Usually the copyright problems don't come from the games themselves, but from the ROM images. For example you can download Elite but not the BBC Micro MOS and Basic ROMs needed to play it.
Amstrad behaved very well here - many years ago they released the Spectrum ROM code into the public domain, or at least allowed people to use it. But Acorn (the makers of the BBC Micro) kept a tight grip on the ROMs from 1981. Some say this was because the operating system on their 32-bit Archimedes machines (Arthur, which later became RISC OS) was originally a straight port of the BBC Micro's operating system, and had many features in common. But that's a pretty poor reason to stop people using the original 8-bit code for a machine which hasn't been sold since 1986.
Acorn kept this up until their eventual demise a couple of years ago. I don't know who inherited the copyright to the BBC ROMs - probably RISC OS Limited (who almost develop newer versions of RISC OS) or Pace Micro Technology (who make set-top boxes). They could do the world a favour, and themselves no harm, by letting emulator users run the original ROM images.
This could be useful for a console which runs Linux. Store an individual copy of the OS on each game CD; of course it is already set up for exactly the hardware you have. Having a central copy of the operating system stored in the console itself would be too much work to maintain.
In any case, if you buy a game with Linux already on the CD, nobody says you _have_ to use that copy of Linux. If you're part of the unlucky 20% whose hardware is not detected correctly, you can just run it from your own installation.
Yes. That's exactly what it means. Well actually, it doesn't mean you're a worse driver, just more 'likely' (whatever that means) to have an accident and make a claim.
Discriminating against (or in favour of) an individual based on what similar individuals have done in the past is actually fundamental to society, and I'm surprised people don't notice this.
For example: a university offers a maths degree. After a few years they notice that people who didn't get an A grade in maths at school tend to fail the course. Not all of them, but most of them. Then two students apply for one place - one of them got an A in his exam and the other a B. Assuming other factors are equal, who gets the place?
Maybe 'student B' would actually do better at the course than student A - he could be a mathematical genius who just didn't get on well with the limited school course. (Everyone knows about Einstein's school maths achievement.) It is grossly unfair to judge this one student just based on what other people happened to do in the past - people he is totally unrelated to except by getting the same grade. It's by no means certain that the B-grade student will fail the course - but I'm sure you'll admit it's more likely.
>So would you outlaw no-claimes bonuses?
Yes. This is ridiculous: Someone hits me, and my insurance premiums go up because a claim was filed. Your insurance rates should never be increased because of something that was beyond your control.
So... if you were an insurer, and you had two customers...
Customer A: 'I have never made a claim on my insurance.'
Customer B: 'I've made five claims over the past ten years, but I can prove none of them were my fault.'
Can you honestly say that you'd be just as willing to insure B as you are to insure A at the same price? Would it be too much to admit that just maybe, B is more accident-prone than A? Perhaps he drives more often, or on more dangerous roads. Perhaps he's just unfortuate because he lives in an area full of idiotic drivers. But wouldn't any of these factors increase the risk?
I sympatize with you over the fence episode. All I can say is: choose a different insurer, or don't rely on the no-claims bonus. And remember that the company isn't obliged to sell you insurance at any particular price, any more than a shop is obliged to sell you goods at a particular price.
>Lower car insurance premiums for women and older drivers?
I'd outlaw those in a heartbeat. The statistics that say women and older drivers are less likely to get into accidents are pure bullshit.
A lot of insurance companies seem to disagree. They employ a lot of people whose job it is to work out how risky particular groups are, and they seem to have all reached the same conclusion. Are you saying that all the insurance companies are wrong, and that differences in premiums are a marketing gimmick? What research have you done in this area?
A major factor is that women and old people drive LESS, so no wonder they're going to get into less accidents.
Right. They do have fewer accidents. You admit that this is a fact.
Insurance premiums aren't based on any sort of principles or sense of fairness. They're based on how much the company can get away with charging, which in a competitive market means how likely the customer is to make a claim. You may not like the fact that women or older people have fewer driving accidents than young men, but you'll just have to accept it.
How would you deal with the problem of somebody having a genetic test, finding out they're likely to get a particular illness, and then buying life insurance before any symptoms show? If that happened a lot, it would push premiums through the roof, or more likely make the whole life insurance business impossible.
Most of the issue you cite are based on observed behaviour of a (potential) client.
No, they're based on observed behaviour of a particular group of people. Lower premiums for a new woman driver than for a man are based on the past record of other women. Of course there's nothing to say that the past record of a particular class of people will tell you anything about one member of that class, but in practice it works out pretty well.
So imagine a male driver who has just bought his first car and wants insurance. The high premiums he pays cannot be based on his past behaviour - he has never driven before except in lessons and the driving test. Yet it is a fact that young men are the most likely to have accidents, and that higher premiums are needed for them to be profitable customers. It's not this driver's 'fault' that he belongs to a high risk group. He may even be a very careful driver. It's not 'fair'. But why should everyone else pay extra to underwrite the risks of one particular group? And they won't, unless you pass laws to restrict everyone to buying from 'approved', non-discriminatory insurance companies.
It's similar with genetic predisposition to particular illnesses - it's not based on past actions, it's not anybody's fault, but indisputably there is a higher risk.
I chose to live in a low crime area.
You mean you can afford to.
Car insurance is ten times the cost (for me) in the UK that it is in New Zealand, and the main difference is that I'm required by law to have it in the UK and I'm not in New Zealand.
Maybe in New Zealand only the naturally cautious, risk-averse type of person buys car insurance. These people are probably safer drivers than the average.
So would you outlaw no-claimes bonuses? Lower car insurance premiums for women and older drivers? How about insurance 'discrimination' based on whether you live in a high-crime or low-crime area?
Why do you need to transfer physical hardware at all? Just stick some kind of Unix on the machine and give away remote logins to people who ask for it. They can have all the fun of using an old machine, without the hassle of having to go and collect it!
Re:The ultimate win/lin compatibility already exis
on
User Mode Linux
·
· Score: 2
It might be possible to re-link a Linux binary against the Cygwin libraries instead of glibc, and then run it. Like how Wine links Windows binaries against its own implementation of Win32 - I think. I'm not a binutils guru, so could someone say whether this would be possible?
What 'European version of Red Hat 7'? What's the difference between that and the US version?
(Any large institution will have users whose native language is not English... so IMHO it's best to always install internationalized versions of everything, if you have the disk space. So why not a single distribution where you pick the language at install time? Or is there a separate European version for legal reasons?)
Bandwidth isn't free, but it's cheap enough. Mail - even with lots of dummy messages flying about - uses a fraction of the bandwidth of Web browsing with images, downloading software, or any of the more whizzy things you might do. So cost isn't a big issue, unless you need to buy a new mail server.
It would be better if the Java source were downloaded to your browser and then compiled locally. Your browser could check that the source really is unchanged from the 'source code' posted on Hushmail's site (which you keep locally). (In fact, when Java first came out I assumed that applets would be human-readable, like HTML and pretty much everything else on the Web.)
Or you could check the binary class files - 'only run this applet if the class file matches file X' - but who is going to decompile the Java bytecode and check it does what it claims to do?
You should be able to send 'dummy' encrypted messages to people. In other words the string 'DUMMY' followed by some random-length amount of random data, encrypted with the recipient's public key. Of course the recipient's MUA would be configured to silently discard all such messages. And somebody looking at the encrypted message en route could not tell whether it was a dummy or not.
If you set up your mailreader to harvest addresses and public keys off the net and send a few dummy messages each day, it would be harder to find out which people you are sending real messages to. You could even send a dummy message several times a day to one particular address, like the police. Then if you really needed to contact the police, you could do so without eavesdroppers becoming suspicious.
This isn't spam because it doesn't waste any of the recipient's time - only a small amount of bandwidth. If anything it increases the recipient's security because anyone trying to brute-force their mail or monitor who they are getting messages from will have to wade through all the dummies.
What happens if there isn't enough space for everything at install time? Will dselect intelligently drop emacs, tex or whatever to make the system fit on your disk?
(I'm fed up with RedHat's installer which will happily start installing 600Mbyte of stuff on a 300Mbyte partition, even though it knows the size before starting. I'm sure Debian isn't as stupid.)
How can PS/2 keyboard and mouse ports be 'part of ISA'? Most PS/2s are MCA-based! Serial and parallel ports, mouse and keyboard ports, and ISA slots are three separate issues. You can get rid of one without the others.
Slashdot Mirror
C:\ -> root#
====
email -> mail
emails -> messages
Not that I'd consider tchrist's opinion definitive :-)
You send some mail, or you send three messages.
Usually the copyright problems don't come from the games themselves, but from the ROM images. For example you can download Elite but not the BBC Micro MOS and Basic ROMs needed to play it.
Amstrad behaved very well here - many years ago they released the Spectrum ROM code into the public domain, or at least allowed people to use it. But Acorn (the makers of the BBC Micro) kept a tight grip on the ROMs from 1981. Some say this was because the operating system on their 32-bit Archimedes machines (Arthur, which later became RISC OS) was originally a straight port of the BBC Micro's operating system, and had many features in common. But that's a pretty poor reason to stop people using the original 8-bit code for a machine which hasn't been sold since 1986.
Acorn kept this up until their eventual demise a couple of years ago. I don't know who inherited the copyright to the BBC ROMs - probably RISC OS Limited (who almost develop newer versions of RISC OS) or Pace Micro Technology (who make set-top boxes). They could do the world a favour, and themselves no harm, by letting emulator users run the original ROM images.
This could be useful for a console which runs Linux. Store an individual copy of the OS on each game CD; of course it is already set up for exactly the hardware you have. Having a central copy of the operating system stored in the console itself would be too much work to maintain.
In any case, if you buy a game with Linux already on the CD, nobody says you _have_ to use that copy of Linux. If you're part of the unlucky 20% whose hardware is not detected correctly, you can just run it from your own installation.
Are there any Linux distributions that come with support for this out of the box?
Yes. That's exactly what it means. Well actually, it doesn't mean you're a worse driver, just more 'likely' (whatever that means) to have an accident and make a claim.
Discriminating against (or in favour of) an individual based on what similar individuals have done in the past is actually fundamental to society, and I'm surprised people don't notice this.
For example: a university offers a maths degree. After a few years they notice that people who didn't get an A grade in maths at school tend to fail the course. Not all of them, but most of them. Then two students apply for one place - one of them got an A in his exam and the other a B. Assuming other factors are equal, who gets the place?
Maybe 'student B' would actually do better at the course than student A - he could be a mathematical genius who just didn't get on well with the limited school course. (Everyone knows about Einstein's school maths achievement.) It is grossly unfair to judge this one student just based on what other people happened to do in the past - people he is totally unrelated to except by getting the same grade. It's by no means certain that the B-grade student will fail the course - but I'm sure you'll admit it's more likely.
So... if you were an insurer, and you had two customers...
Customer A: 'I have never made a claim on my insurance.'
Customer B: 'I've made five claims over the past ten years, but I can prove none of them were my fault.'
Can you honestly say that you'd be just as willing to insure B as you are to insure A at the same price? Would it be too much to admit that just maybe, B is more accident-prone than A? Perhaps he drives more often, or on more dangerous roads. Perhaps he's just unfortuate because he lives in an area full of idiotic drivers. But wouldn't any of these factors increase the risk?
I sympatize with you over the fence episode. All I can say is: choose a different insurer, or don't rely on the no-claims bonus. And remember that the company isn't obliged to sell you insurance at any particular price, any more than a shop is obliged to sell you goods at a particular price.
A lot of insurance companies seem to disagree. They employ a lot of people whose job it is to work out how risky particular groups are, and they seem to have all reached the same conclusion. Are you saying that all the insurance companies are wrong, and that differences in premiums are a marketing gimmick? What research have you done in this area?
Right. They do have fewer accidents. You admit that this is a fact.
Insurance premiums aren't based on any sort of principles or sense of fairness. They're based on how much the company can get away with charging, which in a competitive market means how likely the customer is to make a claim. You may not like the fact that women or older people have fewer driving accidents than young men, but you'll just have to accept it.
How would you deal with the problem of somebody having a genetic test, finding out they're likely to get a particular illness, and then buying life insurance before any symptoms show? If that happened a lot, it would push premiums through the roof, or more likely make the whole life insurance business impossible.
No, they're based on observed behaviour of a particular group of people. Lower premiums for a new woman driver than for a man are based on the past record of other women. Of course there's nothing to say that the past record of a particular class of people will tell you anything about one member of that class, but in practice it works out pretty well.
So imagine a male driver who has just bought his first car and wants insurance. The high premiums he pays cannot be based on his past behaviour - he has never driven before except in lessons and the driving test. Yet it is a fact that young men are the most likely to have accidents, and that higher premiums are needed for them to be profitable customers. It's not this driver's 'fault' that he belongs to a high risk group. He may even be a very careful driver. It's not 'fair'. But why should everyone else pay extra to underwrite the risks of one particular group? And they won't, unless you pass laws to restrict everyone to buying from 'approved', non-discriminatory insurance companies.
It's similar with genetic predisposition to particular illnesses - it's not based on past actions, it's not anybody's fault, but indisputably there is a higher risk.
You mean you can afford to.
Maybe in New Zealand only the naturally cautious, risk-averse type of person buys car insurance. These people are probably safer drivers than the average.
So would you outlaw no-claimes bonuses? Lower car insurance premiums for women and older drivers? How about insurance 'discrimination' based on whether you live in a high-crime or low-crime area?
Why do you need to transfer physical hardware at all? Just stick some kind of Unix on the machine and give away remote logins to people who ask for it. They can have all the fun of using an old machine, without the hassle of having to go and collect it!
It might be possible to re-link a Linux binary against the Cygwin libraries instead of glibc, and then run it. Like how Wine links Windows binaries against its own implementation of Win32 - I think. I'm not a binutils guru, so could someone say whether this would be possible?
How do you pronounce XXXX?
What 'European version of Red Hat 7'? What's the difference between that and the US version?
(Any large institution will have users whose native language is not English... so IMHO it's best to always install internationalized versions of everything, if you have the disk space. So why not a single distribution where you pick the language at install time? Or is there a separate European version for legal reasons?)
Probably hidden deep within Janeway's hairdo.
What are the odds that the only playable part of this game will turn out to be the 'Captain Proton' easter egg?
Bandwidth isn't free, but it's cheap enough. Mail - even with lots of dummy messages flying about - uses a fraction of the bandwidth of Web browsing with images, downloading software, or any of the more whizzy things you might do. So cost isn't a big issue, unless you need to buy a new mail server.
Yeah, but 'reflex' is spelled with an 'x'. I think the difference is that there's no verb 'to crucifict'.
It would be better if the Java source were downloaded to your browser and then compiled locally. Your browser could check that the source really is unchanged from the 'source code' posted on Hushmail's site (which you keep locally). (In fact, when Java first came out I assumed that applets would be human-readable, like HTML and pretty much everything else on the Web.)
Or you could check the binary class files - 'only run this applet if the class file matches file X' - but who is going to decompile the Java bytecode and check it does what it claims to do?
You should be able to send 'dummy' encrypted messages to people. In other words the string 'DUMMY' followed by some random-length amount of random data, encrypted with the recipient's public key. Of course the recipient's MUA would be configured to silently discard all such messages. And somebody looking at the encrypted message en route could not tell whether it was a dummy or not.
If you set up your mailreader to harvest addresses and public keys off the net and send a few dummy messages each day, it would be harder to find out which people you are sending real messages to. You could even send a dummy message several times a day to one particular address, like the police. Then if you really needed to contact the police, you could do so without eavesdroppers becoming suspicious.
This isn't spam because it doesn't waste any of the recipient's time - only a small amount of bandwidth. If anything it increases the recipient's security because anyone trying to brute-force their mail or monitor who they are getting messages from will have to wade through all the dummies.
Well, most people write 'reflection' instead of 'reflexion' these days.
What happens if there isn't enough space for everything at install time? Will dselect intelligently drop emacs, tex or whatever to make the system fit on your disk?
(I'm fed up with RedHat's installer which will happily start installing 600Mbyte of stuff on a 300Mbyte partition, even though it knows the size before starting. I'm sure Debian isn't as stupid.)
What is the point in that? Security through obscurity? Anyone could just buy a RedHat CD and look at these files anyway.
How can PS/2 keyboard and mouse ports be 'part of ISA'? Most PS/2s are MCA-based! Serial and parallel ports, mouse and keyboard ports, and ISA slots are three separate issues. You can get rid of one without the others.