Re:and the seller...
on
Online Revenge
·
· Score: 4, Informative
Truth is an absolute defence to libel in England and Wales and I presume Northern Ireland (I don't know about Scotland). The problem is the reversed burden of proof: the defendent has to prove truth, albeit only to civil standards.
library patrons shouldn't really be looking up porn. Nor playing Flash games. Nor watching movies.
A lending library will lend you a copy of The Da Vinci Code. It will lend up a CD audiobook of The Da Vinci Code. Why shouldn't it offer you the movie as well?
In fact, he performed his demonstration of the unstopability of the tides not to prove his own strength, but to prove to his courtiers that he didn't have divine powers. http://en.wikipedia.org/wiki/Canute_the_Great.
So yeah. My point is that Virtualization will not destroy the native OS market. Not a chance in hades. Too many performance hounds like myself in the work to cause native to fail to be profitable.
I doubt it. The set of people for whom the < 10% overhead of virtualization is a problem will be massively outweighed by the people who would rather buy a big server to run four instances than four smaller servers. In dollar value terms, anyway.
So yeah. My point is that Virtualization will not destroy the native OS market. Not a chance in hades. Too many performance hounds like myself in the work to cause native to fail to be profitable.
That depends on the goal of the loss-leader. If it's to induce collateral purchases and thus still gain a net profit on the gross total, then the effect is not a loss to corner a market.
And it precisely is that. RMS himself makes money out of consulting on (especially) gcc. Redhat give away their work on FC in order to leverage RHEL, and also to provide a lot of beta testing (if you think Sun or MS run beta programmes for Solaris or Windows at zero cost you're insane). IBM give away their work on Linux in the hope you'll figure their boxes are a better bet to run it than Dell's, in the hope that you'll buy some consulting and in order to provide something with no license costs that their consultants can bundle (ie linux pulls consulting, or consulting pushes zero-incremental-cost). Sun give away Solaris in the (not-unreasonable) hope that you'll figure that Sun can make decent hardware to run Sun's software. Random hacker X gives away his random hack X in the hope that he'll (and yes, it's he) meet girls, get a gig with Google or appear on the frontpage of/. And so it goes.
(a) We've been talking about that on ukcrypto. It's a problem, but it's fair to say that we'll need to wait for the first case to hit court of someone claiming to have flat random data that is just flat random data.
(b) There are defenses, it's not strict liability. Again, we need to wait for cases to hit court. Remember, UK law is even more driven by precedent than than US.
You're missing the point. The S.49 notice can only be served in respect of data which has been taken under a warrant, and that warrant requires what you would call probable cause. So the flow is:
Police suspect person X of bad doings.
Police go to a judge and get a search warrant or an interception warrant. And before people start making bogus claims, anything that delivers content requires a warrant: the rows about RIPA weakening protection for traffic data aren't at issue here. That warrant requires justification.
The take is found to include encrypted data.
A S.49 notice is served to recover the encrypted material.
The `probable cause' for want of a better phrase was produced in order to obtain the original cipher text. A S.49 notice only relates to material which has been lawfully obtained.
Actually, the UK does in general allow refusal to answer on the grounds of self-incrimination. There are some very narrow cases where refusal to testify is a crime in and of itself, but they are solely related to testimony under oath in serious fraud cases, and the ECHR will take dim view of it.
The recent change to the law simply allows a judge to point out a defendent's lack of testimony in hus summing up, and permits a jury to make inferences from it. It's likely that juries always did, so this is more a technical change than it appears. The caution upon arrest has been changed (the ``if you do not say something that you later rely on in court'') to again allow inference to be taken if someone refuses to answer questions when arrested but then produces answer later in court.
I defer to no-one in my belief in the need for fair trials, but I've not seen cases advanced yet which show these changes have increased the rate of bad convictions.
That proposal was got into the consultation exercise on Part 2 by various ukcrypto people. The Home Office bill team, who were represented on ukcrypto, did explain the problems with the proposal, but I think there is still some discussion about it.
Perhaps once you've finished congratulating yourself on the amazing protective powers of the fifth amendment, you might explain to us how Joe McCarthy got away with it?
I'm not unfamiliar with US law and politics. The Bill of Rights argument isn't terribly strong: illiberal administrations have ignored it, and (for example) you were quite happy to affix the ``except for the niggers'' clause to most of it until President Johnson strapped on some balls and federalised the Alabama national guard. I don't think a legal system which requires the 101st Airborne to enforce is something to big up too much.
The US legal system offers many theoretical freedoms, but stand for communism in 1952, or atheism in 2006, and you may as well slit your own throat. In the UK our freedoms may not be written on pieces of paper, but they have a long tradition of actually being enforced. We've got a semi-fascist prime minister, but the courts are stopping him on every front. Look up the Scotsboro' Boys and call me back on how much better the US constitution is. Or Sacco and Venzetti.
Note 49(1). They can only demand keys to stuff where the cyphertext has been legally obtained. That requires a warrant. This is not about keys being demanded in isolation, this is about keys being demanded for material which is the product of search or interception warrants. Now I'm opposed to that, but opposition needs to be accurate. There is nothing whatsoever in RIPA that provides for keys to be demanded when there is not cyphertext that has been obtained lawfully.
I'm as opposed to section 3 of RIPA as the next man, but I have the benefit of having read it in detail. What is proposed is that, following a lawful search with a warrant issued by a judge, the police or judiciary can demand the keys to any encrypted material that is seized. Refusal to produce keys can be treated as a crime in its own right. Since in America your government, it would appear, doesn't bother with the ``lawful search with a warrant'' part, I think we can safely tone down the ``UK sucks'' tone.
The basic argument is that the purpose of a search warrant is defeated by encryption. Now I think that's wrong, or at least part wrong, and I think an alternative would be to make material held by the defendant which he does not choose to decrypt something that the jury can take account of, just as refusal to testify is now, under limited circumstances, something the judge can point to during summing up. And the alternative of forcing decryption isn't offered (although quite how someone would demonstrate that plain text they offered really _was_ the decryption is a whole other question).
The is bad, illiberal law, and those of us involved in campaigning against it have been in correspondance with our MPs for some years. But it's not just Britain that is tearing up its freedoms in the face of minor terrorism: the USA collectively shat its pants and ripped up a century of jurisprudence on the 12th of September. It makes far more sense for people with a desire for freedom to work together, rather than to assume that we're a bunch of proto-fascists while Bush Jr defends your constituional rights.
The problem is that a couple of hundred big time spammers are getting rich by shitting into the communal water supply!
I find the whole spam thing quite fascinating.
Firstly, I'm fascinated by where the money comes from. It's taken as axiomatic that spammers get rich because they're paid by unspecified end customers. But all the spam I've seen is for hopeless, obvious scams: are the perpetrators of such scams making so much money they can afford to pay top dollar to spam stupid people? Perhaps they can, because spam paradoxically will preferentially get through to idiots. But are the end users of the spam still making money, even after paying the spammers?
But secondly, I'm fascinated by the logic of spammers. I can see why you'd want to get your spam in front of potential marks, and people too stupid to filter are likely to be just the ticket. But why all the effort to get through filters, when you're only going to be sending mail to people who aren't stupid enough to respond anyway?
So I think spam has become an end in itself. Spammers send more spam because that's what they do, and the return on it has become secondly. The people that pay spammers pay them to send spam because it worked in the past. But they'd all probably make more money working.
Readers are referred to Freakonomics' chapter on how little money drug dealers make for further examples.
So, what's it to be, Billy Boy? Head west from Indianapolis, in which case you're excluding the eastern USA and all of Europe and Scandanavia? Or head east, in which case you're excluding all of Asia, India, Africa and the western USA? For the sake of the neat bit of alliteration, one can't help thinking that the geographic ignorance of the stereotypical USA is reinforced...
The comparison with BMW is spot on, albeit (I suspect) unintentionally: it's not that long ago that a radio (any radio) was a cost option in your BMW. The old joke about your get a steering wheel, one seat and and the option list isn't that far off the truth.
My experience is that (a) keeping clocks aligned to each other and to UTC is easier than keeping them aligned to each other without reference to UTC and (b) keeping clocks aligned to a few milliseconds is easier than keeping them aligned to a second.
What do you mean by ``good performance''? Will an arbitrarily chosen microkernel run on a 3GHz Opteron as fast as, say, SunOS 3.0 on a 15Mhz 68020 with 4MB of RAM? Clearly it will. And that was pretty fast at the time. What performance hit is acceptable in exchange for reliability is a difficult question, but in a lot of spaces a 90% hit would be acceptable, and I can think of applications where a 99% hit would be acceptable if the microkernel did indeed deliver the reliability that's claimed. After all, running at only 25% of the potential performance (and no-one's claiming that's the hit) is only 3 years on Moore's law. Vista's how many years late?
More to the point, ``because it's faster'' has been the bane of Unix. To see that in stark relief, look at the shambles of NFS being in the kernel. Rather than fix the generic problems of providing a user-space nfsd, we saw a race into the kernel for a cheap my-code-only win, plus the horror of system calls that never return. Look at the vogue for in-kernel windowing systems (Suntools, for example) although X mercifully killed that off. Repeatedly we've seen massively complex and invasive kernel subsystems produced, when a generic solution to the problems that going into the kernel allegedly solves would have benefitted everyone for longer.
You've got a problem. You decide to solve it with a kernel extension. Now you've got two problems.
The point is that in the real world, it's the delta, not the multiplier. £100 + 20% is a delta of £20. But £10 + 50% is only a delta of £5. As the baseline price falls, the hard cash value (or problem) of a 50% disparity in price becomes smaller. Is a Mac Mini plus a reasonable screen more expensive than the latest special from Dell? Yes: probably twice the price. But that's only three or four hundred quid, which in the context of computer prices of a few years ago is neither here nor there.
My reading is that this isn't double jeopardy (and isn't taking advantage of the `new and compelling') clause in the recent CJA. Note to Americans: not all court systems work like yours.
It appears that the original defence was that there was no case to answer, because a mailerserver implicitly authorises mail to be sent. No case to answer => no trial and no verdict. The Crown Prosecution Service appealed that ruling, a ruling made in a magistrate's court (ie with an essentially amateur judge) and the High Court (a real judge) said that there should be a trial.
Actually, the strata go all the way up to 15 (or is it 16?) Stratum zero are the reference clocks themselves, so a GPS receiver or similar will appear as Stratum zero to the stratum one machine it's attached to:
remote refid st t when poll reach delay offset jitter
*GPS_NMEA(0).GPS. 0 l 14 16 377 0.000 0.026 0.015
Stratum two talks to the Stratum 1 (so the best you can get without a dedicated clock attached). Stratum three talks to two (fine for diffusion to clients in general). But then you can talk to the three and be stratum four (fine for desktops on fast networks with loose timing requirements) and so on. And then you can set the hardware clock of a lightly used machine in a temperature controlled environment (we use an old Cisco which does little else) to be stratum 7 (or something) to provide an emergency clock reference if all else fails (ie we lose the GPS box, the MSF box and the Internet).
A problem with.xxx for the pr0n vendors is that it would be easy for Google to offer a far better safe search facility, by simply filtering all.xxx domains in the results. It comes and goes (fnarr fnarr), but over the years I've had the most innocuous searches return pr0n results (remember the days when META tags were still respected, and pr0n people just folded every popular search string into them?)
Oracle Collaboration Suite runs on typical Oracle platforms (we run it on Solaris/Sparc, but I think it's OK on Linux x86) and provides a pretty good Exchange work-alike. There's a connector for Outlook which provides pretty much Exchange look and feel (although it's not entirely bug free), an excellent suite of native clients (Solaris, Linux, Mac, Windows) and a Java client and a web client. It's not absolutely seamless, but only the most fervent of Exchange fiends would object. Cheap, too: a few tens of dollars per head.
Slashdot Mirror
Truth is an absolute defence to libel in England and Wales and I presume Northern Ireland (I don't know about Scotland). The problem is the reversed burden of proof: the defendent has to prove truth, albeit only to civil standards.
ian
ian
ian
ian
- (a) We've been talking about that on ukcrypto. It's a problem, but it's fair to say that we'll need to wait for the first case to hit court of someone claiming to have flat random data that is just flat random data.
-
(b) There are defenses, it's not strict liability. Again, we need to wait for cases to hit court. Remember, UK law is even more driven by precedent than than US.
ian- Police suspect person X of bad doings.
- Police go to a judge and get a search warrant or an interception warrant. And before people start making bogus claims, anything that delivers content requires a warrant: the rows about RIPA weakening protection for traffic data aren't at issue here. That warrant requires justification.
- The take is found to include encrypted data.
- A S.49 notice is served to recover the encrypted material.
The `probable cause' for want of a better phrase was produced in order to obtain the original cipher text. A S.49 notice only relates to material which has been lawfully obtained.ian
The recent change to the law simply allows a judge to point out a defendent's lack of testimony in hus summing up, and permits a jury to make inferences from it. It's likely that juries always did, so this is more a technical change than it appears. The caution upon arrest has been changed (the ``if you do not say something that you later rely on in court'') to again allow inference to be taken if someone refuses to answer questions when arrested but then produces answer later in court.
I defer to no-one in my belief in the need for fair trials, but I've not seen cases advanced yet which show these changes have increased the rate of bad convictions.
ian
ian
I'm not unfamiliar with US law and politics. The Bill of Rights argument isn't terribly strong: illiberal administrations have ignored it, and (for example) you were quite happy to affix the ``except for the niggers'' clause to most of it until President Johnson strapped on some balls and federalised the Alabama national guard. I don't think a legal system which requires the 101st Airborne to enforce is something to big up too much.
The US legal system offers many theoretical freedoms, but stand for communism in 1952, or atheism in 2006, and you may as well slit your own throat. In the UK our freedoms may not be written on pieces of paper, but they have a long tradition of actually being enforced. We've got a semi-fascist prime minister, but the courts are stopping him on every front. Look up the Scotsboro' Boys and call me back on how much better the US constitution is. Or Sacco and Venzetti.
ian
ian
ian
The basic argument is that the purpose of a search warrant is defeated by encryption. Now I think that's wrong, or at least part wrong, and I think an alternative would be to make material held by the defendant which he does not choose to decrypt something that the jury can take account of, just as refusal to testify is now, under limited circumstances, something the judge can point to during summing up. And the alternative of forcing decryption isn't offered (although quite how someone would demonstrate that plain text they offered really _was_ the decryption is a whole other question).
The is bad, illiberal law, and those of us involved in campaigning against it have been in correspondance with our MPs for some years. But it's not just Britain that is tearing up its freedoms in the face of minor terrorism: the USA collectively shat its pants and ripped up a century of jurisprudence on the 12th of September. It makes far more sense for people with a desire for freedom to work together, rather than to assume that we're a bunch of proto-fascists while Bush Jr defends your constituional rights.
ian
Firstly, I'm fascinated by where the money comes from. It's taken as axiomatic that spammers get rich because they're paid by unspecified end customers. But all the spam I've seen is for hopeless, obvious scams: are the perpetrators of such scams making so much money they can afford to pay top dollar to spam stupid people? Perhaps they can, because spam paradoxically will preferentially get through to idiots. But are the end users of the spam still making money, even after paying the spammers?
But secondly, I'm fascinated by the logic of spammers. I can see why you'd want to get your spam in front of potential marks, and people too stupid to filter are likely to be just the ticket. But why all the effort to get through filters, when you're only going to be sending mail to people who aren't stupid enough to respond anyway?
So I think spam has become an end in itself. Spammers send more spam because that's what they do, and the return on it has become secondly. The people that pay spammers pay them to send spam because it worked in the past. But they'd all probably make more money working.
Readers are referred to Freakonomics' chapter on how little money drug dealers make for further examples.
ian
ian
ian
ian
More to the point, ``because it's faster'' has been the bane of Unix. To see that in stark relief, look at the shambles of NFS being in the kernel. Rather than fix the generic problems of providing a user-space nfsd, we saw a race into the kernel for a cheap my-code-only win, plus the horror of system calls that never return. Look at the vogue for in-kernel windowing systems (Suntools, for example) although X mercifully killed that off. Repeatedly we've seen massively complex and invasive kernel subsystems produced, when a generic solution to the problems that going into the kernel allegedly solves would have benefitted everyone for longer.
You've got a problem. You decide to solve it with a kernel extension. Now you've got two problems.
ian
ian
It appears that the original defence was that there was no case to answer, because a mailerserver implicitly authorises mail to be sent. No case to answer => no trial and no verdict. The Crown Prosecution Service appealed that ruling, a ruling made in a magistrate's court (ie with an essentially amateur judge) and the High Court (a real judge) said that there should be a trial.
ian
remote refid st t when poll reach delay offset jitter .GPS. 0 l 14 16 377 0.000 0.026 0.015
*GPS_NMEA(0)
Stratum two talks to the Stratum 1 (so the best you can get without a dedicated clock attached). Stratum three talks to two (fine for diffusion to clients in general). But then you can talk to the three and be stratum four (fine for desktops on fast networks with loose timing requirements) and so on. And then you can set the hardware clock of a lightly used machine in a temperature controlled environment (we use an old Cisco which does little else) to be stratum 7 (or something) to provide an emergency clock reference if all else fails (ie we lose the GPS box, the MSF box and the Internet).
ian
ian
ian
ian