Psychotic madman in charge of a weak country contained by UN mandated military restrictions and sanctions, dimwit in charge of the world's most powerful nation.
which is the more dangerous?
Re:Too bad it's still around.. at all!
on
4l-j4z333ra 0wn3d
·
· Score: 1
This country is also the place where the automobile was developed, the lightbulb was invented, computers were born...etc... so think what you may, we have a great army too (as Saddam and his armies are learning).
The automobile was invented by Mr. Benz, a german in germany.
The modern lightbulb is generally attributed to edison, USA. (see below about electrics)
Computers: The first mechanical computer was Babbage's calculation engine (not finished), English and in London. Modern electronics draw heavily on the combined work of many many european physicists/scientists who advanced the field of electrics, Ohm, Voltaire, Kirchoff, etc.. and also very heavily on the logical algebra developed by George Boole an anglo-irish teacher who lived in Cork, Ireland.
The well-known english physicist, mathematician and theoretician Alan Turing was involved in building some of the first computers in WWII to help crack the german enigma code. He is known as the Father of modern computing.
Shockley, Brattain and Bardeen (american) developed the first transistor at Bell Labs which allowed computers to be miniaturised.
The first microchips were perhaps developed by Noyce and Kilby (the founders of Intel) (americans).
The japanese did much to bring miniaturised transistor based products to the mass-markets.
Of course, computing, science and engineering all depend heavily on mathematics. Without which none of the above would have been possible. And quite ironically, we owe a lot to the Arabs. They invented the numerical decimal notation we still use today, they perhaps introduced the very important notion of 0 to the modern world, they first factored quadratic equations and recognised these have 2 roots, and much more, eg they invented cryptanalysis too. Additionally the arab world was the centre of much learning, and preserved much knowledge acquired from the earlier greek civilisations. those of us in europe were in the dark ages.
Basically, your fantasy that the USA invented everything and is greatest and best is wrong. The USA has invented much, but so has the rest of the world. Indeed, everything mankind has done is only possible because, as Newton put it, we "stand on the shoulder of giants" of all those who came before us.
The irony is that the computers (built on ideas and work of americans, english, irish, italian, french and other men and women) on your great army's tanks are calculating shell trajectories using equations very much related to those first described and understood by arabs more than a millenium ago. The strong encryption protecting your comms - it exists because arabs first demonstrated the methodologies needed to crack cyphers.
anyway, apologies for having intruded into your blinkered little world. please return to watching fox news.
Use the rpm port of apt, eg see freshrpms.net who have a large apt repository of rpms. Course this wont help you get rpm updates for EOL RH releases, however, apt-get dist-upgrade is about the least painful way to upgrade a box. (still painful though).
The french held the germans at the Marne in World War I. The germans failed to take Paris, the Von Schlieffen plan failed, WWI then became trench warfare.
because thats his attitude? because he has no imagination? (and MS corporate is infused with his "imagination" - remember balmer's dance?). Also, because the/. story is linking just to a very small/extract/ of the interview - most likely the extract that makes billg appear the most stupid. I remember the full transcript being much longer.
Seriously, there is no reason to think its a fake. Just cause it was on the internet long before most of the slashdot kids were, and just cause Gates sounds like an arse doesnt mean its fake (maybe billg really is a "cartoon villain" - come to think of it, i cant think of anything to prove the contrary, well ok, i can for the cartoon part). Note too the fact that quite a few other people are commenting that they remember reading this article in/print/ (i read it on the web).
Its a funny read alright, but i distinctly remember that article from years ago. Quoted in quite a few places. I dont recall of any reason to consider it fake.
I'm reasonably sure this is real. At least i remember this doing the "rounds" a long long time ago and i dont remember it being discredited. I also remember the full transcription of the interview being a lot longer. Gates has evidently had a good bit of "PR for dummies" coaching since then.
Also, it credits a few people for translations, eg inaky perez gonzalez for translation into spanish - iirc he wrote the original USB stack for linux and he now works for intel (unless there's 2 of him:) ). You could ask him to verify.
Well, according to article 4 of its constitution, it's the official name of the country where you live in its native language. Now you know!
So it is and all. I always thought Eire was the name the english mistakingly used instead of Eireann, which i thought was the correct name in irish. Hmm.. (i was exempt from irish in school).
That's a matter of opinion...
It isnt, the IEDR dont. Applications are as welcome fron NI as from the south. The general criteria for a ie. domain apply to all 32 counties. (Those criteria may be rubbish, and the IEDR may have a tendency to change its rules/after/ people ask why it isnt following its own rules, but still..:) )
Having different cultures does not have to mean not getting on.
Well, i meant that even those in the north who would consider themselves more english than irish get on fine down south. And in general i feel the scots, northern irish and irish have more in common with each other culturally than they do with the english. Finally, you seem to forget that a significant part of the population of Northern Ireland are "Taigs" and/would/ consider themselves culturally closer to the South than to the english. Further, going by trends in the NI census figures, the "Taigs" look like they may one day be in the majority. (They tend to have more kids than the other half of the NI population). One reason why the Unionists were a bit anxious about the self-determination clause in the Good Friday agreement.
I think you're confusing "independance" and "wanting to be able to bully anyone they like with no fear of public opinion". Rather like today's "Loyalists": their cause was one of convienience in which only a few intellectuals had any genuine interest. Most people, then and now, would prefer them all to piss off and leave them to earning enough cash to retire on.
I was talking more of the likes of Erskine Childers, the first president of the Free State, Sir Roger Casement, yer man who married Kitty O'Shea and many many other men who were involved in the long struggle for indepedence but whose names i cant remember (was O'Connell protestant? probably not, but OTOH he was rich and in his times it was nearly impossible to be rich and catholic).
Your view on the modern paramilitaries i'd agree on. They're a curse.
Yes, and zero influence on the EU, just like the people in Northern Ireland. Big deal! Ireland votes the "wrong way"? Have another vote. Keep on having votes until they give in and vote the "right way". Still, as long as the hand-outs keep coming who gives a toss about independance now, eh?
Well, the president of the European Parliament is an Irish man (like yourself:) ). But who has influence in the EU anyway? That's a general problem with the EU at the moment, very little accountability and even lower perception of the accountability there is. The Nice referendum, well because of the initial "No" we were able to get some concessions, chiefly on neutrality. Once that contentious issue was dealt with the second vote was much more likely to be "yes". Handouts: yes, we took them, but Ireland was very under developed. I'll note that we will soon become net contributors to the EU soon if you exclude CAP (which benefits the big 2 quite a lot too). As for independence, we're still a sovereign state, however we/agreed/ to enter the EU and to the processes, overseeing institutions and international law that go with it, just as has the UK i'll note.
Anyway, on the national level, 25% of Dail Eireann beats 2% of English parliament anytime. And the unionists'd be welcomed the same as any other to it. Didnt Pat Kenny have Ian Paisley on his Friday night talk show on RTE1 once upon a time?:) (his son has been on TV talk shows down here too). Personally, i rate the leader of the PUP (the guy with the moustache) very very highly.
stop pretending that crack sellers are politically motivated.
Actually, one of the best arguments i've heard for the 6 counties / Northern Ireland ceding from the UK and joining "Ireland" was that immediately all reasons for republican paramilitarism would cease, and presumably loyalist too, if they wanted this. This argument actually came from a sitting Unionist MP, who was promptly kicked out of his party:)
Which works on a similar principle, but by checksum of reported spam mails rather than by volume. The more times a checksum is reported (and who by) the more likely it is to be spam - beyond a certain level checksums will be considered spam. Razor catches a good amount of spam for me.
The best way to fight spam imo is to employ a mix of anti-UCE tools, ie DNSBls to block connections + rbl-milter to 'tag' mail based on a very wide range of DNSBls + Spam checksum clearing house (eg Razor) + a content filter to rate mails according to content and whether they have headers inserted by aforementioned anti-uce tools.
The ie. domain does not make distinctions between the South and North.
As for culture and NI not having anything in common with the South, that really depends on your perspective. There are lots of Northern Irish down here, many of whom'd be of the more orange persuasion, and they get on fine. but i'd still say they're closer in culture to the South than to those in England. (same applies to Scotland). There is one cultural aspect that Northern Ireland and Scotland share which the south does not, and which i'm glad of, and that is the secterian bigotry that sadly still infects the two.
There's a reason that orange is one of the colours of the flag of the Republic: Irish independence was originally a middle-class protestant cause and many of the well-known names involved in gaining that independence were protestant. Orange for the Orangemen in the North, Green for the south and white for peace between them.
And imo, the unionists in NI would ultimately be better off sitting in Dail Eireinn than in English Parliament. They'd have a lot more seats and influence for a start - ~25%. However, strange sentimentalities for a dutch prince who lived 400 years ago seem to have clouded their thinking.
And btw, I have a protestant dutch father, and i've lived in scotland.:)
1. 3.5 million people in the south, ~2.5M in the north. Presuming 60% of the NI population votes for some unionist party, then 1.5M/6M = 25%. Lot better than the measly representation NI has in English parliament. (course, on occassion the unionists will hold balance of power and the "orange card" will be played).
Re:The Ultimate Car For Tinkerers
on
Cars for Tinkerers?
·
· Score: 2, Interesting
eek. 230HP and 460Kg. 0.5 hp/kg. owww..
Another neat car is the Westfield XTR 180BHP and 440Kg for 0.41hp/kg. The Le Man prototype looks are very nice. Though it probably isnt as practical as the Caterham. (which isnt very practical in itself).
Course, none of these cars have anything on motorbikes, even a piddly little 250cc can beat the above cars for power/weight with a small amount of tuning. And the 1 litre bikes blow them away completely, eg Yamaha R1 150hp/177kg = 0.85 hp/kg! 0-120mph+ in 10s or less.:)
Most of the features that AGP adds to a PCI slot aren't useful for other devices.
Such as speed and an IO-MMU?
Very little except a video card can use the GART
A device doesnt use the GART, the OS programmes the GART. The GART is an IO-MMU, same as the more sophisticated types of IO-MMUs that can be found on many 64bit systems, eg Sparc64, PPC64 and Alpha. Used to, eg, work around limitations of devices/buses without having to resort to performance sapping tricks (eg copying data around in system memory). eg 32 bit address capable bus and 64 bit system address space, or hardware that cant do scatter gather, or to avoid use of PCI's DAC (Dual Address Cycle -> 64 bit addressing, but takes 2 bus cycles).
So GART, or at least the concept of being able to arbitrarily map bus to physical addresses is not at all limited to video cards. Indeed, IUI, the GART is available to all AGP and PCI devices (presuming they all sit behind the same northbridge).
Main reason, AIUI, that you dont see multiple AGP slots is that AGP is limited to one device per bus. Which would means you'd need dedicated AGP bridge and board traces for each slot - extra expense. And no one makes non-video card AGP devices, so who's going to bother? Though it could be useful for dual-head. Never heard of a board with dual-AGP though.
HT doesn't even have a possibility to slow things down
Yes it does. The 2 'threads' of the CPU share the same bus and cache, in some scenarios pressure on the cache is such that it runs slower. I think anandtech had a good article on this a while ago.
My claim was that this article, despite it's title (HT Speeds Linux), wasn't really showing any proof that linux itself was going faster.
Read the paper. While lmbench results werent faster (and indeed lmbench by design tests very specific kernel paths) look at the other benchmarks - they show huge speedups, a lot of which will be due to parallelism in the kernel file and network paths.
Another thing is that the NT kernel might be re-entrant, but the rules of IRQL still hold... there are many functions that can't be called at elevated IRQLevels. So, it's not as complicating the kernel as you might think.
Well, i dont really know much about the NT kernel. So originally your point was it was so reentrant, depending on the level. Now you've qualified that to say that many functions cant be called if irq level is elevated - hence nullifying much of that supposed reentrancy.:)
kernel programming, and kernel optimizations are very much a black art, rocket sciene, or voodoo
It isnt really. Its quite well understood. Its just there's lots to it - which means most of us will never be able to understand all or even a part of modern highly-featured kernels. But there's nothing magic about it.
I still find the numbers to be skewed to the profit of the publisher.
it means a piece of hardware raising an interupt will launch a driver's ISR, hardware wise, at that point, any other hardware raising an interupt at a lower IRQ level will not be serviced. BUT, any higher IRQLed hardware can take over of that 'thread' that is servicing the first interupt. There are *no* exceptions to this rule. In effect, your ISR is fully interuptible.
Linux doesnt do IRQ priority (beyond any irq priority done in the interrupt controller). Priorities are horrible do correctly - priority inversion for example. Also, not all the platforms linux runs on might support irq priorities. So Linux tries to use a more abstract and portable model, which incidentally allows even better handling than the priority based system you describe (while retaining the advantage of "simplicity").
In linux, a lock is taken relating to the IRQ number, the interrupt is acknowledged, and the driver 'handler' is run (usually with interrupts/enabled/). Hence in linux, running of a driver 'ISR' does not prevent other interrupts from being serviced (eg by another CPU if SMP). No priorities - the only rule is that the/same/ interrupt can not run concurrently. Also, linux interrupt handling is done in 2 halves (on the general Unix fashion) -> Top half and bottom half. The top half is run with the interrupt disabled - its job is to service the interrupt as fast as possible doing the minimum amount of work needed. The bottom half is run after the top-half has finished, but with the interrupt enabled. (NB: tasklets are replacing bottom-halves, but the general idea is the same. The ISR can schedule 'tasklets' to do work, when it finishes, the kernel will run the tasklets.)
NB: a driver ISR can set SA_INTERRUPT which will mean the general kernel interrupt mechanism will call it with all interrupts disabled. (not reccomended).
re-entrant: interrupts in linux can call certain other general kernel functions, but must be very careful. and there are certain things they must not do (and these restrictions impact very heavily on what functions they can call). However, this isnt a bad thing. A kernel that is re-entrent even from within interrupt handling is possibly one that is trying to overreach itself in its design goals. Purpose of an ISR is to do whats needed, do it fast, and get out of the way. Real work should be done outside of interrupt context - otherwise the complexities are horrendous. Ie linux not allowing much reentrancy from interrupt context is quite possibly a very good design decision.
spin locks: as someone points out, they are a well known and long-standing concept. However, fascinating as the concept and use of them may be - they do *not* increase performance in themselves, in fact they have an overhead. And while their use may allow a kernel to perform better by allowing concurrency, they do complicate the kernel. And their use is a tradeoff between simplicity, lack of reentrancy, efficient at serial loads/inefficient at parallel loads Vs complexity, reentrancy and inefficient at serial loads/efficiency of parrallel loads. (but that efficiency is not scalar because of lock overhead - even more so because complexity overhead makes it difficult for programmers to get very grained locking to be efficient and reliable).
That said, Linux runs on systems with far more CPUs than NT does (iirc linux: SGI Origin 64node O2K with 128CPUs Vs NT on 32 CPU Unisys machines, but unisys recc'd partitioning the system, so in production use you'd run maybe 8 to 16 NT partitions with 2 to 4 CPUs each). And SGI and others have done lots of work to refine locking (and other many CPU SMP issues).
Benchmarks: i asked first:) Its been a long while since those benchmarks came out that showed where NT was stronger than linux at SMP (very carefully targeted benchmark too). The scaling issues that threw up/benefited/ linux as the kernel hackers went and solved them.
So show me a recent benchmark that backs up your claim.
HT a win in the general case: not on certain workloads.:) (but in general, probably).
Look at what the LMBench benchmark is doing - in most cases it tests fairly specific OS paths. Eg a copy of data from userspace to kernel. Or a fork + exec. While these fairly shorts paths may not be "multi-threaded" the kernel itself still is.
the NT kernel is fully pre-emptible, fully-interuptible, and re-entrant by design
Care to explain what these mean?
I doubt very much its 100% preemptible and interruptable - eg the initial OS interrupt vector must not be interrupted, drivers have often have a requirement to turn off interrupts (usually of the one they handle, but sometimes all interrupts). And even if by some magic NT is "fully interruptable and preemtible" - that does not mean it has much of a gain. You must lock data before you can access it, the kernel might preempt one task with another, but that second task might find that data it needs is locked and so has to spin or sleep.
Highly threaded and preemtible kernel's also suffer from complexity (affects maintainability and stability) and this complexity and locking overhead can quite easily lead to/worse/ performance for many workloads.
The linux kernel is not
False. Linux acquired fine grained locking in 2.2, and has moved ever away from the big kernel lock since then. 2.5 is almost bkl free.
But it also has the advantage that it makes *much* better use of SMP. (than linux you mean presumably.)
Perhaps you may want to run some benchmarks and then come back and revisit that claim. Eg, which OS holds the SPECWeb record? Linux process/thread creation is an order of magnitude better than that of NT. etc..
You've forgotten the biggest one by far - Intel out in Leixlip (just outside County Dublin). 2 Fabs there already and they're working on a 3rd.
As for tax "less is better", are you mad??:) Ok, its not scandivian level high, but i wouldnt say we pay little tax. Corporation tax OTOH is very low - esp for foreign companies. However, that's a concession Ireland got from the EU, and its due to finish in 2010 iirc. We'll have to come inline with mainstream EU corporate tax levels after that.
I did state there was no publically known attack against the full rounds of MD5. However, that was 7 years ago, it is not safe to assume that no one else continued on with this work.
that does not mean that finding them is computationally easy or even possible.
Actually, there are well known issues with MD5 that make it susceptible to collission searches, see:
H. Dobbertin, "The Status of MD5 After a Recent Attack", RSA Labs' CryptoBytes, Vol. 2 No. 2, Summer 1996.
http://www.rsa.com/rsalabs/pubs/cryptobytes.html dont think that URL works anymore. This one does, in which Robshaw gives an overview of the problems:
ftp://ftp.rsa.com/pub/pdfs/bulletn4.pdf
Basically, it has been demonstrated by Dobertin in 1996 that data with a colliding hash can be found with 10 odd hours of processing from a (by now very low powered) PC. Admittedly only for the compression round of MD5, not for the full set of rounds specified by MD5, however it is feared that existing techniques (ie those used to break MD4) can be applied to MD5. (indeed this is what Dobertin demonstrated). TTBOMK there is no known collision attack against the full MD5 algorithm. (least not public knowledge anyway:) )
So your assertion is incorrect with respect to MD5.
SHA-1 is currently considered to be safe from hash collission attacks. However, that is not really relevant as the FBI specifically are using CRC-32 and MD5.
However, presuming that the question is not one of the FBI having deliberately modified the images, it does not/matter/ that MD5 is on shaky ground wrt to strength against collission attacks. The use of MD5 here is to verify that the copies are the same as the original images and that there werent any errors introduced during copying. For this purpose MD5 is fine.
The recorder will be/integrated/ into the ECU, or rather the functions of the ECU will expand to include data-logging. Some ECUs already do limited logging of engine faults for diagnostic purposes, race ECU systems already do extensive data logging of engine parameters.
As there will be no seperate recorder, you will not be able to remove it. If you remove the ECU - car doesnt work - its an Engine/Control/ Unit.
So, as the man says... it'd definitely be a good idea if we had the choice of Open (ie accessible, hackable) ECU systems. Hopefully in the future ECUs will move towards ever more standardised general-purpose computers, and hence the problem hopefully will just be one of software. Hopefully..
Directive still stands - will still be implemented
on
Euro DMCA Fails
·
· Score: 2
Errmmm.... has noone cottoned on to the fact that the directive *still* stands? And as such all EU countries are still obligated to pass into local law legislation that implements the Directive. While a directive may contain a deadline for local implementation, missing this deadline does not remove the obligation to implement it.
Slashdot Mirror
Psychotic madman in charge of a weak country contained by UN mandated military restrictions and sanctions, dimwit in charge of the world's most powerful nation.
which is the more dangerous?
This country is also the place where the automobile was developed, the lightbulb was invented, computers were born...etc... so think what you may, we have a great army too (as Saddam and his armies are learning).
The automobile was invented by Mr. Benz, a german in germany.
The modern lightbulb is generally attributed to edison, USA. (see below about electrics)
Computers: The first mechanical computer was Babbage's calculation engine (not finished), English and in London. Modern electronics draw heavily on the combined work of many many european physicists/scientists who advanced the field of electrics, Ohm, Voltaire, Kirchoff, etc.. and also very heavily on the logical algebra developed by George Boole an anglo-irish teacher who lived in Cork, Ireland.
The well-known english physicist, mathematician and theoretician Alan Turing was involved in building some of the first computers in WWII to help crack the german enigma code. He is known as the Father of modern computing.
Shockley, Brattain and Bardeen (american) developed the first transistor at Bell Labs which allowed computers to be miniaturised.
The first microchips were perhaps developed by Noyce and Kilby (the founders of Intel) (americans).
The japanese did much to bring miniaturised transistor based products to the mass-markets.
Of course, computing, science and engineering all depend heavily on mathematics. Without which none of the above would have been possible. And quite ironically, we owe a lot to the Arabs. They invented the numerical decimal notation we still use today, they perhaps introduced the very important notion of 0 to the modern world, they first factored quadratic equations and recognised these have 2 roots, and much more, eg they invented cryptanalysis too. Additionally the arab world was the centre of much learning, and preserved much knowledge acquired from the earlier greek civilisations. those of us in europe were in the dark ages.
Basically, your fantasy that the USA invented everything and is greatest and best is wrong. The USA has invented much, but so has the rest of the world. Indeed, everything mankind has done is only possible because, as Newton put it, we "stand on the shoulder of giants" of all those who came before us.
The irony is that the computers (built on ideas and work of americans, english, irish, italian, french and other men and women) on your great army's tanks are calculating shell trajectories using equations very much related to those first described and understood by arabs more than a millenium ago. The strong encryption protecting your comms - it exists because arabs first demonstrated the methodologies needed to crack cyphers.
anyway, apologies for having intruded into your blinkered little world. please return to watching fox news.
Use the rpm port of apt, eg see freshrpms.net who have a large apt repository of rpms. Course this wont help you get rpm updates for EOL RH releases, however, apt-get dist-upgrade is about the least painful way to upgrade a box. (still painful though).
Your sig is incorrect.
The french held the germans at the Marne in World War I. The germans failed to take Paris, the Von Schlieffen plan failed, WWI then became trench warfare.
because thats his attitude? because he has no imagination? (and MS corporate is infused with his "imagination" - remember balmer's dance?). Also, because the /. story is linking just to a very small /extract/ of the interview - most likely the extract that makes billg appear the most stupid. I remember the full transcript being much longer.
/print/ (i read it on the web).
Seriously, there is no reason to think its a fake. Just cause it was on the internet long before most of the slashdot kids were, and just cause Gates sounds like an arse doesnt mean its fake (maybe billg really is a "cartoon villain" - come to think of it, i cant think of anything to prove the contrary, well ok, i can for the cartoon part). Note too the fact that quite a few other people are commenting that they remember reading this article in
Its a funny read alright, but i distinctly remember that article from years ago. Quoted in quite a few places. I dont recall of any reason to consider it fake.
I'm reasonably sure this is real. At least i remember this doing the "rounds" a long long time ago and i dont remember it being discredited. I also remember the full transcription of the interview being a lot longer. Gates has evidently had a good bit of "PR for dummies" coaching since then.
:) ). You could ask him to verify.
Also, it credits a few people for translations, eg inaky perez gonzalez for translation into spanish - iirc he wrote the original USB stack for linux and he now works for intel (unless there's 2 of him
wowser kev... maybe you should stop watching Fox news so much? doesnt seem to be having a good effect on your blood pressure. :)
Well, according to article 4 of its constitution, it's the official name of the country where you live in its native language. Now you know!
/after/ people ask why it isnt following its own rules, but still.. :) )
/would/ consider themselves culturally closer to the South than to the english. Further, going by trends in the NI census figures, the "Taigs" look like they may one day be in the majority. (They tend to have more kids than the other half of the NI population). One reason why the Unionists were a bit anxious about the self-determination clause in the Good Friday agreement.
:) ). But who has influence in the EU anyway? That's a general problem with the EU at the moment, very little accountability and even lower perception of the accountability there is. The Nice referendum, well because of the initial "No" we were able to get some concessions, chiefly on neutrality. Once that contentious issue was dealt with the second vote was much more likely to be "yes". Handouts: yes, we took them, but Ireland was very under developed. I'll note that we will soon become net contributors to the EU soon if you exclude CAP (which benefits the big 2 quite a lot too). As for independence, we're still a sovereign state, however we /agreed/ to enter the EU and to the processes, overseeing institutions and international law that go with it, just as has the UK i'll note.
:) (his son has been on TV talk shows down here too). Personally, i rate the leader of the PUP (the guy with the moustache) very very highly.
:)
So it is and all. I always thought Eire was the name the english mistakingly used instead of Eireann, which i thought was the correct name in irish. Hmm.. (i was exempt from irish in school).
That's a matter of opinion...
It isnt, the IEDR dont. Applications are as welcome fron NI as from the south. The general criteria for a ie. domain apply to all 32 counties. (Those criteria may be rubbish, and the IEDR may have a tendency to change its rules
Having different cultures does not have to mean not getting on.
Well, i meant that even those in the north who would consider themselves more english than irish get on fine down south. And in general i feel the scots, northern irish and irish have more in common with each other culturally than they do with the english. Finally, you seem to forget that a significant part of the population of Northern Ireland are "Taigs" and
I think you're confusing "independance" and "wanting to be able to bully anyone they like with no fear of public opinion". Rather like today's "Loyalists": their cause was one of convienience in which only a few intellectuals had any genuine interest. Most people, then and now, would prefer them all to piss off and leave them to earning enough cash to retire on.
I was talking more of the likes of Erskine Childers, the first president of the Free State, Sir Roger Casement, yer man who married Kitty O'Shea and many many other men who were involved in the long struggle for indepedence but whose names i cant remember (was O'Connell protestant? probably not, but OTOH he was rich and in his times it was nearly impossible to be rich and catholic).
Your view on the modern paramilitaries i'd agree on. They're a curse.
Yes, and zero influence on the EU, just like the people in Northern Ireland. Big deal! Ireland votes the "wrong way"? Have another vote. Keep on having votes until they give in and vote the "right way". Still, as long as the hand-outs keep coming who gives a toss about independance now, eh?
Well, the president of the European Parliament is an Irish man (like yourself
Anyway, on the national level, 25% of Dail Eireann beats 2% of English parliament anytime. And the unionists'd be welcomed the same as any other to it. Didnt Pat Kenny have Ian Paisley on his Friday night talk show on RTE1 once upon a time?
stop pretending that crack sellers are politically motivated.
Actually, one of the best arguments i've heard for the 6 counties / Northern Ireland ceding from the UK and joining "Ireland" was that immediately all reasons for republican paramilitarism would cease, and presumably loyalist too, if they wanted this. This argument actually came from a sitting Unionist MP, who was promptly kicked out of his party
An excellent idea, and such a system already exists, see the Distributed Checksum Clearing house:
http://www.rhyolite.com/anti-spam/dcc/
There's also Vipul's Razor:
http://razor.sourceforge.net/
Which works on a similar principle, but by checksum of reported spam mails rather than by volume. The more times a checksum is reported (and who by) the more likely it is to be spam - beyond a certain level checksums will be considered spam. Razor catches a good amount of spam for me.
The best way to fight spam imo is to employ a mix of anti-UCE tools, ie DNSBls to block connections + rbl-milter to 'tag' mail based on a very wide range of DNSBls + Spam checksum clearing house (eg Razor) + a content filter to rate mails according to content and whether they have headers inserted by aforementioned anti-uce tools.
Eire? I dont know of a country called Eire.
:)
The ie. domain does not make distinctions between the South and North.
As for culture and NI not having anything in common with the South, that really depends on your perspective. There are lots of Northern Irish down here, many of whom'd be of the more orange persuasion, and they get on fine. but i'd still say they're closer in culture to the South than to those in England. (same applies to Scotland). There is one cultural aspect that Northern Ireland and Scotland share which the south does not, and which i'm glad of, and that is the secterian bigotry that sadly still infects the two.
There's a reason that orange is one of the colours of the flag of the Republic: Irish independence was originally a middle-class protestant cause and many of the well-known names involved in gaining that independence were protestant. Orange for the Orangemen in the North, Green for the south and white for peace between them.
And imo, the unionists in NI would ultimately be better off sitting in Dail Eireinn than in English Parliament. They'd have a lot more seats and influence for a start - ~25%. However, strange sentimentalities for a dutch prince who lived 400 years ago seem to have clouded their thinking.
And btw, I have a protestant dutch father, and i've lived in scotland.
1. 3.5 million people in the south, ~2.5M in the north. Presuming 60% of the NI population votes for some unionist party, then 1.5M/6M = 25%. Lot better than the measly representation NI has in English parliament. (course, on occassion the unionists will hold balance of power and the "orange card" will be played).
It does: ie..
How do you 'burn-in' /black/ bars?
eek. 230HP and 460Kg. 0.5 hp/kg. owww..
:)
Another neat car is the Westfield XTR 180BHP and 440Kg for 0.41hp/kg. The Le Man prototype looks are very nice. Though it probably isnt as practical as the Caterham. (which isnt very practical in itself).
Course, none of these cars have anything on motorbikes, even a piddly little 250cc can beat the above cars for power/weight with a small amount of tuning. And the 1 litre bikes blow them away completely, eg Yamaha R1 150hp/177kg = 0.85 hp/kg! 0-120mph+ in 10s or less.
Most of the features that AGP adds to a PCI slot aren't useful for other devices.
Such as speed and an IO-MMU?
Very little except a video card can use the GART
A device doesnt use the GART, the OS programmes the GART. The GART is an IO-MMU, same as the more sophisticated types of IO-MMUs that can be found on many 64bit systems, eg Sparc64, PPC64 and Alpha. Used to, eg, work around limitations of devices/buses without having to resort to performance sapping tricks (eg copying data around in system memory). eg 32 bit address capable bus and 64 bit system address space, or hardware that cant do scatter gather, or to avoid use of PCI's DAC (Dual Address Cycle -> 64 bit addressing, but takes 2 bus cycles).
So GART, or at least the concept of being able to arbitrarily map bus to physical addresses is not at all limited to video cards. Indeed, IUI, the GART is available to all AGP and PCI devices (presuming they all sit behind the same northbridge).
Main reason, AIUI, that you dont see multiple AGP slots is that AGP is limited to one device per bus. Which would means you'd need dedicated AGP bridge and board traces for each slot - extra expense. And no one makes non-video card AGP devices, so who's going to bother? Though it could be useful for dual-head. Never heard of a board with dual-AGP though.
HT doesn't even have a possibility to slow things down
Yes it does. The 2 'threads' of the CPU share the same bus and cache, in some scenarios pressure on the cache is such that it runs slower. I think anandtech had a good article on this a while ago.
My claim was that this article, despite it's title (HT Speeds Linux), wasn't really showing any proof that linux itself was going faster.
:)
:)
Read the paper. While lmbench results werent faster (and indeed lmbench by design tests very specific kernel paths) look at the other benchmarks - they show huge speedups, a lot of which will be due to parallelism in the kernel file and network paths.
Another thing is that the NT kernel might be re-entrant, but the rules of IRQL still hold... there are many functions that can't be called at elevated IRQLevels. So, it's not as complicating the kernel as you might think.
Well, i dont really know much about the NT kernel. So originally your point was it was so reentrant, depending on the level. Now you've qualified that to say that many functions cant be called if irq level is elevated - hence nullifying much of that supposed reentrancy.
kernel programming, and kernel optimizations are very much a black art, rocket sciene, or voodoo
It isnt really. Its quite well understood. Its just there's lots to it - which means most of us will never be able to understand all or even a part of modern highly-featured kernels. But there's nothing magic about it.
I still find the numbers to be skewed to the profit of the publisher.
So come up with some benchmarks.
it means a piece of hardware raising an interupt will launch a driver's ISR, hardware wise, at that point, any other hardware raising an interupt at a lower IRQ level will not be serviced. BUT, any higher IRQLed hardware can take over of that 'thread' that is servicing the first interupt. There are *no* exceptions to this rule. In effect, your ISR is fully interuptible.
/enabled/). Hence in linux, running of a driver 'ISR' does not prevent other interrupts from being serviced (eg by another CPU if SMP). No priorities - the only rule is that the /same/ interrupt can not run concurrently. Also, linux interrupt handling is done in 2 halves (on the general Unix fashion) -> Top half and bottom half. The top half is run with the interrupt disabled - its job is to service the interrupt as fast as possible doing the minimum amount of work needed. The bottom half is run after the top-half has finished, but with the interrupt enabled. (NB: tasklets are replacing bottom-halves, but the general idea is the same. The ISR can schedule 'tasklets' to do work, when it finishes, the kernel will run the tasklets.)
:) Its been a long while since those benchmarks came out that showed where NT was stronger than linux at SMP (very carefully targeted benchmark too). The scaling issues that threw up /benefited/ linux as the kernel hackers went and solved them.
:) (but in general, probably).
Linux doesnt do IRQ priority (beyond any irq priority done in the interrupt controller). Priorities are horrible do correctly - priority inversion for example. Also, not all the platforms linux runs on might support irq priorities. So Linux tries to use a more abstract and portable model, which incidentally allows even better handling than the priority based system you describe (while retaining the advantage of "simplicity").
In linux, a lock is taken relating to the IRQ number, the interrupt is acknowledged, and the driver 'handler' is run (usually with interrupts
NB: a driver ISR can set SA_INTERRUPT which will mean the general kernel interrupt mechanism will call it with all interrupts disabled. (not reccomended).
re-entrant: interrupts in linux can call certain other general kernel functions, but must be very careful. and there are certain things they must not do (and these restrictions impact very heavily on what functions they can call). However, this isnt a bad thing. A kernel that is re-entrent even from within interrupt handling is possibly one that is trying to overreach itself in its design goals. Purpose of an ISR is to do whats needed, do it fast, and get out of the way. Real work should be done outside of interrupt context - otherwise the complexities are horrendous. Ie linux not allowing much reentrancy from interrupt context is quite possibly a very good design decision.
spin locks: as someone points out, they are a well known and long-standing concept. However, fascinating as the concept and use of them may be - they do *not* increase performance in themselves, in fact they have an overhead. And while their use may allow a kernel to perform better by allowing concurrency, they do complicate the kernel. And their use is a tradeoff between simplicity, lack of reentrancy, efficient at serial loads/inefficient at parallel loads Vs complexity, reentrancy and inefficient at serial loads/efficiency of parrallel loads. (but that efficiency is not scalar because of lock overhead - even more so because complexity overhead makes it difficult for programmers to get very grained locking to be efficient and reliable).
That said, Linux runs on systems with far more CPUs than NT does (iirc linux: SGI Origin 64node O2K with 128CPUs Vs NT on 32 CPU Unisys machines, but unisys recc'd partitioning the system, so in production use you'd run maybe 8 to 16 NT partitions with 2 to 4 CPUs each). And SGI and others have done lots of work to refine locking (and other many CPU SMP issues).
Benchmarks: i asked first
So show me a recent benchmark that backs up your claim.
HT a win in the general case: not on certain workloads.
You're assuming the kernel is one thread =) Of course it wouldn't boost up if all kernel calls were like this:
/worse/ performance for many workloads.
void myKernelFunc( long param ) { return param * 2 - 23; }
urrgg... what balderdash.
Look at what the LMBench benchmark is doing - in most cases it tests fairly specific OS paths. Eg a copy of data from userspace to kernel. Or a fork + exec. While these fairly shorts paths may not be "multi-threaded" the kernel itself still is.
the NT kernel is fully pre-emptible, fully-interuptible, and re-entrant by design
Care to explain what these mean?
I doubt very much its 100% preemptible and interruptable - eg the initial OS interrupt vector must not be interrupted, drivers have often have a requirement to turn off interrupts (usually of the one they handle, but sometimes all interrupts). And even if by some magic NT is "fully interruptable and preemtible" - that does not mean it has much of a gain. You must lock data before you can access it, the kernel might preempt one task with another, but that second task might find that data it needs is locked and so has to spin or sleep.
Highly threaded and preemtible kernel's also suffer from complexity (affects maintainability and stability) and this complexity and locking overhead can quite easily lead to
The linux kernel is not
False. Linux acquired fine grained locking in 2.2, and has moved ever away from the big kernel lock since then. 2.5 is almost bkl free.
But it also has the advantage that it makes *much* better use of SMP. (than linux you mean presumably.)
Perhaps you may want to run some benchmarks and then come back and revisit that claim. Eg, which OS holds the SPECWeb record? Linux process/thread creation is an order of magnitude better than that of NT. etc..
You've forgotten the biggest one by far - Intel out in Leixlip (just outside County Dublin). 2 Fabs there already and they're working on a 3rd.
:) Ok, its not scandivian level high, but i wouldnt say we pay little tax. Corporation tax OTOH is very low - esp for foreign companies. However, that's a concession Ireland got from the EU, and its due to finish in 2010 iirc. We'll have to come inline with mainstream EU corporate tax levels after that.
As for tax "less is better", are you mad??
I did state there was no publically known attack against the full rounds of MD5. However, that was 7 years ago, it is not safe to assume that no one else continued on with this work.
that does not mean that finding them is computationally easy or even possible.
:) )
/matter/ that MD5 is on shaky ground wrt to strength against collission attacks. The use of MD5 here is to verify that the copies are the same as the original images and that there werent any errors introduced during copying. For this purpose MD5 is fine.
Actually, there are well known issues with MD5 that make it susceptible to collission searches, see:
H. Dobbertin, "The Status of MD5 After a Recent Attack", RSA Labs' CryptoBytes, Vol. 2 No. 2, Summer 1996.
http://www.rsa.com/rsalabs/pubs/cryptobytes.html
dont think that URL works anymore. This one does, in which Robshaw gives an overview of the problems:
ftp://ftp.rsa.com/pub/pdfs/bulletn4.pdf
Basically, it has been demonstrated by Dobertin in 1996 that data with a colliding hash can be found with 10 odd hours of processing from a (by now very low powered) PC. Admittedly only for the compression round of MD5, not for the full set of rounds specified by MD5, however it is feared that existing techniques (ie those used to break MD4) can be applied to MD5. (indeed this is what Dobertin demonstrated). TTBOMK there is no known collision attack against the full MD5 algorithm. (least not public knowledge anyway
So your assertion is incorrect with respect to MD5.
SHA-1 is currently considered to be safe from hash collission attacks. However, that is not really relevant as the FBI specifically are using CRC-32 and MD5.
However, presuming that the question is not one of the FBI having deliberately modified the images, it does not
You havn't understood the post you replied to.
/integrated/ into the ECU, or rather the functions of the ECU will expand to include data-logging. Some ECUs already do limited logging of engine faults for diagnostic purposes, race ECU systems already do extensive data logging of engine parameters.
/Control/ Unit.
The recorder will be
As there will be no seperate recorder, you will not be able to remove it. If you remove the ECU - car doesnt work - its an Engine
So, as the man says... it'd definitely be a good idea if we had the choice of Open (ie accessible, hackable) ECU systems. Hopefully in the future ECUs will move towards ever more standardised general-purpose computers, and hence the problem hopefully will just be one of software. Hopefully..
Errmmm.... has noone cottoned on to the fact that the directive *still* stands? And as such all EU countries are still obligated to pass into local law legislation that implements the Directive. While a directive may contain a deadline for local implementation, missing this deadline does not remove the obligation to implement it.
"although the routing servers still get pounded"
WTF? What is a "routing server"? How does it get "pounded" because people use P2P clients.
??