I think I understand the thinking behind having patents - as a way to prevent others from profiting from a patent owner's hard work on researching and implementing a solution to some problem merely by seeing their solution and copying it.
HOWEVER - it seems that many new inventions come about due to the combination of current scientific knowledge, current technology and current problems. This often leads to the same (or very similar) inventions being independantly made by several people across the world without them necessarily 'stealing' the idea from one another, and yet the patent system as currently implemented appears to try and deny all except one of the discovers from using their invention.
Now on the one hand you could argue that instead of bothering with all this research and risking making an unusable independant discovery you should put all that effort in to searching for and licensing other's patents (which would ultimately lead to stagnation - see Asimov's foundation series)
More likely the solution is either not published when you start the research, or (in the case of programming, much of which is researching and 'inventing' ways for a computer to do some task) you do not think of your solution as an 'invention' that has a patent but rather just 'one obvious (to me) algorithm / UI to crack this sub-problem in creating something to meet my current project's requirements'
re: They do not, and never been able to, provide any verification of who is on either end.
They do provide a _little_ verification - so long as you permanently remember the certificate it verifies that whoever is at the other end of the connection on subsequent sessions is the same as whoever is at the other end now (so long as they've kept their private cert suitably secure)
aside: is there _any_ way of ever verifying 'who' someone / something is beyond 'you are the same person that I saw that other time' or 'you are the same person that someone intermediary claims they saw in the past'-> which if you follow it through leads to a better understanding of what a person's identity is and is not, and what you can/can't achieve when requiring people to present 'identification' for security purposes - and a misunderstanding of which leads to a lot of silly security 'theatre'
Exactly - and this highlights the problems we currently face with _not_ having some universally (or at least nationally) accepted form of identity proof.
Ideally we need to have some form of secret ID proving information (the only one I can think of is to use RSA style certificates with proof of ID involving possession of a physical device that can authenticate using its private key). The nearest that most people have at the moment and are understood to be private are :
- PIN numbers for credit/debit/cashpoint card (problem here being that you have to type in the 4 digit PIN in plain-text to various untrusted devices, so it can easily be stolen : though with chip-and-pin I don't know if the chip is doing some fancy public key handshake thing or not)
- Demonstrating that you can write your signature in a reasonably natural looking manner
- Possession of a (relatively) difficult document to forge / copy such as passport or driving license (the obvious problem - they are not really _that_ difficult to copy, and secondly there is no trusted relationship between your identity as verified by possession of the passport and other day to day activities such as trying to arrange to transfer money from your bank account to a shopkeeper to pay for goods at the checkout)
Unfortunately many places erroneously trust various other pieces of non-private data to authenticate people such as :
- Use of national-insurance number (is that equivilent to american SSN?)
- bank account sort code and account number
- knowledge of name, address and phone number
- a facsimile of a signature
- your company employee numbeer, salary, favourite colour and pet's name. Information such as this is not 'secret' and so should not be treated as private data sufficient for authentication (and subsequent authorization) from a security perspective, despite many people thinking of it as 'private' in social terms.
So in conclusion I sympathise with Clarkson - the information he published should not have been sufficient for identity theft to occur, the problem is with those that accepted the information as valid proof of having authority to transfer money out of his bank account.
A much more noticable effect is the horizontal scan frequency (15.625 KHz for PAL) - this is in the range that younger people can hear, but older people can't (your hearing gets worse as you age - but at different rates for different people). So as a teeneger / mid-20s, at some point you'll find yourself in the company of other people of the same age where some of you can hear the (annoying) noise, and some can't.
I think a lot of switched mode power supplies also tend to operate in the same frequency at their standard load (their frequency can change depending on how much current is drawn) leading to a similar effect with all sorts of electronics : including perhaps the circuitry for the 30KV(ish) high voltage part of the CRT (which LCDs don't require)
From your description of the noise as a "whine", it seems more likely to be one of these two effects rather than the 50Hz or 60Hz vertical scan frequency (which most people would describe as a low "buzzing" sound rather than a "whining sound")
To add to that - if you have a good algorithm library available, make sure that you familiarize yourself with its documented performance - and use it : many man years of effort have (hopefully) been put in to ensuring that it works accurately and efficiently
The best example I can think of that is great for many mundane things is probably the C++ STL's algorithm and container template libraries - you have plenty of choice of containers to choose from, all with documented big O performance, and the fact that they are templates (should) let the compiler do as good a job at compile time optimization as you're ever likely to get
<rant>one of my pet annoyances at the moment is that with the.NET libraries seem to have taken a step several years backwards on this front (compared to C++ STL or Java's class libraries) - out of the box their 'collections' libraries appear to basically give you a choice of a vector (Array) or a hash table : nothing comparable to the STL's set, deque, linkedlist or map (red-black tree based dictionary) which must form the foundations of many algorithms for simple data processing</rant>
I saw an article in an IEEE magazine a few years ago about a French project to trial wireless power transmission for a remote village on a mountainous island - It may be small scale, but it does seem to be a real working system, which is a step towards what would be required for whats being discussed here
Q1. Do Intel products contain lead?
A1. Yes, most of our products contain lead in very small amounts. The use of lead in very small quantities in electronic products is ubiquitous. Lead is found throughout electronic components, component packaging, printed circuit boards, and other products. Intel estimates that approximately 90% of all electronic components contain some lead -- mostly due to the use of solder that contains lead. For a desktop computer with a CRT monitor, Intel's products contribute <1% of the total lead in the computer system. A typical Intel microprocessor contains approximately 0.2 grams of lead. An Intel motherboard contains roughly 2-3 grams of lead. For comparison purposes, a house key contains about the same amount of lead (2-3 g) as an Intel motherboard.
What could help turn them in to real spectator sports (if its not been done already - I haven't seriously played any games for a while so might have missed it) is to add special "camera man" players, who unlike in the physical world can be invisible, indestructable and able to move, fly and/or see through walls - with a competent commentator and good production crew an exciting spectacle for the crowds can then be put on in real time (rather than just in the after-game playback/review mode, which have been in games for a considerable time)
Obviously there are problems in doing this in a way which doesn't clash with cheat-prevention when players are distributed across the internet - but in an organised event the players screens will be being closely monitored anyway, so they cannot use that kind of cheat, or be fed information from spectators in the crowd who can see the other teams positions.
I'd assummed that the estimates of greenhouse gas output were simply based on raw fuel consumption figures for the country - doesn't burning a litre of petrol produce pretty much the same amount of CO2 at the end of the day whether it is used in a 2 stroke petrol lawn mower or the most modern of 'clean' cars? (I don't know the answer to this, in an inefficient/badly tuned engine what sort of percentage of the carbon ends up as soot or something else rather than CO2?)
Really large fuel consumers may take steps to reduce emmissions, but it is not unreasonable to measure these up individually (e.g. lime scrubbers in some power stations), but small scale consumers tend not to do anything like this (or the effect could just be averaged out if they do)
I tried Dia some time ago and found it very frustating to use, with most of built in objects being too restrictive and lacking enough variety of connection points to terminate lines and arrows.
For line diagrams (e.g. for documenting 3D graphics algorithms), I've moved on to qcad, which is a 2D CAD package.
For OO design, you can try doing this backwards : prototype the interfaces and classes and run doxygen (with dot) to generate inheritance and collaboration diagrams or XML output for post-processing. This probably fits in more with an extreme programming prototype/refactor style methodology than a complete up-front design methodology.
For state diagrams, I've had some success using dot on its own.
Using doxygen and dot has the significant additional feature that the files you edit are all plain-text files, so you can use your existing source revision control system (e.g. CVS)
Even with these packages however, I usually still find a pen and paper (or a white-board or blackboard if available) the easiest and fastest option for sketching out an initial design, or during discussions of various design alternatives with colleagues.
Ignoring your (sarcastic?) main point, instead re:
I mean, seriously, most BIOSes are bigger than 1.44mb now!
How long then before mainstrean manufacturers start shipping a standardized small "boot" or "rescue" O.S in their bios, with some kind of user interface, core device drivers and a few tools, available at boot time either in addition to or as a replacement for the bios options screen.
It would seem pretty simple to just include DOS (or some open source alternative) using part of the bios as a bootable ROM-disk.
I have used a SUN computer some years ago that provided something like this - I can't remember the precise details but I think that it had a forth-like command line environment available at boot time which was stored in ROM / flash RAM
The article linked to in the parent post has quite a lot more interesting data in it than the main article (except for the lack of quoted error margins on the results)
For most of those tests, the two compilers seem to show quite similar results, with one outperforming the other by only small margins of 10% or less (gcc has marginally better scores for the FFT routine on PIII and MazeBench, intel has slightly better scores for the Stepanov benchmark and LU decomposition for example). A few interesting differences do appear though that it would be nice to know more about:
The Intel compiler scores approximately 2.5 times as much as gcc 3.2.1 for the Monte-Carlo simulation, which is a considerably larger margin than for any of the other parts of the SciMark 2.0 benchmark
The executable size created for the MazeBench executable by the Intel compiler is around 3.5 times as large as that produced by gcc, wheras most of the other executables are within about 25%
g++'s output has more than a factor 17 performance difference in the OOPack Complex number benchmark on the P4 between its C and C++ implementations : all of the other tests have performance differences within a factor of 2 (with most having a negligible difference, some in favour of C, some in favour of C++)
Target specific object code and executable file format notwithstanding, what really makes the Intel compiler so very interesting is the fact that it delivers the same instruction sequences on both Linux and Windows platforms.
Doesn't gcc also have the same property when running under Linux or windows? (for the OS-independant parts of your code, obviously)
Although the main thrust of the article is to compare the new intel compiler to gcc and visual studio, they are also talking about comparing performance accross platforms - their initial comparison is gcc on Linux vs Visual Studio on Windows (no mention of what optimizations are performed, what other service are running, what nice levels the programs are executed at, whether they are measuring user cpu time or total execution time, and so on)
It would be nice to see tests varying each of these variables individually, or showing all the possible combinations:
Operating System : Linux 2.4, Linux 2.5, Windows XP. I wouldn't expect the O.S. to have any impact on the performance of computationally heavy stuff that the compiler can affect (but would make a big difference when it comes to memory management, and any IO)
Compiler : Gcc, Intel C++, Visual Studio
Optimizations : no optimizations (should theoretically all be the same?), max optimizations without violating ANSI/IEEE rules, max optimizations not caring about rules (e.g. using -ffast-math for gcc), enabling or disabling different targets (ie compile for 386 against compile for 686). Maybe even anti-optimizations such as turning on some extra debug/profiling options under each compiler (the time taken to create and debug a program can be more important the time it takes to execute)
On a (big) plus point - this article does at least show us some error margins on their measurements (none of this "Graphics card A lagged behind with a mere 47.6, but graphics card B stormed ahead with a score of 47.7" nonsense which seems to be all too common in most reviews)
They've really simply discovered that dynamically generating essentially static content is a bad idea : the 'dynamic' pages they are talking of are just articles which once written stay the same, and so are serving identical pages to each user.
Using scripting with database look ups to create such pages is obviously not good - much better is to compile your data in to static pages and serve those. I have done this for my own website using XSLT to generate the html pages with consistant links and menu's etc. - but you do have to remember to re-build it after making any changes or adding new content (I use gnu make to handle the dependancies of one page upon another so it doesn't rebuild the entire site everytime.)
They've taken the alternative approach of still using a database for the requests, but then caching future requests for the same page-id's, which has the advantage of being compatible with their original dynamic generation system, but they don't mention how they handle the dependancy / cascading alterations problem if they change the content (though they could always flush the entire cache of course....).
Neither of these approaches can help you though if you have real dynamic pages where every request is unique or there are are too many possible pages for caching to be feasible (for example amazon or google).
Artificial gravity????
on
Robocoaster
·
· Score: 2
We could try doing artificial gravity with diamagnetism if the Levitating frog experiment can be scaled up to work for Humans.
This same mechanism could also do star-trek style inertial damping...
Most of the data on your hard disk is just your local cache of software/data from 'the internet' - so you could treat your hard drive as mostly cache, with a small partition with your own/original data which you are hosting locally.
There would be many many problems to resolve before doing it (especially regarding security) - but the advantages would be that you'd effectivly have the latest versions of _all_ the worlds software available to you in your/usr or "Program Files" directory (though obviously you might not be able to use all of it without paying for licence keys)
I have used freeuk's dial up service (with both Windows and Linux - even though its not listed as supported by them)
Although I haven't used it for a while (I think the last time was 9 months ago whilst setting up a new office whilst waiting for our DSL line to be fitted) one used to be able to sign up instantly for a free account.
There are many other free ISP's - just do a google search
You have to register (your name and email address) and agree to some terms and conditions to be allowed to read it.
Interestingly they also thought it important to answer the seemingly pointless question : how many API's are there? with a bar chart showing the number in a variety of specifications!
Example: Mt. St. Helens rocks were dated at (varies by sample, that alone should raise red flags about the radiometric dating concept) values ranging from 350,000 to 2,800,000 years
<RantMode>
Why do people (the 'media' in particular) always miss out the error margins when quoting scientific results?
The first thing you should learn when doing any form of quantitative science is error analysis: without this all the results are meaningless as you have no idea of the certainty or significance of them.
For example, with these results you quote, if the measured result were actually 2.8 Million years +/- 2 million years, then that would not give any cause for concern, or 350 thousand years +/- 10 million years would be as good as spot on. OTOH if the results were 2.8 Million +/- 1 year then we should definitely be questioning the accuracy of such measurements.
but the original point in the thread was that if the manufacturer is giving you a 3 year warranty then the manufacturer is confident that the drive will probably not fail for at least 3 years.
By reducing the warranty period, they're giving the impression to us customers that that they are not confident that the new drives will last 3 years in operation, or to extrapolate further, that new drives are quite likely to develop faults between 1 and 3 years from installation.
If you can't imagine keeping a drive in active use for three years, then it seems that hard drive manufacturers have already caused you to lose confidence in their reliability.
or china, who claimed something similar a few years ago. from this 1999 article:
The Chinese appear to have turned things around. Their new Passive Coherent Location system allegedly tracks fluctuations in normal or civilian broadcasts, which function at a different frequency, and looks for distortions, reports Newsweek.
Out of interest (but possibly off-topic) do you know of anywhere that has a survey of what different chipsets and different monitors are capable of?
For my current computer I purchased a CTX PV880C which was spec'ed to do 1280x1024 at 75Hz (and has a DVI-digital input, unlike many other cheap panels...), and a Matrox G550 to drive it (it was the cheapest card I could find that advertised Linux drivers and DVI digital output), but was initally surprised that if I run it at much more than 60Hz then I get interference / noise appearing on the screen (but not in analogue mode).
Not that I'm complaining much - I'd have thought that on an LCD anything above cinama-film rate (around 24Hz?) would be sufficient.
Obviously you can't parse the http header for the browser type until after you've already set up the ssl connection, which you won't have been able to do if the browser was not susceptible.....
However, the attack would still work, you just rely on grabbing enough passwords and stealing enough money before being discovered and shutdown to make it workwhile.
also re: > that your conducting should read "that you're conducting"
outlay a bit of cash to set up or buy in to a business that let me provide a bit of the internet backbone.
One day substitute a computer doing a man-in-the-middle attack on a selection of banks and online share-dealers for one of the routers (or simply insert it between two running routers, if there is nobody around to notice the extra box). It doesn't need to do anything clever - it can just act as a proxy to the to the intended receipient (forwarding all the incoming http requests in new ssl connections) and log everything.
After a suitable length of time, swap out the router and take the logs away, hopefully before anybody discovers the attack so that nobody even knows that it has happened.
At you leisure, go through the logs and pull out peoples account information, usernames and passwords from the logs, then use them to log in and buy/sell/transfers shares and money to your hearts contents (obviously anyone doing organized crime on this scale will also need to set up some kind of money laundering scheme to make the money untraceable)
The chances of being discovered during the time that your conducting the attack can be minimized by parsing the http headers for the browser type, and only attempting the attack for clients using vulnerable browsers. This way you could leave it in operation for longer, and steal more information.
So what have I missed here? is there some other aspect to this that makes it more complicated than I've made out? stealing the certificates was meant to be the difficult part, getting access to the network is not difficult if you are big enough, and creating a transparent-proxy is going to be relativly easy.
After this story, and one yesterday aboutt he size of drinks cups also in ounces, I've looked up the conversion rates to see what you're all talking about
1 American fluid ounce = 29.57 millilitres.
1 British fluid ounce = 28.41 millilitres
So the American 8oz cup is just under a quarter litre (or around halfway between a 1/3 and a 1/2 English pint, if that's more your kind of reference size....)
Slashdot Mirror
I think I understand the thinking behind having patents - as a way to prevent others from profiting from a patent owner's hard work on researching and implementing a solution to some problem merely by seeing their solution and copying it.
HOWEVER - it seems that many new inventions come about due to the combination of current scientific knowledge, current technology and current problems. This often leads to the same (or very similar) inventions being independantly made by several people across the world without them necessarily 'stealing' the idea from one another, and yet the patent system as currently implemented appears to try and deny all except one of the discovers from using their invention.
A google search for something like 'simultaneous invention' will turn up many articles listing some of the more famous examples such as Calculus (Newton and Leibniz), the car (Daimler and Benz), the telephone and so on.. (e.g. http://www.newyorker.com/reporting/2008/05/12/080512fa_fact_gladwell?currentPage=all or http://german.about.com/library/weekly/aa030501a.htm)
Now on the one hand you could argue that instead of bothering with all this research and risking making an unusable independant discovery you should put all that effort in to searching for and licensing other's patents (which would ultimately lead to stagnation - see Asimov's foundation series)
More likely the solution is either not published when you start the research, or (in the case of programming, much of which is researching and 'inventing' ways for a computer to do some task) you do not think of your solution as an 'invention' that has a patent but rather just 'one obvious (to me) algorithm / UI to crack this sub-problem in creating something to meet my current project's requirements'
re: They do not, and never been able to, provide any verification of who is on either end.
They do provide a _little_ verification - so long as you permanently remember the certificate it verifies that whoever is at the other end of the connection on subsequent sessions is the same as whoever is at the other end now (so long as they've kept their private cert suitably secure)
aside: is there _any_ way of ever verifying 'who' someone / something is beyond 'you are the same person that I saw that other time' or 'you are the same person that someone intermediary claims they saw in the past'-> which if you follow it through leads to a better understanding of what a person's identity is and is not, and what you can/can't achieve when requiring people to present 'identification' for security purposes - and a misunderstanding of which leads to a lot of silly security 'theatre'
Exactly - and this highlights the problems we currently face with _not_ having some universally (or at least nationally) accepted form of identity proof.
Ideally we need to have some form of secret ID proving information (the only one I can think of is to use RSA style certificates with proof of ID involving possession of a physical device that can authenticate using its private key). The nearest that most people have at the moment and are understood to be private are :
- PIN numbers for credit/debit/cashpoint card (problem here being that you have to type in the 4 digit PIN in plain-text to various untrusted devices, so it can easily be stolen : though with chip-and-pin I don't know if the chip is doing some fancy public key handshake thing or not)
- Demonstrating that you can write your signature in a reasonably natural looking manner
- Possession of a (relatively) difficult document to forge / copy such as passport or driving license (the obvious problem - they are not really _that_ difficult to copy, and secondly there is no trusted relationship between your identity as verified by possession of the passport and other day to day activities such as trying to arrange to transfer money from your bank account to a shopkeeper to pay for goods at the checkout)
Unfortunately many places erroneously trust various other pieces of non-private data to authenticate people such as :
- Use of national-insurance number (is that equivilent to american SSN?)
- bank account sort code and account number
- knowledge of name, address and phone number
- a facsimile of a signature
- your company employee numbeer, salary, favourite colour and pet's name.
Information such as this is not 'secret' and so should not be treated as private data sufficient for authentication (and subsequent authorization) from a security perspective, despite many people thinking of it as 'private' in social terms.
So in conclusion I sympathise with Clarkson - the information he published should not have been sufficient for identity theft to occur, the problem is with those that accepted the information as valid proof of having authority to transfer money out of his bank account.
A much more noticable effect is the horizontal scan frequency (15.625 KHz for PAL) - this is in the range that younger people can hear, but older people can't (your hearing gets worse as you age - but at different rates for different people).
So as a teeneger / mid-20s, at some point you'll find yourself in the company of other people of the same age where some of you can hear the (annoying) noise, and some can't.
I think a lot of switched mode power supplies also tend to operate in the same frequency at their standard load (their frequency can change depending on how much current is drawn) leading to a similar effect with all sorts of electronics : including perhaps the circuitry for the 30KV(ish) high voltage part of the CRT (which LCDs don't require)
From your description of the noise as a "whine", it seems more likely to be one of these two effects rather than the 50Hz or 60Hz vertical scan frequency (which most people would describe as a low "buzzing" sound rather than a "whining sound")
re: pick the right algorithms
To add to that - if you have a good algorithm library available, make sure that you familiarize yourself with its documented performance - and use it : many man years of effort have (hopefully) been put in to ensuring that it works accurately and efficiently
The best example I can think of that is great for many mundane things is probably the C++ STL's algorithm and container template libraries - you have plenty of choice of containers to choose from, all with documented big O performance, and the fact that they are templates (should) let the compiler do as good a job at compile time optimization as you're ever likely to get
<rant>one of my pet annoyances at the moment is that with the .NET libraries seem to have taken a step several years backwards on this front (compared to C++ STL or Java's class libraries) - out of the box their 'collections' libraries appear to basically give you a choice of a vector (Array) or a hash table : nothing comparable to the STL's set, deque, linkedlist or map (red-black tree based dictionary) which must form the foundations of many algorithms for simple data processing</rant>
I saw an article in an IEEE magazine a few years ago about a French project to trial wireless power transmission for a remote village on a mountainous island - It may be small scale, but it does seem to be a real working system, which is a step towards what would be required for whats being discussed here
A little googling for it eventually turns up this English language page : Grand Bassin (Réunion Island) Wireless Power Transmission , but I couldn't find very much technical information on it.
from http://www.intel.com/research/silicon/leadfree.ht
What could help turn them in to real spectator sports (if its not been done already - I haven't seriously played any games for a while so might have missed it) is to add special "camera man" players, who unlike in the physical world can be invisible, indestructable and able to move, fly and/or see through walls - with a competent commentator and good production crew an exciting spectacle for the crowds can then be put on in real time (rather than just in the after-game playback/review mode, which have been in games for a considerable time)
Obviously there are problems in doing this in a way which doesn't clash with cheat-prevention when players are distributed across the internet - but in an organised event the players screens will be being closely monitored anyway, so they cannot use that kind of cheat, or be fed information from spectators in the crowd who can see the other teams positions.
I'd assummed that the estimates of greenhouse gas output were simply based on raw fuel consumption figures for the country - doesn't burning a litre of petrol produce pretty much the same amount of CO2 at the end of the day whether it is used in a 2 stroke petrol lawn mower or the most modern of 'clean' cars? (I don't know the answer to this, in an inefficient/badly tuned engine what sort of percentage of the carbon ends up as soot or something else rather than CO2?)
Really large fuel consumers may take steps to reduce emmissions, but it is not unreasonable to measure these up individually (e.g. lime scrubbers in some power stations), but small scale consumers tend not to do anything like this (or the effect could just be averaged out if they do)
I tried Dia some time ago and found it very frustating to use, with most of built in objects being too restrictive and lacking enough variety of connection points to terminate lines and arrows.
For line diagrams (e.g. for documenting 3D graphics algorithms), I've moved on to qcad, which is a 2D CAD package.
For OO design, you can try doing this backwards : prototype the interfaces and classes and run doxygen (with dot) to generate inheritance and collaboration diagrams or XML output for post-processing. This probably fits in more with an extreme programming prototype/refactor style methodology than a complete up-front design methodology.
For state diagrams, I've had some success using dot on its own.
Using doxygen and dot has the significant additional feature that the files you edit are all plain-text files, so you can use your existing source revision control system (e.g. CVS)
Even with these packages however, I usually still find a pen and paper (or a white-board or blackboard if available) the easiest and fastest option for sketching out an initial design, or during discussions of various design alternatives with colleagues.
Ignoring your (sarcastic?) main point, instead re:
How long then before mainstrean manufacturers start shipping a standardized small "boot" or "rescue" O.S in their bios, with some kind of user interface, core device drivers and a few tools, available at boot time either in addition to or as a replacement for the bios options screen.
It would seem pretty simple to just include DOS (or some open source alternative) using part of the bios as a bootable ROM-disk.
I have used a SUN computer some years ago that provided something like this - I can't remember the precise details but I think that it had a forth-like command line environment available at boot time which was stored in ROM / flash RAM
The article linked to in the parent post has quite a lot more interesting data in it than the main article (except for the lack of quoted error margins on the results)
For most of those tests, the two compilers seem to show quite similar results, with one outperforming the other by only small margins of 10% or less (gcc has marginally better scores for the FFT routine on PIII and MazeBench, intel has slightly better scores for the Stepanov benchmark and LU decomposition for example). A few interesting differences do appear though that it would be nice to know more about:
re:
Doesn't gcc also have the same property when running under Linux or windows? (for the OS-independant parts of your code, obviously)
Although the main thrust of the article is to compare the new intel compiler to gcc and visual studio, they are also talking about comparing performance accross platforms - their initial comparison is gcc on Linux vs Visual Studio on Windows (no mention of what optimizations are performed, what other service are running, what nice levels the programs are executed at, whether they are measuring user cpu time or total execution time, and so on)
It would be nice to see tests varying each of these variables individually, or showing all the possible combinations:
On a (big) plus point - this article does at least show us some error margins on their measurements (none of this "Graphics card A lagged behind with a mere 47.6, but graphics card B stormed ahead with a score of 47.7" nonsense which seems to be all too common in most reviews)
They've really simply discovered that dynamically generating essentially static content is a bad idea : the 'dynamic' pages they are talking of are just articles which once written stay the same, and so are serving identical pages to each user.
Using scripting with database look ups to create such pages is obviously not good - much better is to compile your data in to static pages and serve those. I have done this for my own website using XSLT to generate the html pages with consistant links and menu's etc. - but you do have to remember to re-build it after making any changes or adding new content (I use gnu make to handle the dependancies of one page upon another so it doesn't rebuild the entire site everytime.)
They've taken the alternative approach of still using a database for the requests, but then caching future requests for the same page-id's, which has the advantage of being compatible with their original dynamic generation system, but they don't mention how they handle the dependancy / cascading alterations problem if they change the content (though they could always flush the entire cache of course....).
Neither of these approaches can help you though if you have real dynamic pages where every request is unique or there are are too many possible pages for caching to be feasible (for example amazon or google).
We could try doing artificial gravity with diamagnetism if the Levitating frog experiment can be scaled up to work for Humans.
This same mechanism could also do star-trek style inertial damping...
Most of the data on your hard disk is just your local cache of software/data from 'the internet' - so you could treat your hard drive as mostly cache, with a small partition with your own/original data which you are hosting locally.
There would be many many problems to resolve before doing it (especially regarding security) - but the advantages would be that you'd effectivly have the latest versions of _all_ the worlds software available to you in your /usr or "Program Files" directory (though obviously you might not be able to use all of it without paying for licence keys)
I have used freeuk's dial up service (with both Windows and Linux - even though its not listed as supported by them)
Although I haven't used it for a while (I think the last time was 9 months ago whilst setting up a new office whilst waiting for our DSL line to be fitted) one used to be able to sign up instantly for a free account.
There are many other free ISP's - just do a google search
If you follow through a few levels of links from the artice, the standard is available online (for free) at:
http://www.unix.org/version3/online.html
You have to register (your name and email address) and agree to some terms and conditions to be allowed to read it.
Interestingly they also thought it important to answer the seemingly pointless question : how many API's are there? with a bar chart showing the number in a variety of specifications!
re:
<RantMode>
Why do people (the 'media' in particular) always miss out the error margins when quoting scientific results? The first thing you should learn when doing any form of quantitative science is error analysis: without this all the results are meaningless as you have no idea of the certainty or significance of them.
For example, with these results you quote, if the measured result were actually 2.8 Million years +/- 2 million years, then that would not give any cause for concern, or 350 thousand years +/- 10 million years would be as good as spot on. OTOH if the results were 2.8 Million +/- 1 year then we should definitely be questioning the accuracy of such measurements.
<RantMode>but the original point in the thread was that if the manufacturer is giving you a 3 year warranty then the manufacturer is confident that the drive will probably not fail for at least 3 years.
By reducing the warranty period, they're giving the impression to us customers that that they are not confident that the new drives will last 3 years in operation, or to extrapolate further, that new drives are quite likely to develop faults between 1 and 3 years from installation.
If you can't imagine keeping a drive in active use for three years, then it seems that hard drive manufacturers have already caused you to lose confidence in their reliability.
or china, who claimed something similar a few years ago. from this 1999 article:
The Chinese appear to have turned things around. Their new Passive Coherent Location system allegedly tracks fluctuations in normal or civilian broadcasts, which function at a different frequency, and looks for distortions, reports Newsweek.
Out of interest (but possibly off-topic) do you know of anywhere that has a survey of what different chipsets and different monitors are capable of?
For my current computer I purchased a CTX PV880C which was spec'ed to do 1280x1024 at 75Hz (and has a DVI-digital input, unlike many other cheap panels...), and a Matrox G550 to drive it (it was the cheapest card I could find that advertised Linux drivers and DVI digital output), but was initally surprised that if I run it at much more than 60Hz then I get interference / noise appearing on the screen (but not in analogue mode).
Not that I'm complaining much - I'd have thought that on an LCD anything above cinama-film rate (around 24Hz?) would be sufficient.
Somewhere there I wasn't thinkging straight:
.....
Obviously you can't parse the http header for the browser type until after you've already set up the ssl connection, which you won't have been able to do if the browser was not susceptible
However, the attack would still work, you just rely on grabbing enough passwords and stealing enough money before being discovered and shutdown to make it workwhile.
also re:
> that your conducting
should read
"that you're conducting"
then I would:
The chances of being discovered during the time that your conducting the attack can be minimized by parsing the http headers for the browser type, and only attempting the attack for clients using vulnerable browsers. This way you could leave it in operation for longer, and steal more information.
So what have I missed here? is there some other aspect to this that makes it more complicated than I've made out? stealing the certificates was meant to be the difficult part, getting access to the network is not difficult if you are big enough, and creating a transparent-proxy is going to be relativly easy.
After this story, and one yesterday aboutt he size of drinks cups also in ounces, I've looked up the conversion rates to see what you're all talking about
from this web site :
1 American fluid ounce = 29.57 millilitres.
1 British fluid ounce = 28.41 millilitres
So the American 8oz cup is just under a quarter litre (or around halfway between a 1/3 and a 1/2 English pint, if that's more your kind of reference size....)