Use a good VersionControl system, git, murcurial or Subversion if you must, __but__ keep the meta-data out of the code. If you use sensible tools you get all that, and its much easier to work with the upstream,
The fact is that Mark to Market generates false valuations, especially in volatile markets, which is what we clearly have and is (a) easily manipulated (b) daft. It is a BAD rule that only Americans would defend, and it needs to go, not to nothing but to a properly regulated "fair value" rule.
But it is not only MTM, but also the toleration of Naked Shorts, the abolition of the uptick rule that has set the stage for unprecedented volatility whereby the same people will extract fees and profits from the bailout money.
You need some regulators who arn't dishonest dumb or blind and prosecutors with a robust view of fraud and misrepresentation. That way walking dead banks can contribute to the infrastructure programs by having their management in road-gangs.
Believe me, the rest of the world, having been nagegged and bullied by the US have had enough so dont think that the EU and China wont insist that Obama establish a much more transparent system.
Only you (American) idiots don't realize that you have regulation (eg fu.. cun.) on radio, TV and trying on Internet _BUT_ at the same time have the bigest set of crooks on Wall Street... just Google for Wall Street Fraud, with no regulation whatsoever.
If you did not notice, the rest of the world is pissed, and we are clearly aware that it is not just George W Bush, it is YOU.
Fix it, but remember the world moves on anyway, eg Rome Portugal UK USSR... USA?
On three continents I only need to run M$ W-x, under virtualisation, to work round bugs in IE-y, which BTW has an uncondonably poor ACID compliance.
Windows is insecure junk.
Since I buy my machines from the Taiwan/China factories that supply HP & Dell, and I havn't bought a Linux Box distro in 10 years no install counts in the Gartner/IDG returns, which exist only to mislead end users.
Computers are like the finance market, if you don't know what you are doing you get clipped.
The problem, as so clearly seen in the final days of the Bush administration, is that the government is, in fact, far too powerful.
And makes more mistakes all the time. The histrionic BS about Banking and Automotive bail-outs show this very very clearly.
You need to replace Presidential assent by electronic referendum so bad law, and flawed decisions, are slowed down. The media howls 'do something', and that is almost always wrong.
The present economic crisis is the result of that.
As I watch, for example CNN or read the FT one of the things that most annoys me are the lies that are told by the main-stream media about the financial crisis:
The mad fluctuations in corporate value are a consequence of a US accounting mistake: The mark-to-market rule, which causes the value of real commodities eg gold, houses, oil to evaporate over night, and
The continued tolerance of Naked-short-Selling, and Short-Selling, at all, without the up-tick rule, are the reason for the vast market volatility ie +-5% daily swings. The only purpose of this is to drain tax-payers money as profits and fees, and is why the 'bail-outs' are not working.
The media are either knaves or fools and liers and most likely all three. We urgently need a free, non-corporate press and honest competant regulators in business, the market and scientific endevour.
Sorry, look at the Berkeley paper and you will see we have been significantly behind Moors law since 2004, and the performance increase rate is slowing, and it isnt going to get better.
Intel _had_ to fix their memory bandwidth problem, which strangled their performance, to sell any server chips into the current market, but sadly AMD's hyperchannel dosnt have the bandwidth to cope with (Open)MP interchanges past 32/128 cores.
We need fundamental work on compiler optimization for commonly manufacturable high multicore architectures. This is much more than a kludge to GCC or ICC and Intel dont have a big enough team. This is a classic case for the Bazar!
People building airplanes, managing oil fields, simulating climate, designing fusion reactors and trying to price CDIs based on morgage debt would like solutions today or tomorrow.
The fact that scalable throughput eg map-reduce, gridables gets better dosnt help them, but at least those guys are getting where they want to go.
The __PROBLEM__ with the US Civil System is there is really no downside to filing really dumb ie frivolous and vexatious motions since costs are each party or in cause so people like the RIAA file to play to the main stream media who, puppy like quote the motion but ignore its dismissal.
Since, except that it is filed in court, the actual allegation is defamatory (it is in fact privilidged by reason of its forum) and un-constitutional if it were dismissed with an order for costs in any event, this nonsense would come to an abrupt stop and US civil actions would take half as long. The same can be said about un-preparedness and endless continuances.
Academic and other work has often focused on language design with incremental language facility addition eg C -> C++ -> (java, C#) or Dr.N Worth's Pascal, Modula-x.
This is the exact opposite of what we need, we have C as a given, and we need to make the code generated for parallelisable cases _much_ better.
First, the scope for mono-processors is now strictly limited and we _will_not_ see x2/18 months again, there may be x10 to x100 possible within this basic technology but thats just a few Moor's law cycles; second, the (commercial) problems, as described elsewhere, involve the solution of partial differtial (Heat, Navier Stokes, Elasticity) Equations or Stochastic Simulations --- all of which are inherently clusterable, but not gridable).
It is the cache coherency and memory bandwidth problems with existing architectures that are the problem. We need better low latency data transfer and significant improvement in auto-parallelism technology in compilers.
It should be clear that there has been very little serious investment in basic compiler technology and that is now needed. Academics have realised this but it takes time. The bandwidth issues are solvable else-when with more transistors.
Finally, we have a variety of programming paradigms OO, Functional & procedural and more each of which has a problem niche.
One thing we will certainly have to get away fom is the idea that 'legacy' code can carelessly be re-written in the flavor of month interpreted language eg Java, C#, Perl, Python or Ruby. You can write 95% of your code in a programmet friendly language. But the critical sections need to be in C, FORTRAN or Assembler and need to be very carefully optimized. That can give you x100 on the same architecture.
The fact of the matter is that almost all hardware devices use some kind of firmware to manage design complexity. This includes CPUs as well as many device chips.
This debate is about whether there is any real difference if the firmware is in ROM, (flash-)ROM or RAM, needing re-loading at each re-set from manufacturer provided and developed data which determines the nature of his device. This could easily be a generic cpu+io device with firmware determined personality.
I am sorry, firmware that you must download does not make software less free. If you want a free HARDWARE debate, do that, but dont waste all our time because of mis-understanding.
As with almost all problems where electronic/internet technologies bump into real life issues eg privacy, non-repudiability and simple confidence it is because the Law has not kept up with technology, and that in the USA is the responsibility of the Congress. Writing was thousands of years old, and the printing-press more than 300 years old when the Constitution was adopted in September 17, 1787. The drafters understood the technology.
Today we are blessed with ignorant self serving legislators who do not, and are far too happy to follow hard-case makes bad law hurd thought, eg children, porn, paedophilia, drugs and terrorism. The courts have long held that you can read post-cards, but that if your letter-in-an-envelope is opened then a felony is committed or the information is normally in-admissible.
For this to work people have to start encrypting and signing their e-mails and the Congress and the SCOTUS must enforce identical rules for electronic and hand-written communication.
Specifically you can not go out and discover the entire contents of someone's library and papers in a law suite, and expect to go on a search-engine enabled fishing expedition.
People can be trained to ignore their inner ear, all instrument rated pilots, for example, and you can easily test and train for this ability.
The real problem is that NASA dosn't have a job description of _repair_man_ and tries to use specialists in other areas as practical engineers with funny results. My father, with a PhD in mechanical engineering, would hit his head on the hood and put finger in the fan blades everytime he went to check the oil level.
Assuming you do not need extream accuracy, ie to do General Relativity because you are near the event horizon of a black hole you can yse the very well understood Newton's laws of motion, which should be second nature to a pilot, astronaut or general engineer.
This is it, I have had enough, I can read about 100 times faster than I can make sense of rambeling audio. Whichever marketing-droid thinks unprepared semi-monologues can substitute for well thought out papers needs to fail his MBA, and be drummed out of the community in ignominy.
The only place for this is good lectures, think Hawking, Richard_Feynman and short bio bits, eg I am Linus, this is Tove and thats the dog... 5 seconds
The recent Obama campaign shows that audio-visual clips can be very good, with expensive speach writers, tele-prompters and good producers, this takes lots of preparaton and money but all pointing a cam at an un-prepared speaker results in is sad junk.
Not if they loose the database each week and screw up _both_ the biometric data signing and UID, which given the history of UK government seems most likely.
While a Brit, thank God I live in Switzerland, where the populace is educated, public data secure and FOSS is ever more popular while the Bundesrat can't pass laws the people don't like.
What the USA and UK need is Universal Democracy, and the Internet would allow large populations to get there.
What the democracies also need to is to issue X509 certificates, free, to everyone at birth keeping the key card till children come of age.
Slashdot Mirror
I hate code full of company comment cruft.
Use a good VersionControl system, git, murcurial or Subversion if you must, __but__ keep the meta-data out of the code. If you use sensible tools you get all that, and its much easier to work with the upstream,
In Switzerland too, Obligatory in the mountains in winter.
You must have an American care else you would have an "ABS aus" switch.
What the UK sorely needs is Impeachment to deal with people like Smith.
This story is click-through junk and should not be on slashdot, even on a slow news day
The fact is that Mark to Market generates false valuations, especially in volatile markets, which is what we clearly have and is (a) easily manipulated (b) daft. It is a BAD rule that only Americans would defend, and it needs to go, not to nothing but to a properly regulated "fair value" rule.
But it is not only MTM, but also the toleration of Naked Shorts, the abolition of the uptick rule that has set the stage for unprecedented volatility whereby the same people will extract fees and profits from the bailout money.
You need some regulators who arn't dishonest dumb or blind and prosecutors with a robust view of fraud and misrepresentation. That way walking dead banks can contribute to the infrastructure programs by having their management in road-gangs.
Believe me, the rest of the world, having been nagegged and bullied by the US have had enough so dont think that the EU and China wont insist that Obama establish a much more transparent system.
Only you (American) idiots don't realize that you have regulation (eg fu.. cun.) on radio, TV and trying on Internet _BUT_ at the same time have the bigest set of crooks on Wall Street ... just Google for Wall Street Fraud, with no regulation whatsoever.
... USA?
If you did not notice, the rest of the world is pissed, and we are clearly aware that it is not just George W Bush, it is YOU.
Fix it, but remember the world moves on anyway, eg Rome Portugal UK USSR
On three continents I only need to run M$ W-x, under virtualisation, to work round bugs in IE-y, which BTW has an uncondonably poor ACID compliance.
Windows is insecure junk.
Since I buy my machines from the Taiwan/China factories that supply HP & Dell, and I havn't bought a Linux Box distro in 10 years no install counts in the Gartner/IDG returns, which exist only to mislead end users.
Computers are like the finance market, if you don't know what you are doing you get clipped.
The computer industry seems compelled to repeat the same mistakes, think
Ken Olsen (CEO DEC) Unix is snakeoil
Immer geleich!
Shill, again.
It is the SAME, extended code base.
NONE of the Cr.. has been re-witten, it si easy to pay for puff, but, at last, after many fired CIOs enterprises test.
Work on better chairs!
The problem, as so clearly seen in the final days of the Bush administration, is that the government is, in fact, far too powerful.
And makes more mistakes all the time. The histrionic BS about Banking and Automotive bail-outs show this very very clearly.
You need to replace Presidential assent by electronic referendum so bad law, and flawed decisions, are slowed down. The media howls 'do something', and that is almost always wrong.
The present economic crisis is the result of that.
See Direct Democracy in Switzerland
Does someone want to post a standard US contract of this type. Ie simple!
As I watch, for example CNN or read the FT one of the things that most annoys me are the lies that are told by the main-stream media about the financial crisis:
The mad fluctuations in corporate value are a consequence of a US accounting mistake: The mark-to-market rule, which causes the value of real commodities eg gold, houses, oil to evaporate over night, and
The continued tolerance of Naked-short-Selling, and Short-Selling, at all, without the up-tick rule, are the reason for the vast market volatility ie +-5% daily swings. The only purpose of this is to drain tax-payers money as profits and fees, and is why the 'bail-outs' are not working.
The media are either knaves or fools and liers and most likely all three. We urgently need a free, non-corporate press and honest competant regulators in business, the market and scientific endevour.
Sorry, look at the Berkeley paper and you will see we have been significantly behind Moors law since 2004, and the performance increase rate is slowing, and it isnt going to get better.
Intel _had_ to fix their memory bandwidth problem, which strangled their performance, to sell any server chips into the current market, but sadly AMD's hyperchannel dosnt have the bandwidth to cope with (Open)MP interchanges past 32/128 cores.
We need fundamental work on compiler optimization for commonly manufacturable high multicore architectures. This is much more than a kludge to GCC or ICC and Intel dont have a big enough team. This is a classic case for the Bazar!
People building airplanes, managing oil fields, simulating climate, designing fusion reactors and trying to price CDIs based on morgage debt would like solutions today or tomorrow.
The fact that scalable throughput eg map-reduce, gridables gets better dosnt help them, but at least those guys are getting where they want to go.
The __PROBLEM__ with the US Civil System is there is really no downside to filing really dumb ie frivolous and vexatious motions since costs are each party or in cause so people like the RIAA file to play to the main stream media who, puppy like quote the motion but ignore its dismissal.
Since, except that it is filed in court, the actual allegation is defamatory (it is in fact privilidged by reason of its forum) and un-constitutional if it were dismissed with an order for costs in any event, this nonsense would come to an abrupt stop and US civil actions would take half as long. The same can be said about un-preparedness and endless continuances.
At the cost of replying to myself:
Academic and other work has often focused on language design with incremental language facility addition eg C -> C++ -> (java, C#) or Dr.N Worth's Pascal, Modula-x.
This is the exact opposite of what we need, we have C as a given, and we need to make the code generated for parallelisable cases _much_ better.
First, the scope for mono-processors is now strictly limited and we _will_not_ see x2/18 months again, there may be x10 to x100 possible within this basic technology but thats just a few Moor's law cycles; second, the (commercial) problems, as described elsewhere, involve the solution of partial differtial (Heat, Navier Stokes, Elasticity) Equations or Stochastic Simulations --- all of which are inherently clusterable, but not gridable).
It is the cache coherency and memory bandwidth problems with existing architectures that are the problem. We need better low latency data transfer and significant improvement in auto-parallelism technology in compilers.
It should be clear that there has been very little serious investment in basic compiler technology and that is now needed. Academics have realised this but it takes time. The bandwidth issues are solvable else-when with more transistors.
Finally, we have a variety of programming paradigms OO, Functional & procedural and more each of which has a problem niche.
One thing we will certainly have to get away fom is the idea that 'legacy' code can carelessly be re-written in the flavor of month interpreted language eg Java, C#, Perl, Python or Ruby. You can write 95% of your code in a programmet friendly language. But the critical sections need to be in C, FORTRAN or Assembler and need to be very carefully optimized. That can give you x100 on the same architecture.
The fact of the matter is that almost all hardware devices use some kind of firmware to manage design complexity. This includes CPUs as well as many device chips.
This debate is about whether there is any real difference if the firmware is in ROM, (flash-)ROM or RAM, needing re-loading at each re-set from manufacturer provided and developed data which determines the nature of his device. This could easily be a generic cpu+io device with firmware determined personality.
I am sorry, firmware that you must download does not make software less free. If you want a free HARDWARE debate, do that, but dont waste all our time because of mis-understanding.
As with almost all problems where electronic/internet technologies bump into real life issues eg privacy, non-repudiability and simple confidence it is because the Law has not kept up with technology, and that in the USA is the responsibility of the Congress. Writing was thousands of years old, and the printing-press more than 300 years old when the Constitution was adopted in September 17, 1787. The drafters understood the technology.
Today we are blessed with ignorant self serving legislators who do not, and are far too happy to follow hard-case makes bad law hurd thought, eg children, porn, paedophilia, drugs and terrorism. The courts have long held that you can read post-cards, but that if your letter-in-an-envelope is opened then a felony is committed or the information is normally in-admissible.
For this to work people have to start encrypting and signing their e-mails and the Congress and the SCOTUS must enforce identical rules for electronic and hand-written communication.
Specifically you can not go out and discover the entire contents of someone's library and papers in a law suite, and expect to go on a search-engine enabled fishing expedition.
People can be trained to ignore their inner ear, all instrument rated pilots, for example, and you can easily test and train for this ability.
The real problem is that NASA dosn't have a job description of _repair_man_ and tries to use specialists in other areas as practical engineers with funny results. My father, with a PhD in mechanical engineering, would hit his head on the hood and put finger in the fan blades everytime he went to check the oil level.
If not they should, or not only tool bags could go for a walk! and the main tether should incorporate an an emergency _let_out_ extension.
Assuming you do not need extream accuracy, ie to do General Relativity because you are near the event horizon of a black hole you can yse the very well understood Newton's laws of motion, which should be second nature to a pilot, astronaut or general engineer.
Use a weak, inelastic tie eg string!
A ball of string should be mandatory issue, and be used!
This is it, I have had enough, I can read about 100 times faster than I can make sense of rambeling audio. Whichever marketing-droid thinks unprepared semi-monologues can substitute for well thought out papers needs to fail his MBA, and be drummed out of the community in ignominy.
... 5 seconds
The only place for this is good lectures, think Hawking, Richard_Feynman and short bio bits, eg I am Linus, this is Tove and thats the dog
The recent Obama campaign shows that audio-visual clips can be very good, with expensive speach writers, tele-prompters and good producers, this takes lots of preparaton and money but all pointing a cam at an un-prepared speaker results in is sad junk.
Not if they loose the database each week and screw up _both_ the biometric data signing and
UID, which given the history of UK government seems most likely.
While a Brit, thank God I live in Switzerland, where the populace is educated, public data secure and FOSS is ever more popular while the Bundesrat can't pass laws the people don't like.
What the USA and UK need is Universal Democracy, and the Internet would allow large populations to get there.
What the democracies also need to is to issue X509 certificates, free, to everyone at birth
keeping the key card till children come of age.