While I agree that the poster was being stupid, there is one problem that still persists with nuclear waste - while it can be stored safely, every single form of transport known to man has safety problems. If the waste dump is a thousand miles away from the plant, then the transport of the waste across that thousand miles is the really risky part.
It's for the same reason dirt doesn't stay on the top of the road. Rain and wind. The radioactive dust particles that were thrown into the air washed off the road, but where they hit earth, they became part of the soil just like any other dust would.
I hate it when people put things they know are controversial in signatures. For some reason moderators think the post with the sig is on topic, but anyone responding to the sig is off-topic. That's really unfair. It means if you want to get away with propaganda - all you have to do is put it in your sig instead of in your post's body and it's considered taboo for anyone to question you on it.
Oddly enough this is one case where Microsoft is actually more correct according to the standards than their competitors. The ASCII meaning of carriage return is supposed to be *just* moving the cursor to column zero of the current line. The ASCII meaning of linefeed is supposed to be just to scroll down a line without moving the cursor horizontally. So the combination of both of them actually fits what happens at end-of-line. And of course, most TCP/IP protocols, which were certainly not designed with Windows in mind, say the same thing - lines end with a CR/LF pair.
It does make writing a program to read text files a bit annoying, though, doesn't it? I mean, how often have you *just* wanted to linefeed or *just* carriage return in a text file? How come the only place MS adheres to the standard is in a place where the standard is a bit stupid?
If the whistleblower is legitimate (and I'm leaving that decision up for grabs in my mind still), then whoever it is would have no difficulties in life after having been fired from SCO. The publicity would help in a number of areas. Yes, some companies wouldn't want to hire a known whistleblower, but others would be quite grateful, especially companies SCO was suing.
A little less, in reality - because if you have 42.6 million in investments, the moment you start selling them off like mad, your investments start losing value before you get around to selling off the rest of them. Other owners of stock start thinking, "holy crap, what do they know that we don't?", and they start selling too, lowering the prices. If they tried using that 42.6 million as if it was liquid cash, it would be worth a lot less than 42.6 million by the time they extracted all of it.
1. In a car you don't *want* too much stiffness in the frame. They are designed to crumple so when you run into something you are more likely to ruin the car instead of your body. (The car crumples to soften the impulse of the hit.) 2. The hood is irrelevant to the strength of the car anyway, since the strength doesn't come from the body but from the skeletal frame the body is attached to.
While I hate Microsoft for all the same reasons most of you do, I think in this case there's a non-malicious explanation. That redirection is probably triggered by the following conditions:
(1) The search term contains certain things that tend to find X-rated content. The algorithm might look for dictionary terms and try to form seperate words out of the serch phrase (so if you looked for hothornybabes it would notice that it contains the words "hot" and "horny" and "babes".) So, "Xfree86" probably gets flagged because it's "X" followed by "free" and some irrelevant number. But, wait, you say 'Xfree87" and "Xfree85" don't trigger, so that can't be it, right? Well, it still could be because of the next point:
2) It probably *also* only triggers the redirection if the search result returns a lot of hits.
So 'Xfree86' triggers a lot of hits, and contains red-flagged terms, while 'Xfree87' has the same flagged terms, but triggers few hits, and so isn't assumed to be porn.
Anyway, that's one possible explanation. I'd attribute this to stupidity on the part of the algorithm before attributing it to maliciousness.
Good idea, but pick somebody dead. It's less dangerous because they can't actually take office, but it definately proves to the public that the system is broken. (But be prepared to spend a *LOONG* time in jail over it if it gets discovered. The government doesn't take kindly to people proving it's stupid.)
What do you mean "would be"? It already *is* the way it works in a lot of voting districts. I've voted that way in the last 3 presidential elections. Yes, the systems used in Florida were terrible and needed fixing - but there were already good systems (like the above system) that could have been used, instead of this push toward unverifiable trust in a company to not rig an election count.
Results of French elections are usually known a few hours after the votes, and after-voting polls usually give the result right at closure time.
Given that the US covers several time zones, I really wish they'd either stagger the vote times so that all the polls close at the same exact time regardless of what the local time may be, or put a moratorium on reporting results before all the polls are closed. It really annoys me that the news shows start announcing predictions of elections while some people can still vote yet. Everyone's vote should be just as "blind" as everyone else's - nobody should get to vote using extra information that was available to them and not to others, just because they live farther to the west.
- Tallying of complicated voting system. (I.e. preferential voting.)
If that was needed, then filling in numbers would work just as well (fill in the '1' circle next to the candidate which is your first choice, the '2' next to the one that is your second...)
- Efficient handling of LARGE ballots.
The ballots are *all* printed large - It's in a huge font, on a sheet of thick paper twice the size of typing paper (something like 17x11 inches). - Presentation of voting options in a way that minimizes confusion and mistakes.
There isn't any confusion over "make a line by the candidate you want" that is any worse than "touch the screen by the candidate you want". (Assuming the ballots are printed correctly. Unlike that Florida fiasco where the arrows and the candidates didn't match up because they were printed seperately and then bound together, in this case it's all printed on one paper.
The system you describe sounds exactly like the one used in the precinct where I vote (Fitchburg, Wisconsin). More than likely they are the same company that made the machines. I too have used these machines and thought - "This is so easy, and it's the way that makes the most sense. The ballot is paper, is kept for manual recounts, but the tallies are counted by a computer scanner and electronically available immediately - best of both worlds.
I am not sure what the system does when you fill in the arrow for "other" and write a candidate's name down. I would hope it sorts the ballot into a seperate pile for a human to look at instead of just ignoring it, or counting all write-ins as the same collective vote.
(Actually, now that I think of it, counting all write-ins as a collective single candidate would kind of work in a way - if "Write-in" collectively wins the election, then that means they would have to manually go back and look at all the ballots and see who everybody wrote down (in which case maybe "write-in" wouldn't actually win if "write-in" was split several ways, but at least the need for the recount would be known and thus the real tally could be eventually determined.) If "write-in" collectively doesn't win the election, then it doesn't matter who people wrote for their write-in, because if all of them combined together into one total didn't "win", then none of them individually would have enough votes to win either.
How timely. I recently wrote an essay (read: rant) on why E-Voting is inevitable, and why we should all just suck it up and work to make the system better, instead of fighting it and trying to preserve an antiquated and inadequate pen-and-paper system.
I don't think anyone really has been arguing against electronic voting. They've been arguing against *BAD* eleectronic voting systems, and saying paper b allots are better than a *bad* electronic system, not that paper ballots are better than *all* possible electronic systems.
Did you even read the responses? The reason the poster said the court of public opinion is more important is because the poster wasn't talking about the verdict, or the legal damages, but the public relations damages.
Even today you "membership" in the Free Software Movement depends on your acceptance of a particular political ideology.
No, it doesn't. I do agree that the PERCEPTION that this is the case did hold free software back quite a bit, and that a new term was needed to get around that. But that perception was a misconception.
"Open source" was coined because it doesn't have the same problem that "free" does, that being the inability to effectively communicate that it's about "free as in speech" not "free as in beer". Saying "open source" gets the idea across that the speaker is talking about being able to openly access the source code, not about getting the program for no charge. Now, most of the time a free program has *both* features - that you get the source with it *and* that the program is also distributed at no charge. That just added to the confusion, because trypically *both* meanings of "free" were applicable. It became important to have a term to specify which aspect of the software you are talking about - the price or the openness of the code. Consider: Under the "gratis" definition of "free", Internet Explorer is "free software". It's because of this sort of confusion over the word "free" (that the closed-source companies were deliberately taking advantage of in their FUD), that ESR felt the need to coin a new term.
IBM isn't squeaky clean on this either. They often patent ideas that are so obvious that the only reason people haven't patented them before is because they never thought it would even be an issue. (For example, they patented the idea in the '80s, already in common use in a variety of programs, of making a graphical cursor on a screen without using a 'save mask', and instead using bitwise XOR on the color codes - XOR the color once to make the cursor appear, and XOR the color again to reset the screen under it to the same colors again. Pretty simple stuff - everyone was doing it - and IBM went and patented it.)
But what the grandparent poster doesn't realize is that IBM of today and IBM of yesterday are very different kinds of company. They had to change a lot of their ways to survive in the days where everyone hated big iron. I think that if MS is ever brought to the humble position it deserves, it will probably end up being like that - a change into a new kind of company rather than the company dying altogether.
Ugly though it may be, MS did invent the FAT system. So bitching about patent infringement when others use the same system is within their rights (but a really stupid thing to do since it makes it harder for companies to make devices that interoperate nicely with their OS). If they finally get around to catching up to what unix did over a decade ago, by making a desktop pager, then they can patent that, but only THAT specific implementation of it. If they try to push for patent infringement on the likes of fvwmpager, then that would NOT be like what they do with the FAT system. It would be like them trying to go after anyone who makes any sort of filesystem of any kind, which isn't what they are doing. (So what I'm trying to say is that the analogy doesn't really hold.)
Now, I think there *is* a problem with using patents to deliberately make open interoperability harder, which is what they are doing with FAT, it's not the same kind of problem as the one of claiming credit for prior work, which is what people are bitching about with this.
There is certainly a baseline amount of services you can allow for it to operate as a server on the network, while having services unneeded for that base funtionality either uninstalled or disabled by default.
No, there isn't. For example, I would consider some kind of remote commandline access to be a part of that bare minimum. Microsoft does not.
You (and most other linux people) are trying to stick to your guns with the unrealistic idea that all linux users should be experts.
Get off your high horse. I said no such thing, and I implied no such thing.
No, they said that indirect attacts, such as viruses and worms, were not counted.
And those happen to be precisely the type that are more common on Windows. So it's a biased count. Hopefully that wont be the case, but until an OS is secure by default (which your average Linux distro is not)
The only way to make an OS secure by default is to make it do absolutely nothing by default. Functionality and security are contrary. The purpose of security is to limit the functionality the system provides, such that only the functionality you want to have is present and no more. Since different customers want different functionality, there is no such thing as a one-size-fits-all secure OS configuration. That is why you *don't* get a secure configuration on an OS designed for 'average joes'. "Average Joes" will think that having to "jump through hoops" to enable a feature they want equates to a bad OS. This is what leads to misfeatures like automatically executing attachments when you click on them (instead of defaulting to saying the file like it *should* do.) Misfeatures like that exist specifically to placate the non-expert users.
Lets ignore the fact that I do computer security or a living
Do you understand basic statistics? The population of all server installations does not contain an equal number of each type of OS, not even close. So the fact that MacOS had the least hack incidents out of that population doesn't necessarily mean it's more secure. It could just be because it's a smaller sliver of the population being looked at.
Without the information about how many OSes of each type *weren't* cracked in the population, their numbers mean nothing.
Now, if they had done something like pick 10,000 random Linux servers, and 10,000 random Windows servers, and 10,000 random MacOS servers, and 10,000 random BSD servers, and so on, and then compared the break-in counts from *those*, that might have meant something useful.
And saying that Windows servers arent a target is laughable. Every script kiddie with an internet connection tries to break into Windows.
The article explicitly mentioned that automated attacks were not counted.
Slashdot Mirror
While I agree that the poster was being stupid, there is one problem that still persists with nuclear waste - while it can be stored safely, every single form of transport known to man has safety problems. If the waste dump is a thousand miles away from the plant, then the transport of the waste across that thousand miles is the really risky part.
It's for the same reason dirt doesn't stay on the top of the road. Rain and wind. The radioactive dust particles that were thrown into the air washed off the road, but where they hit earth, they became part of the soil just like any other dust would.
Technically it's not brighter than the sun. It's a lot dimmer - it's just a hell of a lot CLOSER than the sun.
I hate it when people put things they know are controversial in signatures. For some reason moderators think the post with the sig is on topic, but anyone responding to the sig is off-topic. That's really unfair. It means if you want to get away with propaganda - all you have to do is put it in your sig instead of in your post's body and it's considered taboo for anyone to question you on it.
Oddly enough this is one case where Microsoft is actually more correct according to the standards than their competitors. The ASCII meaning of carriage return is supposed to be *just* moving the cursor to column zero of the current line. The ASCII meaning of linefeed is supposed to be just to scroll down a line without moving the cursor horizontally. So the combination of both of them actually fits what happens at end-of-line. And of course, most TCP/IP protocols, which were certainly not designed with Windows in mind, say the same thing - lines end with a CR/LF pair.
It does make writing a program to read text files a bit annoying, though, doesn't it? I mean, how often have you *just* wanted to linefeed or *just* carriage return in a text file? How come the only place MS adheres to the standard is in a place where the standard is a bit stupid?
If the whistleblower is legitimate (and I'm leaving that decision up for grabs in my mind still), then whoever it is would have no difficulties in life after having been fired from SCO. The publicity would help in a number of areas. Yes, some companies wouldn't want to hire a known whistleblower, but others would be quite grateful, especially companies SCO was suing.
A little less, in reality - because if you have 42.6 million in investments, the moment you start selling them off like mad, your investments start losing value before you get around to selling off the rest of them. Other owners of stock start thinking, "holy crap, what do they know that we don't?", and they start selling too, lowering the prices. If they tried using that 42.6 million as if it was liquid cash, it would be worth a lot less than 42.6 million by the time they extracted all of it.
1. In a car you don't *want* too much stiffness in the frame. They are designed to crumple so when you run into something you are more likely to ruin the car instead of your body. (The car crumples to soften the impulse of the hit.)
2. The hood is irrelevant to the strength of the car anyway, since the strength doesn't come from the body but from the skeletal frame the body is attached to.
While I hate Microsoft for all the same reasons most of you do, I think in this case there's a non-malicious explanation. That redirection is probably triggered by the following conditions:
(1) The search term contains certain things that tend to find X-rated content. The algorithm might look for dictionary terms and try to form seperate words out of the serch phrase (so if you looked for hothornybabes it would notice that it contains the words "hot" and "horny" and "babes".) So, "Xfree86" probably gets flagged because it's "X" followed by "free" and some irrelevant number. But, wait, you say 'Xfree87" and "Xfree85" don't trigger, so that can't be it, right? Well, it still could be because of the next point:
2) It probably *also* only triggers the redirection if the search result returns a lot of hits.
So 'Xfree86' triggers a lot of hits, and contains red-flagged terms, while 'Xfree87' has the same flagged terms, but triggers few hits, and so isn't assumed to be porn.
Anyway, that's one possible explanation. I'd attribute this to stupidity on the part of the algorithm before attributing it to maliciousness.
Good idea, but pick somebody dead. It's less dangerous because they can't actually take office, but it definately proves to the public that the system is broken. (But be prepared to spend a *LOONG* time in jail over it if it gets discovered. The government doesn't take kindly to people proving it's stupid.)
What do you mean "would be"? It already *is* the way it works in a lot of voting districts. I've voted that way in the last 3 presidential elections. Yes, the systems used in Florida were terrible and needed fixing - but there were already good systems (like the above system) that could have been used, instead of this push toward unverifiable trust in a company to not rig an election count.
Results of French elections are usually known a few hours after the votes, and after-voting polls usually give the result right at closure time.
Given that the US covers several time zones, I really wish they'd either stagger the vote times so that all the polls close at the same exact time regardless of what the local time may be, or put a moratorium on reporting results before all the polls are closed. It really annoys me that the news shows start announcing predictions of elections while some people can still vote yet. Everyone's vote should be just as "blind" as everyone else's - nobody should get to vote using extra information that was available to them and not to others, just because they live farther to the west.
- Tallying of complicated voting system. (I.e. preferential voting.)
If that was needed, then filling in numbers would work just as well (fill in the '1' circle next to the candidate which is your first choice, the '2' next to the one that is your second...)
- Efficient handling of LARGE ballots.
The ballots are *all* printed large - It's in a huge font, on a sheet of thick paper twice the size of typing paper (something like 17x11 inches).
- Presentation of voting options in a way that minimizes confusion and mistakes.
There isn't any confusion over "make a line by the candidate you want" that is any worse than "touch the screen by the candidate you want". (Assuming the ballots are printed correctly. Unlike that Florida fiasco where the arrows and the candidates didn't match up because they were printed seperately and then bound together, in this case it's all printed on one paper.
The system you describe sounds exactly like the one used in the precinct where I vote (Fitchburg, Wisconsin). More than likely they are the same company that made the machines. I too have used these machines and thought - "This is so easy, and it's the way that makes the most sense. The ballot is paper, is kept for manual recounts, but the tallies are counted by a computer scanner and electronically available immediately - best of both worlds.
I am not sure what the system does when you fill in the arrow for "other" and write a candidate's name down. I would hope it sorts the ballot into a seperate pile for a human to look at instead of just ignoring it, or counting all write-ins as the same collective vote.
(Actually, now that I think of it, counting all write-ins as a collective single candidate would kind of work in a way - if "Write-in" collectively wins the election, then that means they would have to manually go back and look at all the ballots and see who everybody wrote down (in which case maybe "write-in" wouldn't actually win if "write-in" was split several ways, but at least the need for the recount would be known and thus the real tally could be eventually determined.) If "write-in" collectively doesn't win the election, then it doesn't matter who people wrote for their write-in, because if all of them combined together into one total didn't "win", then none of them individually would have enough votes to win either.
How timely. I recently wrote an essay (read: rant) on why E-Voting is inevitable, and why we should all just suck it up and work to make the system better, instead of fighting it and trying to preserve an antiquated and inadequate pen-and-paper system.
I don't think anyone really has been arguing against electronic voting. They've been arguing against *BAD* eleectronic voting systems, and saying paper b allots are better than a *bad* electronic system, not that paper ballots are better than *all* possible electronic systems.
Did you even read the responses? The reason the poster said the court of public opinion is more important is because the poster wasn't talking about the verdict, or the legal damages, but the public relations damages.
Even today you "membership" in the Free Software Movement depends on your acceptance of a particular political ideology.
No, it doesn't. I do agree that the PERCEPTION that this is the case did hold free software back quite a bit, and that a new term was needed to get around that. But that perception was a misconception.
"Open source" was coined because it doesn't have the same problem that "free" does, that being the inability to effectively communicate that it's about "free as in speech" not "free as in beer". Saying "open source" gets the idea across that the speaker is talking about being able to openly access the source code, not about getting the program for no charge. Now, most of the time a free program has *both* features - that you get the source with it *and* that the program is also distributed at no charge. That just added to the confusion, because trypically *both* meanings of "free" were applicable. It became important to have a term to specify which aspect of the software you are talking about - the price or the openness of the code. Consider: Under the "gratis" definition of "free", Internet Explorer is "free software". It's because of this sort of confusion over the word "free" (that the closed-source companies were deliberately taking advantage of in their FUD), that ESR felt the need to coin a new term.
IBM isn't squeaky clean on this either. They often patent ideas that are so obvious that the only reason people haven't patented them before is because they never thought it would even be an issue. (For example, they patented the idea in the '80s, already in common use in a variety of programs, of making a graphical cursor on a screen without using a 'save mask', and instead using bitwise XOR on the color codes - XOR the color once to make the cursor appear, and XOR the color again to reset the screen under it to the same colors again. Pretty simple stuff - everyone was doing it - and IBM went and patented it.)
But what the grandparent poster doesn't realize is that IBM of today and IBM of yesterday are very different kinds of company. They had to change a lot of their ways to survive in the days where everyone hated big iron. I think that if MS is ever brought to the humble position it deserves, it will probably end up being like that - a change into a new kind of company rather than the company dying altogether.
Ugly though it may be, MS did invent the FAT system. So bitching about patent infringement when others use the same system is within their rights (but a really stupid thing to do since it makes it harder for companies to make devices that interoperate nicely with their OS). If they finally get around to catching up to what unix did over a decade ago, by making a desktop pager, then they can patent that, but only THAT specific implementation of it. If they try to push for patent infringement on the likes of fvwmpager, then that would NOT be like what they do with the FAT system. It would be like them trying to go after anyone who makes any sort of filesystem of any kind, which isn't what they are doing. (So what I'm trying to say is that the analogy doesn't really hold.)
Now, I think there *is* a problem with using patents to deliberately make open interoperability harder, which is what they are doing with FAT, it's not the same kind of problem as the one of claiming credit for prior work, which is what people are bitching about with this.
If you're paying $1000 a month in rent, you aren't poor. If you're poor you'll move somewhere cheaper than that.
There is certainly a baseline amount of services you can allow for it to operate as a server on the network, while having services unneeded for that base funtionality either uninstalled or disabled by default.
No, there isn't. For example, I would consider some kind of remote commandline access to be a part of that bare minimum. Microsoft does not.
You (and most other linux people) are trying to stick to your guns with the unrealistic idea that all linux users should be experts.
Get off your high horse. I said no such thing, and I implied no such thing.
No, they said that indirect attacts, such as viruses and worms, were not counted.
And those happen to be precisely the type that are more common on Windows. So it's a biased count.
Hopefully that wont be the case, but until an OS is secure by default (which your average Linux distro is not)
The only way to make an OS secure by default is to make it do absolutely nothing by default. Functionality and security are contrary. The purpose of security is to limit the functionality the system provides, such that only the functionality you want to have is present and no more. Since different customers want different functionality, there is no such thing as a one-size-fits-all secure OS configuration. That is why you *don't* get a secure configuration on an OS designed for 'average joes'. "Average Joes" will think that having to "jump through hoops" to enable a feature they want equates to a bad OS. This is what leads to misfeatures like automatically executing attachments when you click on them (instead of defaulting to saying the file like it *should* do.) Misfeatures like that exist specifically to placate the non-expert users.
If you wait to solve all problems before exploring further, you never end up exploring further - EVER.
Lets ignore the fact that I do computer security or a living
Do you understand basic statistics? The population of all server installations does not contain an equal number of each type of OS, not even close. So the fact that MacOS had the least hack incidents out of that population doesn't necessarily mean it's more secure. It could just be because it's a smaller sliver of the population being looked at.
Without the information about how many OSes of each type *weren't* cracked in the population, their numbers mean nothing.
Now, if they had done something like pick 10,000 random Linux servers, and 10,000 random Windows servers, and 10,000 random MacOS servers, and 10,000 random BSD servers, and so on, and then compared the break-in counts from *those*, that might have meant something useful.
And saying that Windows servers arent a target is laughable. Every script kiddie with an internet connection tries to break into Windows.
The article explicitly mentioned that automated attacks were not counted.