Firstly, this is a relevant question and I don't see why it was marked "offtopic".
I have been a debian stable user for many years, and recently installed gentoo on my main workstation, so here's the comparison as I see it:
In debian stable, as in gentoo x86, the packages are tested before being released. The difference is that in debian stable, the distribution as a whole is tested, so you don't get integration problems. When I installed gentoo, I got a good working version of gimp, and a good working version of sane (scanner tools), but they weren't compatible with each other -- to scan an image from gimp I'd need to upgrade to a newer "unstable" sane or downgrade to an older gimp.
Secondly, debian generally tries to configure packages for you. You decide you want a service, you install the package, and that service is installed and running. It might need to ask you a couple of questions, but that's all. Gentoo explicitly leaves the runtime configuration of a package to the user. The "zero-administration" paradigm of debian is extremely ambitious, but they are surprisingly successful at it -- probably more so than any other distribution or for that matter any other operating system. Mostly everything Just Works.
Then there's the obvious differences: debian stable is very old, and in Gentoo you have to wait for stuff to compile.
The reason Gentoo can have the "rolling stable" distribution is because it is a source distribution. debian testing is the same kind of idea, but it doesn't work nearly as well because binary packages necessarily have more dependencies on specific versions of other packages. It is the loosening of dependencies, not the tiny optimisations, that is the real benefit of distributing as source.
Of course, if you want the "stable distribution" rather than "distribution of stable software", then debian has the advantage. The gimp/sane problem was the only issue of that kind that I've encountered so far, though, so from my point of view the integration question seems fairly manageable.
Copyright laws give the power back to the people, as it were.
No they don't.
The flaw with this scheme is that while it tries to stop you from being spammed, you have no recourse if you are spammed. The only party that can act is this essentially uninvolved third party which holds the copyright.
In other words, it has exactly the same problem you've (correctly) identified in CAN-SPAM.
Secondly, when it succeeds it's a bad precedent. It eats away at the principle of "reverse engineering for compatibility", that was upheld in the garage-door-opener case. Exactly the same technique could be used to restrict access to other kinds of services. The fact that this instance is in a "good cause" doesn't change the principles.
It comes back to my first point: the only person with authority to say who accesses my servers is me.
That is correct. Contrary to the parent, citizens of the Republic of Ireland who are not UK citizens do not qualify.
There is a concept of "Honorary Knighthood", which can be given to non-subjects of the crown. Some recipients who come to mind are Caspar Weinberger, Stephen Spielberg, Rudolph Giuliani a couple of years ago, Bob Geldof and Spike Milligan (last 2 both Irish). As the knighthoods are honorary, they are not entitled to be called Sir Caspar etc.
I seem to remember "Planet of the Damned" by Harry Harrison started with a kind of "decatholon" involving weightlifting, running, poetry composition and chess, among other events. It took place on a planet that was ice-bound for most of the year and had a brief summer when everyone went wild, or something like that.
I'm in two minds about this. It's true that a drive-through is a different situation than a restuarant -- handling a very hot drink in a flexible cup in a car is intrinsically slightly dangerous.
But "how hot 190 degrees is"... please!
"Not hot enough" is how I would describe it if I was buying it in a restaurant to drink at a table. Water boils at 212 degrees, and that's what I make my hot drinks with at home. I'm more of a tea drinker than a coffee drinker, and if you don't make your tea with boiling water, it doesn't taste right.
In fact, I've spilled boiling water over myself quite often (and I mean just a few seconds out of the kettle -- definitely more than 190F). Worst case is it makes a patch which is sore for a day or so. Of course, I'm not 80, and that makes a big difference. If boiling water hurt me more, I'd be more careful with it than I generally am.
There's a pile of contributory factors here: Victim elderly, therefore more likely to be injured. Coffee hotter than normal. Non-rigid cup. Trying to open the lid, because the vendor doesn't adjust the drink for you. Being in a car, where you don't have room to do things safely.
I suppose the real question is: whose responsibility is it to combine that list of factors and realise that there's a dangerous situation? My instinct is to say "the customer", but that's just me, and really it's a matter of custom. By custom, you could suggest that a large chain has a greater responsibility to think ahead about these things than Bob's drive-thru would have. On the other hand, you might argue that bottom-end-of-the-market vendor should have less responsibility than a family-owned joint where there is an expectation of "looking after the customer." So, definitely not frivolous, even if I don't agree with the verdict.
That's occurred to me too. I've noticed it mostly in my own posts, that if it gets to 3, it's likely to get to 4, and if it gets to 4 it's almost certain to get to 5, even if it obviously doesn't deserve it.
Fully logarithmic might be excessive; 1 vote to +1, 2 more to +2, 3 more to +3 would be worth a try, sort of like the foodbox in civ2.
The other approach would be to not allow moderation directly when viewing comments in a manner that emphasises already-high-scoring comments. Just don't give the links. The moderator could still select a post and then moderate it, but it's not encouraging moderators to apply points to posts that have already received moderation, as the current system does.
They've got to catch the French ship--so? Why should I, as a moviegoer, care about that for its own sake?
Um, because it's French. You know, like, from France. French. Get it?
Re:Another important exception: acid rain
on
Global Dimming
·
· Score: 1
Possibly... It is plausible that a strong prevailing air current could act almost like a river and carry air pollutants a long distance without dispersing them much. It is by no means established.
I was not aware of the North American acid rain situation, but in the Scandinavian case I had the impression that attention had turned more to local causes, particularly forest fires, which put acid into the atmosphere but deposit alkaline ash on the soil.
Re:Old-style environmentalism
on
Global Dimming
·
· Score: 1
Absolutely not. The thing about the air, and for that matter the sea, is that as pollutants move, they disperse. You just don't realise how big the earth is. One volcano can throw more assorted crud into the air (including SOx, NOx, the works) than the whole of the human race. It might be nice, in a way, to think we're bigger than that, but we're not.
Unlike the sea and the atmosphere, stuff in rivers doesn't have three dimensions to disperse into -- it only has one. If you dump mercury, or pesticides, or just excessive nitrates into the river, it won't disperse until it's out into the open sea, and can cause a lot of problems all the way to there.
Yes, women are generally involved too. But a relatively small pool of "available" women can infect a majority of young men.
Old-style environmentalism
on
Global Dimming
·
· Score: 4, Insightful
Indeed. That is what environmentalism used to be about. Real, obvious problems that you could point to and do something about.
Unfortunately, real environmental problems are usually created locally*. Fixing them means taking the economic hit locally -- losing factory jobs in your own city, reducing the fertilizer-driven crop yield on your own farm, having a smaller engine in your own car, whatever.
It's much better to deal with global environmental issues, which are, by definition, somebody else's fault. "It's not me, it's those darned Amazonian loggers! I can't do anything by myself, the world's governments need to get together and make everyone do things differently."
[* important exception: rivers. Rivers carry and in some cases even concentrate pollution from large distances upstream]
Yeah, I think it was during 2002 that multi-region players became more or less universal in the UK. Certainly my year-old one from Tesco is in the "Tap the secret code into the remote" category.
Current low end is GBP30-40, so if you think you'll get landed with 2 more R1 DVD's, it's worth it. 1 more, and it's marginal. If you have any use at all for having 2 players, that probably tilts it to be worth getting one. (I can't believe prices can fall much further).
Releases in Britain are "localised" to the extend of being approved by the BBFC, which can involve tailoring of the content.
Also, it would be impossible to have a region boundary going through the EU: under EU law there is free movement of goods between member states, so anyone in (say) Sweden can import British DVD players, personally or commercially. It would be illegal for the manufacturers to attempt to prevent this with contracts (see the endless battles with car manufacturers).
US consumers are least affected by region codes: they watch virtually only US content, and have small risk of wanting to play a non-region-1 DVD. (obviously there are exceptions, but I'm talking about the mass of consumers here).
Outside the US, where most consumers watch a mixture of domestic and US produced content, multi-region players are the norm. I think I read that all players in New Zealand are multi-region, and I know for a fact it would be hard to get one here in the UK that isn't.
So it's mainly a problem for Blockbuster: they can't rent out an out-of-region DVD even if 90% of consumers can watch it, because the other 10% will cause them so much trouble.
And you've got the right answer. It's 2. Some things are too complicated to push into a couple of dozen slides. Some things do have to be actually studied before you can make good decisions. You might want to give a half-hour talk to a group to tell them what they have to study and what they have to decide, and Powerpoint or equivalent is a great help to that, but you can't say that keeping the space shuttle in the sky isn't rocket science. You give them maybe twenty pages of analysis, including charts or whatever, that is going to take a few hours to study and understand, and if you need to talk to them you make a brief presentation that talks about the report.
To many readers of/., SCO's claim that this strengthens their argument seems more plausible than their earlier gibberish over Linux.
Fear not.
On the history of UNIX and on the details of intellectual property law, the average Nasdaq investor is less knowledgable than the average slashdotter.
But on the mechanics of business disputes and the running of companies, they are more knowledgable. They are not idiots, and they will interpret this as "HP's lawyers have spent a lot of time looking at this, and HP are betting BIG MONEY that SCO are full of shit."
SCO's desperate pleading in the Red Hat case -- "don't make us defend our FUD in court, we never dreamed of suing Red Hat for merely distributing Linux, they have no reasonable apprehension of being sued." -- will have given HP huge confidence for this move. SCO as good as said "We will make no demands of anyone who has the resources, expertise, and incentive to prove in court that Linux does not infringe our copyrights. We merely intend to shake down those for whom it is cheaper to send us a few thousand dollars than take the effort and risk of opposing us".
If SCO had the guts to take on HP, they would have had the guts to take on Red Hat. In actively running away from Red Hat they have invited this.
So they're giving weight to it by saying "SCO's claims are false. If we're wrong, we'll pay our customers' bills" ?
SCO are being offered credence, by the markets at least. The current market price can only be justified by an assumption that SCO are going to make hundreds of millions of dollars either from IBM or from Linux users.
This attacks that assumption. Nothing anyone can say carries as much weight as HP putting its money where its mouth is.
Fundamentally, you are 100% correct. NAT provides no security that simple filtering can't do better.
However, practical security is about more than fundamentals. It is about what happens when you screw up.
I have a gateway with filtering. If I mess up a config, my gateway might come up without the iptables rules, and I might not notice for ages.
However, all the workstations I use are NATted, (All things being equal I would prefer that they weren't, but the charges from my ISP would be far from equal), and if the NAT is accidentally deactivated, I'll notice pretty damn quick because nothing will work.
Yes, you could say "If you screw up your config that's your own fault, you're incompetent", but a large proportion of real-world security breaches are caused by administrator errors, and a security system that is more resistant to errors is "more secure" by reasonable definitions.
Having said all that, your point about the ISP being able to route directly to your internal IPs is a good one. Luckily (because I hadn't really thought about it), my iptables setup will reject any such packets.
you are allowed to make copies of your friend's CD/tape/whatever--as long as you are not selling it
Your logic is fine, your premiss is wrong. In most jurisdictions (including the USA), casual copying for friends is still an infringement.
The only differences are (a) it is impossible for them to catch you, and (b) if they do, they can only say you made one copy, and therefore claim miniscule damages. If you run kazaa, or upload to a ftp site, they can claim thousands of people downloaded, or might have downloaded, the music, and multiply the damages they claim from you accordingly.
It depended a lot on the platform. On the orignal BBC B version, those vipers were terrifying. You really had to be well gunned up and pretty experienced to be prepared to get a "fugitive" rating. On the Spectrum version, however, and the DOS "Elite II" for that matter, the vipers were less threatening and came in smaller numbers, so it really was simpler to shoot everything.
I don't know how much involvement David Braben had in the ports.
The technical workaround is good, but I think this is one rare case where legal action might be reasonable.
If you don't want deep linking, you're objecting to how various random individuals on the internet interact with your computers. You should restrict that interaction on your own computer and not whine about the rest of the world.
Verisign are not some random external party - they exclusively control chunks of the internet infrastructure. They should be held to a higher standard of behaviour.
Of course, the real technical solution is for everyone to use an alternative root server. Given the economic network effects in the internet, that's very difficult to arrange. (If Verisign's abuse got much worse, it would be just conceivable).
SCO's press releases have no relationship with their statements to the court. SCO have made no claims to any court that support the claims made in press releases regarding Linux users.
"can't afford to upgrade" is not the main reason why large institutions often run on old software infrastructure.
The discussion here is how much can be broken in such an environment by a single security fix. How much do you think could be broken by upgrading from NT4 to W2K or XP? If the NT4 servers work, why would admins even consider going through this?
And that doesn't change the fact that the RIAA isn't suing him for violating MIT's AUP.
If he's got an alibi, then he didn't infringe the plaintiff's copyright. If they say "but you must have broken the AUP!", he can say "I guess I must have, but I didn't download those files."
If MIT then want to have a go at him for breaking the AUP, that's fine, but I imagine they're not in the habit of suing their students.
Slashdot Mirror
I have been a debian stable user for many years, and recently installed gentoo on my main workstation, so here's the comparison as I see it:
In debian stable, as in gentoo x86, the packages are tested before being released. The difference is that in debian stable, the distribution as a whole is tested, so you don't get integration problems. When I installed gentoo, I got a good working version of gimp, and a good working version of sane (scanner tools), but they weren't compatible with each other -- to scan an image from gimp I'd need to upgrade to a newer "unstable" sane or downgrade to an older gimp.
Secondly, debian generally tries to configure packages for you. You decide you want a service, you install the package, and that service is installed and running. It might need to ask you a couple of questions, but that's all. Gentoo explicitly leaves the runtime configuration of a package to the user. The "zero-administration" paradigm of debian is extremely ambitious, but they are surprisingly successful at it -- probably more so than any other distribution or for that matter any other operating system. Mostly everything Just Works. Then there's the obvious differences: debian stable is very old, and in Gentoo you have to wait for stuff to compile.
The reason Gentoo can have the "rolling stable" distribution is because it is a source distribution. debian testing is the same kind of idea, but it doesn't work nearly as well because binary packages necessarily have more dependencies on specific versions of other packages. It is the loosening of dependencies, not the tiny optimisations, that is the real benefit of distributing as source.
Of course, if you want the "stable distribution" rather than "distribution of stable software", then debian has the advantage. The gimp/sane problem was the only issue of that kind that I've encountered so far, though, so from my point of view the integration question seems fairly manageable.
The flaw with this scheme is that while it tries to stop you from being spammed, you have no recourse if you are spammed. The only party that can act is this essentially uninvolved third party which holds the copyright.
In other words, it has exactly the same problem you've (correctly) identified in CAN-SPAM.
Secondly, when it succeeds it's a bad precedent. It eats away at the principle of "reverse engineering for compatibility", that was upheld in the garage-door-opener case. Exactly the same technique could be used to restrict access to other kinds of services. The fact that this instance is in a "good cause" doesn't change the principles.
It comes back to my first point: the only person with authority to say who accesses my servers is me.
There is a concept of "Honorary Knighthood", which can be given to non-subjects of the crown. Some recipients who come to mind are Caspar Weinberger, Stephen Spielberg, Rudolph Giuliani a couple of years ago, Bob Geldof and Spike Milligan (last 2 both Irish). As the knighthoods are honorary, they are not entitled to be called Sir Caspar etc.
I seem to remember "Planet of the Damned" by Harry Harrison started with a kind of "decatholon" involving weightlifting, running, poetry composition and chess, among other events. It took place on a planet that was ice-bound for most of the year and had a brief summer when everyone went wild, or something like that.
But "how hot 190 degrees is" ... please!
"Not hot enough" is how I would describe it if I was buying it in a restaurant to drink at a table. Water boils at 212 degrees, and that's what I make my hot drinks with at home. I'm more of a tea drinker than a coffee drinker, and if you don't make your tea with boiling water, it doesn't taste right.
In fact, I've spilled boiling water over myself quite often (and I mean just a few seconds out of the kettle -- definitely more than 190F). Worst case is it makes a patch which is sore for a day or so. Of course, I'm not 80, and that makes a big difference. If boiling water hurt me more, I'd be more careful with it than I generally am.
There's a pile of contributory factors here: Victim elderly, therefore more likely to be injured. Coffee hotter than normal. Non-rigid cup. Trying to open the lid, because the vendor doesn't adjust the drink for you. Being in a car, where you don't have room to do things safely.
I suppose the real question is: whose responsibility is it to combine that list of factors and realise that there's a dangerous situation? My instinct is to say "the customer", but that's just me, and really it's a matter of custom. By custom, you could suggest that a large chain has a greater responsibility to think ahead about these things than Bob's drive-thru would have. On the other hand, you might argue that bottom-end-of-the-market vendor should have less responsibility than a family-owned joint where there is an expectation of "looking after the customer." So, definitely not frivolous, even if I don't agree with the verdict.
Fully logarithmic might be excessive; 1 vote to +1, 2 more to +2, 3 more to +3 would be worth a try, sort of like the foodbox in civ2.
The other approach would be to not allow moderation directly when viewing comments in a manner that emphasises already-high-scoring comments. Just don't give the links. The moderator could still select a post and then moderate it, but it's not encouraging moderators to apply points to posts that have already received moderation, as the current system does.
Um, because it's French. You know, like, from France. French. Get it?
article on acid rain
I was not aware of the North American acid rain situation, but in the Scandinavian case I had the impression that attention had turned more to local causes, particularly forest fires, which put acid into the atmosphere but deposit alkaline ash on the soil.
Unlike the sea and the atmosphere, stuff in rivers doesn't have three dimensions to disperse into -- it only has one. If you dump mercury, or pesticides, or just excessive nitrates into the river, it won't disperse until it's out into the open sea, and can cause a lot of problems all the way to there.
Yes, women are generally involved too. But a relatively small pool of "available" women can infect a majority of young men.
Unfortunately, real environmental problems are usually created locally*. Fixing them means taking the economic hit locally -- losing factory jobs in your own city, reducing the fertilizer-driven crop yield on your own farm, having a smaller engine in your own car, whatever.
It's much better to deal with global environmental issues, which are, by definition, somebody else's fault. "It's not me, it's those darned Amazonian loggers! I can't do anything by myself, the world's governments need to get together and make everyone do things differently."
[* important exception: rivers. Rivers carry and in some cases even concentrate pollution from large distances upstream]
Current low end is GBP30-40, so if you think you'll get landed with 2 more R1 DVD's, it's worth it. 1 more, and it's marginal. If you have any use at all for having 2 players, that probably tilts it to be worth getting one. (I can't believe prices can fall much further).
Also, it would be impossible to have a region boundary going through the EU: under EU law there is free movement of goods between member states, so anyone in (say) Sweden can import British DVD players, personally or commercially. It would be illegal for the manufacturers to attempt to prevent this with contracts (see the endless battles with car manufacturers).
Outside the US, where most consumers watch a mixture of domestic and US produced content, multi-region players are the norm. I think I read that all players in New Zealand are multi-region, and I know for a fact it would be hard to get one here in the UK that isn't.
So it's mainly a problem for Blockbuster: they can't rent out an out-of-region DVD even if 90% of consumers can watch it, because the other 10% will cause them so much trouble.
And you've got the right answer. It's 2. Some things are too complicated to push into a couple of dozen slides. Some things do have to be actually studied before you can make good decisions. You might want to give a half-hour talk to a group to tell them what they have to study and what they have to decide, and Powerpoint or equivalent is a great help to that, but you can't say that keeping the space shuttle in the sky isn't rocket science. You give them maybe twenty pages of analysis, including charts or whatever, that is going to take a few hours to study and understand, and if you need to talk to them you make a brief presentation that talks about the report.
Fear not.
On the history of UNIX and on the details of intellectual property law, the average Nasdaq investor is less knowledgable than the average slashdotter.
But on the mechanics of business disputes and the running of companies, they are more knowledgable. They are not idiots, and they will interpret this as "HP's lawyers have spent a lot of time looking at this, and HP are betting BIG MONEY that SCO are full of shit."
If SCO had the guts to take on HP, they would have had the guts to take on Red Hat. In actively running away from Red Hat they have invited this.
SCO are being offered credence, by the markets at least. The current market price can only be justified by an assumption that SCO are going to make hundreds of millions of dollars either from IBM or from Linux users.
This attacks that assumption. Nothing anyone can say carries as much weight as HP putting its money where its mouth is.
Fundamentally, you are 100% correct. NAT provides no security that simple filtering can't do better.
However, practical security is about more than fundamentals. It is about what happens when you screw up.
I have a gateway with filtering. If I mess up a config, my gateway might come up without the iptables rules, and I might not notice for ages.
However, all the workstations I use are NATted, (All things being equal I would prefer that they weren't, but the charges from my ISP would be far from equal), and if the NAT is accidentally deactivated, I'll notice pretty damn quick because nothing will work.
Yes, you could say "If you screw up your config that's your own fault, you're incompetent", but a large proportion of real-world security breaches are caused by administrator errors, and a security system that is more resistant to errors is "more secure" by reasonable definitions.
Having said all that, your point about the ISP being able to route directly to your internal IPs is a good one. Luckily (because I hadn't really thought about it), my iptables setup will reject any such packets.
you are allowed to make copies of your friend's CD/tape/whatever--as long as you are not selling it
Your logic is fine, your premiss is wrong. In most jurisdictions (including the USA), casual copying for friends is still an infringement.
The only differences are (a) it is impossible for them to catch you, and (b) if they do, they can only say you made one copy, and therefore claim miniscule damages. If you run kazaa, or upload to a ftp site, they can claim thousands of people downloaded, or might have downloaded, the music, and multiply the damages they claim from you accordingly.
It depended a lot on the platform. On the orignal BBC B version, those vipers were terrifying. You really had to be well gunned up and pretty experienced to be prepared to get a "fugitive" rating. On the Spectrum version, however, and the DOS "Elite II" for that matter, the vipers were less threatening and came in smaller numbers, so it really was simpler to shoot everything.
I don't know how much involvement David Braben had in the ports.
The technical workaround is good, but I think this is one rare case where legal action might be reasonable.
If you don't want deep linking, you're objecting to how various random individuals on the internet interact with your computers. You should restrict that interaction on your own computer and not whine about the rest of the world.
Verisign are not some random external party - they exclusively control chunks of the internet infrastructure. They should be held to a higher standard of behaviour.
Of course, the real technical solution is for everyone to use an alternative root server. Given the economic network effects in the internet, that's very difficult to arrange. (If Verisign's abuse got much worse, it would be just conceivable).
SCO's press releases have no relationship with their statements to the court. SCO have made no claims to any court that support the claims made in press releases regarding Linux users.
"can't afford to upgrade" is not the main reason why large institutions often run on old software infrastructure.
The discussion here is how much can be broken in such an environment by a single security fix. How much do you think could be broken by upgrading from NT4 to W2K or XP? If the NT4 servers work, why would admins even consider going through this?
If he's got an alibi, then he didn't infringe the plaintiff's copyright. If they say "but you must have broken the AUP!", he can say "I guess I must have, but I didn't download those files."
If MIT then want to have a go at him for breaking the AUP, that's fine, but I imagine they're not in the habit of suing their students.