There was a method of extracting SSL Keys out of a web browser which may be able to be used here. It works on the principle that code is not random but crypto keys appear to be random. So in order to find the key in memory all you have to do is find the areas of the applications memory that has highest entropy and that is highly likely to be the crypto key.
Digital TV is a step forwards because channels can be multiplexed over a smaller number of frequencies. This reduces the amount of the spectrum which is needed for TV broadcasts and hence more can be sold off by governments for newer technologies from which they can make a fortune in licensing.
I would be very concerned here in Hull, UK if sea levels rise as the River Humber is tidal and some areas of the city and nearby area are below sea level. London would have similar problems with a rising in the level of the Thames. There loads of other areas around the world which would suffer similar problems.
My MP3 player can already hold every video ever made (or going to be made for that matter). I'm just having problems storing the keys I need to find them - some of them can be over 700MBytes.
VAT is 17.5% in this country and applied to everything except food, books, children's clothes and a couple of other things. It has been around for so long that people don't really think about it - all prices except wholesale prices are quoted with VAT already added so most people don't think about it.
A lot of problems can be caused by the lack of a coherent timing signal across even logical, let alone physical, processors. This is the main reason behind cut of lines, out of sync video etc bugs which affects a lot of older games designed to only one processor.
Unless virus makers work out how to access the kernel by using the mechanism that has been added to pander to the security companies which would otherwise be impossible. I foresee an incident where sloppy security at a security company means that someone get hold of a private key (I'm assuming this is how this will work) and write a virus and sign it using said key and everyone blames Microsoft for poor security. Also, if this is done by digital certifcate, what constitutes a security company who deserve access to the kernal. What's stopping me from setting up "Oggiejnr's Antivirus" and then claiming that I have to be allowed to hook the kernel as well? Once I have the key I can do what I want in the kernel and the whole system is useless
I think that the greater danger comes when you have two states like India and Pakistan both with small number of nuclear weapons and a pre-existing conflict. All it takes is one side to look at the other's nuclear capability and to think that it could survive an attack (lose a few cities maybe but still be relatively functional) and you have far more likelihood of a nuclear war.
Eventually computer programming languages will be able to scale out even relatively simple systems across multiple execution paths. One way in which I see this progressing is, providing the language is type safe and garbage collected, it should be a relatively simple exercise to determine what objects a particular function will touch given a certain input (providing the execution path is non-branching in the simplest model).
If it is possible to show that two event handlers, for example window messages, will not write to the same object and that one will not read to the same object that one writes, it is safe to execute the handlers in parallel across multiple cores rather than sequentially in the core as would happen now leading to a more responsive user experience as messages (in general) are not waiting to be executed. This is one are in which multiple cores, with a corresponding development in languages, could be beneficial. In some specialized circumstances it could be possible to add this type of parallelisation (is that a word) retrospectively to programs which are either interpretted or run from bytecode. Similar things could be done with running a for loop across multiple processes (as long is it can be shown that the different branches would not write to each other in the wrong order) hence the computation could be done across multiple cores which then either write independently to memory (if the locations are different) or sequentially which the processing already done to an external system. A similar approach could be used for reads.
All of this requires increases in the ability of static analysis to determine relevant code paths and it may be determined that Godel's theorem does not permit the full implementation. However considering the type of thing that MSR and others has been able to prove about managed operating systems it may not be that limiting. It would also likely be needed that one core would have to be reserved for sheduling the others within an application however when 80 cores are being talked about this is not as much of a concern.
LatexSuite at http://vim-latex.sourceforge.net/ is brilliant for editing latex documents. It took me a couple of days to initially learn but the tutorial on their site makes starting easy.
There was a system designed that would allow a voter to keep a receipt of their vote which could be used for verification but does not reveal who the person voted for. It splits up the text of who was voted for into horizontal lines which are distributed over two seperate sheets of paper. Combined they produce readable text but not seperate. The image of the layers is stored in database. The bottom layer is filed into the ballot box and the top layer kept by the voter. In the event of a recont using paper, the layers can be brought from the database and verified against the paper record. This provides one solution for a verifiable electronic election.
In the UK any calculator capable of displaying stored text is automatically disallowed for GCSE and A-Level exams. Any calculator capable of storing programs on it has for wiped in front of a invigilator for the exam and anything that can do symbolic algebra is banned. Of course this requires that the people administering the exams know what they are doing. Unfortunately mine did so I wasn't allowed my Ti-89 anywhere near the exam hall - you could have fitted a fair amount of info in plain text on it 2.3Mb internal memory and it even has a basic e-book reader for it.
As far as I am aware S/PDIF sends clocking information along with the audio meaning that jitter in the connection is not relevant if between two digital stages. The problem is when the DAC relies on incoming signal without reclocking and hence any jitter in the signal is transferred into the anologue realm. If the two clocks worked independently as you describe then you would constantly be getting clicks in the signal. Whether or not jitter can be heard is another issue. If the signal is constantly underclocked or overclocked by a small margin then hearing the difference is unlikely however you,ay have trouble with clicks and other audio artifacts. However the more likely scenario is that the average of the clock is 44.1 kHz but the clock varies or reflections in an optical cable cause variations. In this case, is the DAC does not reclock the data to a steady internal clock before converting back to analog you can get pitch variations, which although minor, some people claim to be able to hear.
(Someone correct me if I'm wrong - is Apple Computer doing exclusive media deals with anyone?)
Judging by their ads on UK TV then yes. There was something about unreleased Miles Davis or similar available exclusively on iTunes. Whether this is online exclusive or genuine excluse I'm not sure and I also don't know if I dreamed it all
First things first - I don't pretend to be cultured I merely comment on things as I see them. I am a scientist (as opposed to an artist) and like Metal music so there is no way that I can possibly be cultured in your narrow sense of the word. Secondly I did not say that people paying money for it alone makes it art, I said that if people pay money for it on the basis of it being art it should be classified as art. In your examples the money is being paid, but not on the basis that they are art and hence are not art. Also I was using "modern art" in the sense that it is commonly used in the mass media and by most people so if you have a problem with that terminology take it to them not me. I hope this settles any points that you may have.
PS don't be a coward, I like to know to which user I'm talking - it makes continued conversation easier.
I've heard of the wonders of John Cage and believe that BBC Radio 3 did a live broadcast of the piece rescored for full orchestra - what a wonderful use ot the licence payers money. Though I believe there was a rational behind it being 433 in that it becomes 273 seconds, -273 degrees C being absolute zero. I also remember him planning on sueing someone for copying his "composition" but I'm not sure about that.
Re:There's no such thing as art
on
Are Videogames Art?
·
· Score: 4, Insightful
Given the current crop of so-called "modern art", I think is safe to say that the only definition of art that can be uniformly applied is that it is art if someone is willing to pay money for it on the basis of it being art.
In some mechanical patents the confusing jargon can be necessary to protect the invention - a patent that could be voided because someone replaced those so carefully labelled springs in your patent with elastic bands would be pretty much worthless - however most of the time I agree that it is used as nothing more than a device to patent the otherwise unpatentable. This could be done through the punishing of rejected applications which could be held to try and confuse their way into the register but you would still need an organisation capable of checking patent applications in far more detail than is done now. This may involve throwing money at the problem but maybe other ways could be put forward.
With regards to the laywers - I agree that lawyers are mostly to blame for this mess and it would be beneficial for professions other than lawyers and the ultra-rich end of the business spectrum to control the legislature but unfortunately I can't see this happening in the near future - one has the bullshitting skills and the other has the money.
One other way is that rather than banning software patents have them patentable for a shorter amount of time and actively enforce the non-obvious rules. In the drug industry, for example, the capital costs of research are much larger than in software and it can also take upto half the patentable life time for a drug to be certified safe for use. In circumstances like this it is argualbe that the current 20 year patent system is appropriate as it promotes research by private companies which can use the patent period to recoup research costs and turn the profit. Whether the profit they manage to turn is unethical is another discussion. In the software industry where, as has been stated has a much higher rate of progress, you could allow patents for say 3-5 years which would protect ideas in teh short term but not overly stifle progress due to the short expiry time. As long as the non-obvious and innovative rules were enforced this may make a compromise between the two opposing viewpoints.
One such example for this may well be the new Ribbon UI in MS Office 2007 which is, as far as I am aware, a new concept in user interfaces in getting rid of menus and simplifying everything down. While a 20 year patent on such an idea is complete overkill, a 3 year patent would permit MS to use it exclusively or licence it for a short amount of time but prevent this continuing for an entirnity (in the computer world at least).
How to turn this idea into a workable system is left as an excersise for overpaid lawyers.
Just a clarification -.Net is NOT interpreted it is JIT Compiled, although it can be compiled using ngen if you don't want the JIT overhead and once a method is JITed it runs fast enough for just about anything (barring high-end 3d games possible). I wrote a program to play sound files in a theatre setting and using C# and Managed DirectSound I could easily playback 30 tracks at once using a streaming buffer system including volume fading and all the rest of the rubbish that was needed. Testing a midi patch bay in C# compared to C++ the difference in speed (once Jitted) was less than could be measured using MIDIOX to generate and record midi data - the timestamp values for both programs were identical. So in the vast majority of situations.NET is perfectly fast enough as long as you take into account JIT overhead which can be precompiled out. There is also the issue of garbage collection but Java and Objective C (I think there was a/. article on this but I maybe wrong) have this as well.
Slashdot Mirror
There was a method of extracting SSL Keys out of a web browser which may be able to be used here. It works on the principle that code is not random but crypto keys appear to be random. So in order to find the key in memory all you have to do is find the areas of the applications memory that has highest entropy and that is highly likely to be the crypto key.
Digital TV is a step forwards because channels can be multiplexed over a smaller number of frequencies. This reduces the amount of the spectrum which is needed for TV broadcasts and hence more can be sold off by governments for newer technologies from which they can make a fortune in licensing.
I would be very concerned here in Hull, UK if sea levels rise as the River Humber is tidal and some areas of the city and nearby area are below sea level. London would have similar problems with a rising in the level of the Thames. There loads of other areas around the world which would suffer similar problems.
My MP3 player can already hold every video ever made (or going to be made for that matter). I'm just having problems storing the keys I need to find them - some of them can be over 700MBytes.
VAT is 17.5% in this country and applied to everything except food, books, children's clothes and a couple of other things. It has been around for so long that people don't really think about it - all prices except wholesale prices are quoted with VAT already added so most people don't think about it.
A lot of problems can be caused by the lack of a coherent timing signal across even logical, let alone physical, processors. This is the main reason behind cut of lines, out of sync video etc bugs which affects a lot of older games designed to only one processor.
Unless virus makers work out how to access the kernel by using the mechanism that has been added to pander to the security companies which would otherwise be impossible. I foresee an incident where sloppy security at a security company means that someone get hold of a private key (I'm assuming this is how this will work) and write a virus and sign it using said key and everyone blames Microsoft for poor security. Also, if this is done by digital certifcate, what constitutes a security company who deserve access to the kernal. What's stopping me from setting up "Oggiejnr's Antivirus" and then claiming that I have to be allowed to hook the kernel as well? Once I have the key I can do what I want in the kernel and the whole system is useless
I haven't experienced the problem you describe. Both VS.NET 2003 C++ and C# standard editions and C# 2005 still work fine with beta 3 of IE7.
Or if you don't mind not having a hard copy its available on steam for $10 pre-launch
I think that the greater danger comes when you have two states like India and Pakistan both with small number of nuclear weapons and a pre-existing conflict. All it takes is one side to look at the other's nuclear capability and to think that it could survive an attack (lose a few cities maybe but still be relatively functional) and you have far more likelihood of a nuclear war.
Eventually computer programming languages will be able to scale out even relatively simple systems across multiple execution paths. One way in which I see this progressing is, providing the language is type safe and garbage collected, it should be a relatively simple exercise to determine what objects a particular function will touch given a certain input (providing the execution path is non-branching in the simplest model).
If it is possible to show that two event handlers, for example window messages, will not write to the same object and that one will not read to the same object that one writes, it is safe to execute the handlers in parallel across multiple cores rather than sequentially in the core as would happen now leading to a more responsive user experience as messages (in general) are not waiting to be executed. This is one are in which multiple cores, with a corresponding development in languages, could be beneficial. In some specialized circumstances it could be possible to add this type of parallelisation (is that a word) retrospectively to programs which are either interpretted or run from bytecode. Similar things could be done with running a for loop across multiple processes (as long is it can be shown that the different branches would not write to each other in the wrong order) hence the computation could be done across multiple cores which then either write independently to memory (if the locations are different) or sequentially which the processing already done to an external system. A similar approach could be used for reads.
All of this requires increases in the ability of static analysis to determine relevant code paths and it may be determined that Godel's theorem does not permit the full implementation. However considering the type of thing that MSR and others has been able to prove about managed operating systems it may not be that limiting. It would also likely be needed that one core would have to be reserved for sheduling the others within an application however when 80 cores are being talked about this is not as much of a concern.
You could always use Eclipse inside VIM http://sourceforge.net/projects/eclim/ though I can't get my head round it someone else might.
LatexSuite at http://vim-latex.sourceforge.net/ is brilliant for editing latex documents. It took me a couple of days to initially learn but the tutorial on their site makes starting easy.
There was a system designed that would allow a voter to keep a receipt of their vote which could be used for verification but does not reveal who the person voted for. It splits up the text of who was voted for into horizontal lines which are distributed over two seperate sheets of paper. Combined they produce readable text but not seperate. The image of the layers is stored in database. The bottom layer is filed into the ballot box and the top layer kept by the voter. In the event of a recont using paper, the layers can be brought from the database and verified against the paper record. This provides one solution for a verifiable electronic election.
In the UK any calculator capable of displaying stored text is automatically disallowed for GCSE and A-Level exams. Any calculator capable of storing programs on it has for wiped in front of a invigilator for the exam and anything that can do symbolic algebra is banned. Of course this requires that the people administering the exams know what they are doing. Unfortunately mine did so I wasn't allowed my Ti-89 anywhere near the exam hall - you could have fitted a fair amount of info in plain text on it 2.3Mb internal memory and it even has a basic e-book reader for it.
As far as I am aware S/PDIF sends clocking information along with the audio meaning that jitter in the connection is not relevant if between two digital stages. The problem is when the DAC relies on incoming signal without reclocking and hence any jitter in the signal is transferred into the anologue realm. If the two clocks worked independently as you describe then you would constantly be getting clicks in the signal. Whether or not jitter can be heard is another issue. If the signal is constantly underclocked or overclocked by a small margin then hearing the difference is unlikely however you ,ay have trouble with clicks and other audio artifacts. However the more likely scenario is that the average of the clock is 44.1 kHz but the clock varies or reflections in an optical cable cause variations. In this case, is the DAC does not reclock the data to a steady internal clock before converting back to analog you can get pitch variations, which although minor, some people claim to be able to hear.
I'm pretty sure that this is more reasonable
http://us.kensington.com/html/6408.html
Though I have to say I won't mind WiredZone's profit margin
(Someone correct me if I'm wrong - is Apple Computer doing exclusive media deals with anyone?)
Judging by their ads on UK TV then yes. There was something about unreleased Miles Davis or similar available exclusively on iTunes. Whether this is online exclusive or genuine excluse I'm not sure and I also don't know if I dreamed it all
In response to an AC
First things first - I don't pretend to be cultured I merely comment on things as I see them. I am a scientist (as opposed to an artist) and like Metal music so there is no way that I can possibly be cultured in your narrow sense of the word. Secondly I did not say that people paying money for it alone makes it art, I said that if people pay money for it on the basis of it being art it should be classified as art. In your examples the money is being paid, but not on the basis that they are art and hence are not art. Also I was using "modern art" in the sense that it is commonly used in the mass media and by most people so if you have a problem with that terminology take it to them not me. I hope this settles any points that you may have.
PS don't be a coward, I like to know to which user I'm talking - it makes continued conversation easier.
I've heard of the wonders of John Cage and believe that BBC Radio 3 did a live broadcast of the piece rescored for full orchestra - what a wonderful use ot the licence payers money. Though I believe there was a rational behind it being 433 in that it becomes 273 seconds, -273 degrees C being absolute zero. I also remember him planning on sueing someone for copying his "composition" but I'm not sure about that.
Given the current crop of so-called "modern art", I think is safe to say that the only definition of art that can be uniformly applied is that it is art if someone is willing to pay money for it on the basis of it being art.
In some mechanical patents the confusing jargon can be necessary to protect the invention - a patent that could be voided because someone replaced those so carefully labelled springs in your patent with elastic bands would be pretty much worthless - however most of the time I agree that it is used as nothing more than a device to patent the otherwise unpatentable. This could be done through the punishing of rejected applications which could be held to try and confuse their way into the register but you would still need an organisation capable of checking patent applications in far more detail than is done now. This may involve throwing money at the problem but maybe other ways could be put forward.
With regards to the laywers - I agree that lawyers are mostly to blame for this mess and it would be beneficial for professions other than lawyers and the ultra-rich end of the business spectrum to control the legislature but unfortunately I can't see this happening in the near future - one has the bullshitting skills and the other has the money.
One such example for this may well be the new Ribbon UI in MS Office 2007 which is, as far as I am aware, a new concept in user interfaces in getting rid of menus and simplifying everything down. While a 20 year patent on such an idea is complete overkill, a 3 year patent would permit MS to use it exclusively or licence it for a short amount of time but prevent this continuing for an entirnity (in the computer world at least).
How to turn this idea into a workable system is left as an excersise for overpaid lawyers.
Reminds of a quote from somewhere that he who has the ability to take power should in no way be allowed it.
Just a clarification - .Net is NOT interpreted it is JIT Compiled, although it can be compiled using ngen if you don't want the JIT overhead and once a method is JITed it runs fast enough for just about anything (barring high-end 3d games possible). I wrote a program to play sound files in a theatre setting and using C# and Managed DirectSound I could easily playback 30 tracks at once using a streaming buffer system including volume fading and all the rest of the rubbish that was needed. Testing a midi patch bay in C# compared to C++ the difference in speed (once Jitted) was less than could be measured using MIDIOX to generate and record midi data - the timestamp values for both programs were identical. So in the vast majority of situations .NET is perfectly fast enough as long as you take into account JIT overhead which can be precompiled out. There is also the issue of garbage collection but Java and Objective C (I think there was a /. article on this but I maybe wrong) have this as well.