I've got moderator points, but where is the "this is the only post you need to read" option when you need it?
Wikipedia needs your penis movies. You would think that with anonymity and with the rare opportunity to flaunt their packages in front of the world, people would be tripping all over themselves to upload their junk, but sadly it just isn't so. </sarcasm>
No, burning the RPM DB to CD isn't sufficient. Any manual that says this is wrong.
The local rpm binary (or any important file it uses) can be trojaned. So, before you run `rpm`, you first have to compare checksums of those files to checksums you've written to CD. The 'md5sum' (or its replacement) program also has to be stored on CD to make sure it hasn't been trojaned.
However, even that isn't sufficient. Even if you make sure you're running a pristine copy of md5sum off the CD, the kernel itself could have been subverted to patch the md5sum binary during loading.
Rule #2 of intrusion detection: Never ever run the intrusion detection program from the same system you're trying to protect. (unless the kernel and everything else are running from read-only media, i.e. a live CD)
rpm -V checks against the LOCAL database of rpm md5 checksums. It uses the LOCAL md5 binary. And the LOCAL rpm binary. All of which are very easy to modify.
The first rule of intrusion detection is: the intrusion detector should NEVER use any files that are writable from the system it's protecting.
it will determine that certtool is provided by gnutls-utils and install that package. IIRC, apt-get can't do that.
apt-file search path/to/myfile
Q: Can I use slydial to connect to the voicemail of a landline phone number?
A: No. slydial only works to connect to the voicemail of a U.S. mobile phone user.
Which is weird, since landlines that have provider electronic voicemail, those have remote numbers too, no?
Anyway, there are some publicly-posted lists of these numbers. [1] [2]
Well, SlyDial has figured out a way to do it across providers (some sort of fancy PBX setup?), so it should be possible for mainstream providers to do the same thing. But I agree that they probably don't.
Yup. There's exactly one place this rabbit hole ends up — exposing the fact that companies with more than a few hundred employees are almost guaranteed to have one of them doing unlawful things.
So, it'll be interesting if Google can get the judge to understand the difference between two or three employees misbehaving, and having an entire Google team tasked with uploading illegal videos.
What's the chance that Google did the latter? It would be unbelievably stupid for Google to do, not merely because they'd risk having a whistleblower, but because so many other people were willing to upload copyrighted videos for them. So, this is really just lawyers trying to gain whatever leverage they can to force an early settlement and increase the settlement payout, nothing more.
At the time, Darwin didn't know about any of the actual mechanisms that enabled the transmission of genes, he just inferred that they must exist via statistics. Since then, we've discovered DNA, and it confirmed most of his findings. We've been able to use population genetics to figure out what route humans took to initially expand to all the continents, and everything else that the actual mitochondrial/nucleic DNA mechanisms taught us.
Otherwise the helium would make their voices squeaky, and having squeaky voices over the radio isn't very manly.
Whereas reading the top 20 comments on Digg or Reddit tell you that conspiracy theories are actually believable, that it's the people who pay their taxes who are the silly ones, and that cats are really cute. (okay, maybe the last one is true)
55 comments so far, and nobody has mentioned welcoming our robot overlords? It's true that the meme is getting a bit old (even Fox news has picked it up — quite the death-knell), but that's never stopped Slashdotters before.
Not really. The internet is great because YouTube spinoffs are a dime a dozen.The internet provides more competition than we've ever had before. But that's almost the mirror opposite of ISPs who own the last mile. Most customers have either one or two choices of ISP. When faced with such a disparity in competitive environments, the ISPs have tons of leverage, and content providers have almost none (even when they group up, they're still < 50% of the suppliers).
What uses 0.2hz waves for acoustic communication? Even an elephant on sulfur hexafluoride doesn't get that low.
AOL+Yahoo doesn't strike me as being able to produce better services than Yahoo alone could. Or MS+Yahoo. Or any other combination.
The bigger a company is, the more cultural inertia it has, the less willing it is to try something new. Would strapping AOL's "never change anything" mentality to any company make it better? At least Microsoft has occasionally given one of its subdivisions such free-reign that it's been able to innovate (Microsoft mice, xbox360's networking features). Still, MS is mostly extra baggage.
Yahoo by itself is already producing tons of different services, on the off-chance that a handful will be successful. Combining with someone larger will certainly slow that down. Would that slowdown be offset by making some more likely to be successful? I doubt it.
My premise is that it's impossible (from a city-wide perspective) to have a short commute without an increase in density.
If you're saying that it's possible for an individual to optimize for commute distance at the expense of higher rent, less floor space, and/or a less desirable neighborhood, then I agree. However, if you advocate that everyone should try to optimize for the same variable, then all the other variables would get dramatically worse for everyone (much higher rent, much less floor space). (competition for scarce resources works best when people (as a whole) don't reject some of the available solutions — ie. sprawl as a substitute for density)
(Yes, in cities that primarily use trains in place of expressways, and buses and/or walking in place of local streets, it takes ~15 minutes to do the walking+bus part, but (for example, in Tokyo) average commutes are ~1 hour, so that's acceptable for those who aren't strongly optimizing for commute time)
That's not a scalable solution (unless you consider living in an ever-shrinking apartment because all of the suburbanites took your advice and moved into the city with you to be scalable). Trains avoid that problem, and they're better than expressways because they transport more people per route.
Ding ding.
Go to Tokyo at rush hour, and observe. The only slowdown that occurs is when one train becomes too full, and people have to wait ~3 minutes for the next one. I never saw a situation where people had to wait for more than one additional train, because the trains can hold a lot of people because they're packed like cattle-cars. On the other hand, Japanese seem to be much better at being fairly quiet and avoiding talking on their cellphones when in such dense quarters, while Americans seem to think that the subway is the best place for talking really loudly on the phone.