I'm aware about the CIA statement about using espionage to monitor foreign companies engaging in corrupt practices to win contracts. OTH, look in certain countries (check out the bottom of the list)that are known for their corrupt practices. Do we see US companies standing back and not paying bribes in the country. Sorry, think again. They are out there leading the pack (they are often financially stonger).
I quote from an unnamed source in Uzbekistan, a US citizen working for a US owned joint venture.
Sure we pay bribes, the (US) govt should get real and exert real pressure at the diplomatic, not the business level.
So why do they really bug software? Just look at the business between Boeing and Airbus. Govt sponsored commercial espionage! Both sides.
The St. Petersburg Times (an English, biweekly paper, published in St. Petersburg, Russia, has an article about winning the the competition.
I quote a student and team coach from the article:
"A large part of their talent was due to very good basic mathematical knowledge given in school," Alexeyev said. "That is one of the strongest points of Russian education. Michigan University, for instance, may have stronger technical facilities, but the smartest guys come from Russia."
What does this mean in reality. Well they very good mathematicians, but unfortunately they learn very little about non-computer related stuff during their education. Their current business education is one acedamic hour per week for just one semester.
Eventually, this will be fixed, (I would guess about five years or so) but not for some time, as education in Russia will not be the fastest to modernise. However, there are some excellent software houses there and they, at least, seem to know something about s/w engineering and running a business.
Oh, yes, I speak about this because I have been working with programmers in St. Pete as well as small and medium sized technology companies. Also, the Mrs, used to help run some educational aid programmes there for Soros.
Please remember, that the ACM thing is about developing a program rather than a system.
I work with people in St. petersburg and can testify how good their coding is. Their week points are the engineering side and project management. I guess that is where the western schools could show their forte.
Without making this into a very different contest, it is difficult to show real software engineering skills though.
An RTOS costs money. Linux doesn't and it has the advantage of being rather more configurable. That means you can use the same basic RT variant of Linux for a number of different jobs.
I have recommended Open Source software quite often when working in poorer economies. For them, the maintenance charges could easily pay for extra programmers so, the cost of support is not an issue.
The only problem is that after saving all those $$$, we have to be sure that the dollars come back to benefit the economy not just somebody's pocket.
I agree, but sometimes they get a bit off-base on the high-tech business scene. Still everyone else was about the.com business and it was just old duffers like Warren Buffet who were crying "Ponzi" and stayed clear. He missed the peak but he definitely didn't get hit by the trough.
The science and technology section is excellent though. I find it better written these days than Scientific American, which seems to have been dumbed down.
I agree, we get lots of bugs in our exciting industry. Usually, it doesn't kill anyone.
How about usability? It is the idea these days to go away from modal dialogues. However try to send something using Outlook and look information up at the same time (say a few contact details). Under some conditions, this will work, on others, Outlook will just refuse to let you out of the current dialogue. Microsoft is just one example though. Some people can single-task, but most of us have to do several jobs at once. The computer doesn't like to let us unless the apps are totally unrelated.
I like the 36 year old "programmer" who listed his skills as c, java, xml, cgi, js, fortran, basic. That sentence is like a giant red flag to any engineer, be they a coder or a manager.
Sorry, I have been working as a contractor. Sure I learned my stuff years ago in Fortran and COBOL so its on my CV but I can also do Java/C/C++. Does it mean that I'm a guru in anyone of these languages? No, but I have no religion and can work in whatever best suits a project. If the language is rusty, then a couple of weeks and I'm back on the ball.
In the end, all we are doing is designing and implementing alorithms, if we are quite good at that then language is not a major issue. yes, I will sprinkle the cool stuff liberally through my CV, because otherwise I get filtered out at first hurdle.
What I am forced to deemphasise on my CV is that I have the ability to manage projects and bring them in on time. There are any number of managers out there who have no real technical background, and they don't like recruiting people that do. So I tell the world that I'm an Analyst/Programmer and I keep pulling the work in.
It has been done elsewhere, but why not price drinks dynamically depending upon demand. When prices go to high, just arrange for a crash after trading stops for a while.
I was in a place like this and it is great fun to 'hack' the market. The only problem was when we came back with a trader who wanted to arrange options on beer!!!!
With a sufficiently large ciomputer and a complete knowledge of quantum mechanics, then the universe is most definitely, deterministic.
As it is, a good old noisy diode, suitably shielded and sampled is a reasonably good source of randomness for practical purposes and remarkably easy to implement in hardware.
Intel even did it as a side function of their 880 chipsets! There are other chips doing the same and not too expensive to add to a router or processor motherboard. Such randomness is invaluable for protocols like TCP and most importantly, cryptographic session keys.
Lets carry on with your use of the automobile as an example. By some lucky chance, we have throttle, brake and clutch, all in a nice little row and completely the same all over the world. Maybe there are better arrangements, but short of automatic transmission, this is a standard.
It means that after learning to drive one auto, the skills to drive another are not wildly different. Personally, I think this is kind of neat.
Look and feel and the wasy of using systems should not be patentable or even copyrightable. The technology behind the one-click or the floating-toolbar may be protected by copyright but never patents which completely monopolise concepts.
If you make a machine that encrypts/decrypts and does electronic signatures then that machine is patentable. It may be that your implementation of PGP is on an ASIC or it might be in a microprocessor controlled by a program. However, you are not patenting PGP, you are patenting a machine that implements PGP.
What is important is that the software component is purely the controller for the patentable application.
Sewers were generally built a long time ago. They didn't exactly have GIS then.
A major issue is that maybe you do have a plan of where the sewers are in relation to eachother but it is difficult to match up with what is on the street that may change, or indeed with other utilitiues diagrams.
The other issue is dimensional instability of paper. Plans are just not very accurate over a period of time because the paper can become distorted.
Law is basically common sense and a certain amount of semantic logic. It doesn't take an expert to grasp the basics. Case-law is where it gets to be very, very complicated and requiring the prodigious feats of memory. The basics of law are somewhat more accessible than say the basics of computing.
The thing is that a lot of technical people like Farber and even to a lesser degree myself (I worked on the regulation of securities markets in Central Asia) can contribute to the policy and regulatory processes.
A lawyer is very specialised though and typically does not understand technology very well. To understand this, just look at some of the stuff on the DMCA, etc.
Perhaps we should have the non technical policymaker preceding their briegfings with I am not a technician or IANATto warn us that they are outside their area of professional competence.
That the legislators are often out of their depth is beyond doubt. If you knopw technology and/emd can communicate, I can only suggest that slashdot readers follow in Farber's footsteps.
We can't have a token Brit like AR on board without having his counterpart, Lister to briung him own to earth. Anyway, it can't be our AR, because this one may have fully qualified to become an officer.
Perhaps this is because you don't need Astrogation to be a weapons technical officer.
I agree. For various reasons, I had to leave SE Linux almost as soon as I tried it - I needed a newer kernel and 2.4.0 was out.
I have used systems with ACL type structures and say that I like it. Fine grained privilege control seems to also catch a lot of software design issues.
If we could get SE Linux or anything else with similar features it would be really good in the server world. I can live with just root on a desktop but it doesn't really help on a server.
Money has definitely leaked out of the Russian Space Program, but think, how much of that is really NASA money. Do we really need to send a full team of US manned spaceflight training specialists to Russia who have more hours in space than any other country?
Please also remember that although the Russians were overbudget and late (which we hear a lot about), wasn't that rather conveniant for NASA who also had contractors delivering late/over-budget?
Oh, I think if you looked at the figures, you would find that Baikonur can put up payloads a lot cheaper than the shuttle can. Unfortunately, we are stuck with the shuttle fleet with a lot of development costs to ammortize over its lifetime.
It was Dave Cutler, designer of RSX-11M and one of the principle architects for VAX/VMS. Great coder, but the Mr. Filesystem and Mr. Security was Andy Goldstein who didn't jump ship with Cutler.
VAX/VMS, despite major parts being written in assmbler was a lot smaller so Cutler and the other architects had mach more say so over the exclusion of cruft.
To give the same example, the protection mechanism on VMS was quite sophisticated, but it didn't even include such things as append-only access. However most security mechanisms were supported, between that and the access modes.
The problem was simply that the ACL Editor couldn't hack-it. However there were third party tools that could work out who could access what and also to fix the poroblem of dangling identifiers. However there was nothing standard.
I like ACLs but they should be well implemented and stuck in the kernel (as part of an SRM), not in a particular files system.
As someone who has attended a KGB presentation (actually, this bit is called the FSB now), they discussed security for companies. They were even advocating the use of lie detectors for key employees. This was quite funny because several defectors have commented that lie detectors are easy to defeat I guess Hanssen was lie-detectored as well during the routine checks.
The thing is, the Russian companies who would get this secret technical information belong effectively still to the state and are not run well. They couldn't use the information if they tried.
The smaller Russian companies with owner-managers who are quite efficient would never, ever see this information. The good technical people wouldn't work for the FSB, because the pay is etrrible and they can make 10 times as much working externally.
Incidentally, the FSB presentation the worst I have ever attended. These are people trained to keep secrets, not to present information. No visual aids, no handouts, just a monotonous monologue as the presenter read from a prepared script. The other funny thing was that it appears that the Russians have perfected cloning as the three FSB persons and the former FSB person who looked and acted similar!
Linux is very important because it links a lot of other free stuff together. Lets go back and look at the beginning though.
Organisations like DECUS (Digital Equipment Corp User Society) started a library and collecting and distributed software. Other such initiatives existed too from other vendors, but because of DEC's position in the educational and research sectors, theirs was one of the more extensive libraries.
Along came a number of companies who gloried in all this code in source form and produced commercial products based upon it. A lot of freeware authors started thinking twice about this.
ALong comes Stallman with the GPL, an editor and a C compiler. Many vendors by this stage were charging serious money for 'development systems'. GCC kept getting better and even companies with money were choosing by preference. However the effect on the PC world was minimal.
Linux is a bit of a johnny-come-lately. He got his kernel up on a PC and gave it away with a bunch of utilities. You no longer needed any licensed product to use a PC. Revolutionary, yes but what came earlier shouldn't be forgotten.
Maybe I'm stupid, but this device needs to be read. The data that comes out can be recorded in whatever form you want, ignoring the content protection.
Why am I exasperated? Well we have seen the same claims again and again. Unless we have a series of tamperproof blackboxes with a fully encrypted I/O (perhaps even with a time code to prevent replay of the encrypted stream) between the storage media and the D/A converter, the content can be copied digitally by anyone with access to the media.
In may stop my son from exchanging stuff with his friends but it will do absolutely nothing to prevent mass piracy.
It depends on whether you really mean a microprocessor based CPU. We were running VAXclusters from the early eighties. This was around VAX 11/750s and 11/780s. Shared file-system, good distributed lock manager and all that stuff.
The earliest clusters had some problems but by the late eighties the only thing stopping a VMScluster (as they later became known, to include the Alpha) was Digital's appallingly stupid management.
Slashdot Mirror
I quote from an unnamed source in Uzbekistan, a US citizen working for a US owned joint venture.
So why do they really bug software? Just look at the business between Boeing and Airbus. Govt sponsored commercial espionage! Both sides.I quote a student and team coach from the article:
What does this mean in reality. Well they very good mathematicians, but unfortunately they learn very little about non-computer related stuff during their education. Their current business education is one acedamic hour per week for just one semester.
Eventually, this will be fixed, (I would guess about five years or so) but not for some time, as education in Russia will not be the fastest to modernise. However, there are some excellent software houses there and they, at least, seem to know something about s/w engineering and running a business.
Oh, yes, I speak about this because I have been working with programmers in St. Pete as well as small and medium sized technology companies. Also, the Mrs, used to help run some educational aid programmes there for Soros.
I work with people in St. petersburg and can testify how good their coding is. Their week points are the engineering side and project management. I guess that is where the western schools could show their forte.
Without making this into a very different contest, it is difficult to show real software engineering skills though.
Also more people know it.
The only problem is that after saving all those $$$, we have to be sure that the dollars come back to benefit the economy not just somebody's pocket.
The science and technology section is excellent though. I find it better written these days than Scientific American, which seems to have been dumbed down.
How about usability? It is the idea these days to go away from modal dialogues. However try to send something using Outlook and look information up at the same time (say a few contact details). Under some conditions, this will work, on others, Outlook will just refuse to let you out of the current dialogue. Microsoft is just one example though. Some people can single-task, but most of us have to do several jobs at once. The computer doesn't like to let us unless the apps are totally unrelated.
Sorry, I have been working as a contractor. Sure I learned my stuff years ago in Fortran and COBOL so its on my CV but I can also do Java/C/C++. Does it mean that I'm a guru in anyone of these languages? No, but I have no religion and can work in whatever best suits a project. If the language is rusty, then a couple of weeks and I'm back on the ball.
In the end, all we are doing is designing and implementing alorithms, if we are quite good at that then language is not a major issue. yes, I will sprinkle the cool stuff liberally through my CV, because otherwise I get filtered out at first hurdle.
What I am forced to deemphasise on my CV is that I have the ability to manage projects and bring them in on time. There are any number of managers out there who have no real technical background, and they don't like recruiting people that do. So I tell the world that I'm an Analyst/Programmer and I keep pulling the work in.
I was in a place like this and it is great fun to 'hack' the market. The only problem was when we came back with a trader who wanted to arrange options on beer!!!!
As it is, a good old noisy diode, suitably shielded and sampled is a reasonably good source of randomness for practical purposes and remarkably easy to implement in hardware.
Intel even did it as a side function of their 880 chipsets! There are other chips doing the same and not too expensive to add to a router or processor motherboard. Such randomness is invaluable for protocols like TCP and most importantly, cryptographic session keys.
It means that after learning to drive one auto, the skills to drive another are not wildly different. Personally, I think this is kind of neat.
Look and feel and the wasy of using systems should not be patentable or even copyrightable. The technology behind the one-click or the floating-toolbar may be protected by copyright but never patents which completely monopolise concepts.
What is important is that the software component is purely the controller for the patentable application.
Partially yes, if this is a repeated offence. The bar owner can and does lose his licence if a a bar becomes known for easy availibility of drugs.
A major issue is that maybe you do have a plan of where the sewers are in relation to eachother but it is difficult to match up with what is on the street that may change, or indeed with other utilitiues diagrams.
The other issue is dimensional instability of paper. Plans are just not very accurate over a period of time because the paper can become distorted.
The thing is that a lot of technical people like Farber and even to a lesser degree myself (I worked on the regulation of securities markets in Central Asia) can contribute to the policy and regulatory processes.
A lawyer is very specialised though and typically does not understand technology very well. To understand this, just look at some of the stuff on the DMCA, etc.
Perhaps we should have the non technical policymaker preceding their briegfings with I am not a technician or IANATto warn us that they are outside their area of professional competence.
That the legislators are often out of their depth is beyond doubt. If you knopw technology and/emd can communicate, I can only suggest that slashdot readers follow in Farber's footsteps.
Perhaps this is because you don't need Astrogation to be a weapons technical officer.
I have used systems with ACL type structures and say that I like it. Fine grained privilege control seems to also catch a lot of software design issues.
If we could get SE Linux or anything else with similar features it would be really good in the server world. I can live with just root on a desktop but it doesn't really help on a server.
Please also remember that although the Russians were overbudget and late (which we hear a lot about), wasn't that rather conveniant for NASA who also had contractors delivering late/over-budget?
Oh, I think if you looked at the figures, you would find that Baikonur can put up payloads a lot cheaper than the shuttle can. Unfortunately, we are stuck with the shuttle fleet with a lot of development costs to ammortize over its lifetime.
Lawyers
Apart from their propensity to bill by the minute, I can't see how this will improve program management.
Of course it could be a wonderful way to kill off the space station by having dozens of lawyers arguing (and billing).
VAX/VMS, despite major parts being written in assmbler was a lot smaller so Cutler and the other architects had mach more say so over the exclusion of cruft.
To give the same example, the protection mechanism on VMS was quite sophisticated, but it didn't even include such things as append-only access. However most security mechanisms were supported, between that and the access modes.
The problem was simply that the ACL Editor couldn't hack-it. However there were third party tools that could work out who could access what and also to fix the poroblem of dangling identifiers. However there was nothing standard.
I like ACLs but they should be well implemented and stuck in the kernel (as part of an SRM), not in a particular files system.
The thing is, the Russian companies who would get this secret technical information belong effectively still to the state and are not run well. They couldn't use the information if they tried.
The smaller Russian companies with owner-managers who are quite efficient would never, ever see this information. The good technical people wouldn't work for the FSB, because the pay is etrrible and they can make 10 times as much working externally.
Incidentally, the FSB presentation the worst I have ever attended. These are people trained to keep secrets, not to present information. No visual aids, no handouts, just a monotonous monologue as the presenter read from a prepared script. The other funny thing was that it appears that the Russians have perfected cloning as the three FSB persons and the former FSB person who looked and acted similar!
Organisations like DECUS (Digital Equipment Corp User Society) started a library and collecting and distributed software. Other such initiatives existed too from other vendors, but because of DEC's position in the educational and research sectors, theirs was one of the more extensive libraries.
Along came a number of companies who gloried in all this code in source form and produced commercial products based upon it. A lot of freeware authors started thinking twice about this.
ALong comes Stallman with the GPL, an editor and a C compiler. Many vendors by this stage were charging serious money for 'development systems'. GCC kept getting better and even companies with money were choosing by preference. However the effect on the PC world was minimal.
Linux is a bit of a johnny-come-lately. He got his kernel up on a PC and gave it away with a bunch of utilities. You no longer needed any licensed product to use a PC. Revolutionary, yes but what came earlier shouldn't be forgotten.
Why am I exasperated? Well we have seen the same claims again and again. Unless we have a series of tamperproof blackboxes with a fully encrypted I/O (perhaps even with a time code to prevent replay of the encrypted stream) between the storage media and the D/A converter, the content can be copied digitally by anyone with access to the media.
In may stop my son from exchanging stuff with his friends but it will do absolutely nothing to prevent mass piracy.
The earliest clusters had some problems but by the late eighties the only thing stopping a VMScluster (as they later became known, to include the Alpha) was Digital's appallingly stupid management.