Yes, the 255-length limit on strings can be bothersome, and to address it a new dynamic string type with a 32-bit length was introduced in Delphi 2: AnsiString.
That is 7 Delphi versions ago, btw, so it's not exactly new any more:-)
However, it is a non-standard extension not supported by recent versions of Pascal on all platforms, which severely limits its usefulness.
were any UK laws (eg the Data Protection Act 1984 or the Regulation of Investigatory Powers Act 2000) broken? Are there any UK cyber law experts on Slashdot with any suggestions...?
The most important one, I suspect, is the Human Rights Act, 1998:
PART II
THE FIRST PROTOCOL
ARTICLE 1
PROTECTION OF PROPERTY
Every natural or legal person is entitled to the peaceful enjoyment of his possessions. No one shall be deprived of his possessions except in the public interest and subject to the conditions provided for by law and by the general principles of international law.
The preceding provisions shall not, however, in any way impair the right of a State to enforce such laws as it deems necessary to control the use of property in accordance with the general interest or to secure the payment of taxes or other contributions or penalties.
So, they can't confiscate property unless some other law specifically allows it. I'm not sure what that law would be.
It's a compromise. It's more difficult to modify the hardware than the software. And the software can easily be compromised without even the owner knowing it by various spyware.
But... this is still susceptible to software compromises. Keyloggers will still work on it. Memory dumps might be able to get sensitive information out of it. Sensitive information might still be written into the OS's paging file.
It's nothing like as good as it sounds like it should be.
Google does great things, but without such opposition, they might not keep all issues in proper perspective. The things they mention are very important.
The problem is, though, that this isn't Google's fault. All they're really doing is drawing attention to a problem that previously existed anyway, and the media are now shooting the messenger.
The article doesn't make it clear how these figures were acquired beyond saying that the research company studied the contents of 40,000 people's hard drives.
Questions to ask:
1. How were these 40,000 people selected? 2. Did they know what exactly was being analysed on their computer? 3. Were there any requirements placed on them? For instance, were only Windows users selected because they had to run some analysis software that only works there?
My main concern is that if these people knew that their usage of digital music files was being studied, then they would be less likely to get involved in unauthorised file sharing, which is, after all, an illegal activity that most people would not want to be monitored doing. Therefore their music files are more likely than average to have been acquired through legitimate means.
Re:Doesn't work. Sorry, do not collect $200.
on
Replacing TCP?
·
· Score: 1
I'd try it myself, but it crashes with "Illegal Instruction" on my old AMD K6 server. ("cmove %eax, %edx", apparently)
Re:Update TCP, don't add new protocol
on
Replacing TCP?
·
· Score: 1
The problem is that this protocol works in an entirely different way to TCP, because it is designed to solve a problem that TCP doesn't work very well with -- unreliable connections that drop packets frequently and have long latencies. I really can't see how TCP can be fixed to work much better in this circumstance, at least not without dropping core aspects of the way it works.
Re:So how does redundancy improve transfer rate?
on
Replacing TCP?
·
· Score: 1
While I haven't looked at their protocol in depth (as the RFCs promised don't seem to be available yet), one possible way of gaining efficiency is if they don't send back (as many?) acknowledgement packets. You're also potentially freed from concerns about window size, which on links with long latencies can seriously slow down TCP transmissions to well under the capacity of the link.
I want to know how they get data transmission at 100% loss!
Oh, that's easy. You just infinite time.
Re:A working wikipedia link for Kademlia
on
Replacing TCP?
·
· Score: 1
The link in the post is coralised. Maybe someone is blocking your outgoing port 8090 for you:-)
Actually, when I was trying it about 5 minutes ago, the service seemed to have been slashdotted (I was getting an error back from the coral server I connected to).
Why do you say you'd have to pay for it? There are free downloads on their web page of code for a very wide variety of operating systems, and the article summary above states they're submitting RFCs for it, so pretty soon it ought to be an open standard, available for anyone to implement.
The question is, though, do we need such things in our cars? Do we really need cars equipped with "home entertainment systems," if these cars will only be used for trips to the nearest shop or driving to work?
Of course we don't _need_ them. However, people want them and are will to pay extra money for them -- a guy I work with runs a business installing this kind of stuff. People will pay a small fortune to keep the kids quiet in the back seats...
Practically, of course, this depends on the browsers you're targetting. According to Microsoft's reference, for instance, the only strings supported by Internet Explorer are "JScript", "javascript", "vbs", "vbscript", and "XML". Seems they've never heard of the HTML 4 recommendation.;)
The cost of hardware and software have nothing to do with it. If there's a way to get a "free copy", some people will always go that route.
They have something to do with it. There are many people who do not enjoy doing what they consider immoral (e.g. software piracy) but consider it a "necessary evil" because they cannot afford to pay for this software.
I usually try to persuade them down the free software route, but that isn't for everyone. Cheaper hardware might allow these people to be able to spend more money on software. But I doubt it -- they have a lot of other things to distribute what they'd save around.
And, yeah, there will always be people who do what they want because they don't see why they shouldn't.
Even if the Taliban were a recognized government, they'd still fail because they didn't have uniforms. (A pedantic point? Sure. But that's law for you; law is nothing more than the rigorous application of pedantism.)
Pedantic? I don't think so. The idea behind this is that if non-uniformed soldiers are not guaranteed the right to be treated in a civilised fashion, this ought to encourage soldiers to wear uniforms. Soldiers wearing uniforms is a good thing, because it reduces the chance that civilians in a war zone will be confused with soldiers and shot by mistake. Call me old fashioned, but I think any step that may protect civilians during a war is a good one.
My biggest problem with the special edition versions was that you could no longer see the stormtrooper banging his head on a door in Episode 4. Took the fun right out of it for me, that did.;)
Personally, I would have thought that after last time they would have upgraded their server before attempting to host the continued story on it. _That_ would be common sense. They must have _known_ that it would attract the same kind of attention as the last installment.
Slashdot Mirror
Yes, the 255-length limit on strings can be bothersome, and to address it a new dynamic string type with a 32-bit length was introduced in Delphi 2: AnsiString.
:-)
That is 7 Delphi versions ago, btw, so it's not exactly new any more
However, it is a non-standard extension not supported by recent versions of Pascal on all platforms, which severely limits its usefulness.
The most important one, I suspect, is the Human Rights Act, 1998:
So, they can't confiscate property unless some other law specifically allows it. I'm not sure what that law would be.
Err. Yeah. If you start messing around with the hardware, there's not a lot that can't be subverted, to be honest with you.
It's a compromise. It's more difficult to modify the hardware than the software. And the software can easily be compromised without even the owner knowing it by various spyware.
But... this is still susceptible to software compromises. Keyloggers will still work on it. Memory dumps might be able to get sensitive information out of it. Sensitive information might still be written into the OS's paging file.
It's nothing like as good as it sounds like it should be.
Google does great things, but without such opposition, they might not keep all issues in proper perspective. The things they mention are very important.
The problem is, though, that this isn't Google's fault. All they're really doing is drawing attention to a problem that previously existed anyway, and the media are now shooting the messenger.
Presumably, if they were concerned, they'd have encrypted their files.
WTF? Get a _LIFE_.
The article doesn't make it clear how these figures were acquired beyond saying that the research company studied the contents of 40,000 people's hard drives.
Questions to ask:
1. How were these 40,000 people selected?
2. Did they know what exactly was being analysed on their computer?
3. Were there any requirements placed on them? For instance, were only Windows users selected because they had to run some analysis software that only works there?
My main concern is that if these people knew that their usage of digital music files was being studied, then they would be less likely to get involved in unauthorised file sharing, which is, after all, an illegal activity that most people would not want to be monitored doing. Therefore their music files are more likely than average to have been acquired through legitimate means.
I'd try it myself, but it crashes with "Illegal Instruction" on my old AMD K6 server. ("cmove %eax, %edx", apparently)
The problem is that this protocol works in an entirely different way to TCP, because it is designed to solve a problem that TCP doesn't work very well with -- unreliable connections that drop packets frequently and have long latencies. I really can't see how TCP can be fixed to work much better in this circumstance, at least not without dropping core aspects of the way it works.
While I haven't looked at their protocol in depth (as the RFCs promised don't seem to be available yet), one possible way of gaining efficiency is if they don't send back (as many?) acknowledgement packets. You're also potentially freed from concerns about window size, which on links with long latencies can seriously slow down TCP transmissions to well under the capacity of the link.
I want to know how they get data transmission at 100% loss!
Oh, that's easy. You just infinite time.
The link in the post is coralised. Maybe someone is blocking your outgoing port 8090 for you :-)
Actually, when I was trying it about 5 minutes ago, the service seemed to have been slashdotted (I was getting an error back from the coral server I connected to).
It has recovered now, though.
well, assuming that the dropped frames aren't sequential in large number
My experience is that this isn't usually the case. How about yours?
Why do you say you'd have to pay for it? There are free downloads on their web page of code for a very wide variety of operating systems, and the article summary above states they're submitting RFCs for it, so pretty soon it ought to be an open standard, available for anyone to implement.
Yeah. In your user preferences you can specify a bonus to add to their score. Make it -6.
The question is, though, do we need such things in our cars? Do we really need cars equipped with "home entertainment systems," if these cars will only be used for trips to the nearest shop or driving to work?
Of course we don't _need_ them. However, people want them and are will to pay extra money for them -- a guy I work with runs a business installing this kind of stuff. People will pay a small fortune to keep the kids quiet in the back seats...
Practically, of course, this depends on the browsers you're targetting. According to Microsoft's reference, for instance, the only strings supported by Internet Explorer are "JScript", "javascript", "vbs", "vbscript", and "XML". Seems they've never heard of the HTML 4 recommendation. ;)
The cost of hardware and software have nothing to do with it. If there's a way to get a "free copy", some people will always go that route.
They have something to do with it. There are many people who do not enjoy doing what they consider immoral (e.g. software piracy) but consider it a "necessary evil" because they cannot afford to pay for this software.
I usually try to persuade them down the free software route, but that isn't for everyone. Cheaper hardware might allow these people to be able to spend more money on software. But I doubt it -- they have a lot of other things to distribute what they'd save around.
And, yeah, there will always be people who do what they want because they don't see why they shouldn't.
It's a sad day on slashdot when Bourne gets a reference to the book/movie but not the shell.
Perhaps we need a new sequel. Let's call it the Bourne Again Identity.
Even if the Taliban were a recognized government, they'd still fail because they didn't have uniforms. (A pedantic point? Sure. But that's law for you; law is nothing more than the rigorous application of pedantism.)
Pedantic? I don't think so. The idea behind this is that if non-uniformed soldiers are not guaranteed the right to be treated in a civilised fashion, this ought to encourage soldiers to wear uniforms. Soldiers wearing uniforms is a good thing, because it reduces the chance that civilians in a war zone will be confused with soldiers and shot by mistake. Call me old fashioned, but I think any step that may protect civilians during a war is a good one.
My biggest problem with the special edition versions was that you could no longer see the stormtrooper banging his head on a door in Episode 4. Took the fun right out of it for me, that did. ;)
Personally, I would have thought that after last time they would have upgraded their server before attempting to host the continued story on it. _That_ would be common sense. They must have _known_ that it would attract the same kind of attention as the last installment.
I don't know about you, but I've seen much worse. I had to read your highlighted section 3 times before I noticed the incongruity of it.
I think you want either 'zcat' or 'gunzip -c'.