The idea of sixth form is to allow you to get a feel for university level subjects without having to take the risk of accepting a university offer that you later regret accepting. That way, if you enjoyed the subjects, and got good exam results, you could skip the first year of university.
I think I disagree with you on one fundamental point here: you seem to imply that the level of material taught in an A-level is equivalent to what is taught in the first year of an undergraduate degree course. While I agree that a lot of the focus at sixth-form level is preparation for university courses, the level is completely different.
For example, while A-level maths covers several of the same subjects as first-year university work, the university level presentation will tend to be significantly more generalised and open-ended. The first year of my undergraduate maths degree was mostly spent "levelling the playing field" since many people had come in with different levels of experience in different fields within mathematics. However, even those who'd studied an unusually large amount of, say, mechanics or statistics at A-level hadn't covered everything taught in the first year courses at university.
To give a common contrast, at A-level you might learn about matrices and simple transformations in the Euclidean plane. Covering the same material at undergraduate level, you would be encouraged to think of those transformations as linear maps over vector spaces. The matrix techniques are still used for computation, of course, but the more advanced work is all based on the properties of those linear maps and vector spaces.
No one's stopping you from taking the stuff or reading about it on your own. But to require every student to get a full liberal arts degree in order to get a job in a specialized area is not simply inefficient, it's idiotic.
I guess that depends on what you think the goal of an education system should be. Is it about more than turning everyone into a productive worker? I think so.
I live in the UK, so my education was fairly general in secondary school (aged 11-16), more specialised in the sixth form (16-18) and then entirely specialised at university (18-21 in my case).
During the secondary stage, I studied not only the fields of maths and science in which I'd later choose to specialise, but also history, English, modern languages, Latin (actually one of the most useful classes I took, notwithstanding the subject matter being "a little outdated"), art, craft, music, and more. This gave me a level of general background knowledge about the world, and an appreciation of what my peers were studying later on. I've found speaking several languages to at least a basic conversational level useful on any number of occasions since, so it's hardly redundant knowledge, either.
During the sixth form, I started focussing on maths, physics and chemistry. This level is the hardest to categorise in the UK; much of the material is beyond what an everyday person would need to know of, say, maths, and the focus is more on preparation for studying a related degree than anything else. It's interesting in its own right as well, of course, even if I never use the knowledge of chemistry I gained there in a job.
Once I got to university, I specialised in maths, and later CS. This was obviously very academic, yet is directly relevant to my chosen profession. Even then, though, it's important to separate this academic training from vocational training. A university course shouldn't be teaching specific tools and today's buzzword techniques, it should be teaching (a) the general knowledge needed to appreciate those tools and techniques, and (b) how to study independently, so you can learn the details of specific areas by yourself later.
It's often said around here that a good programmer can learn a new programming language in a few days, and there's at least an element of truth in that. More importantly, in ten years' time, someone with a good background in the theory and the drive and ability to study independently will still be keeping up with new tools and new buzzwords, while the Java McDegree holders will be wondering what this new language is for, and waiting to be spoon fed over-priced training materials by the commercial entities behind it.
Given what the IT industry has been doing to CS courses in recent years, essentially reducing them to vocational qualifications in buzzword subjects, they are clearly interested in propagating the use of newbie programmers at cheap rates for a few years, then trading them in (firing them) when they get too expense and hiring more cheap newbies instead. From a business perspective, this makes for a pretty good "software construction line", but you're losing the essential higher level of quality, both by neglecting proper training and by giving up your more experienced assets. Ultimately, that sort of behaviour leads to inefficient development processes (one skilled and experienced developer can easily be more productive than three newbies who each cost a quarter as much) and loss of quality (witness the declining performance and security of many modern software projects for obvious examples).
So thanks Bill, but I'd rather you didn't try to convert the rest of secondary education into vocational training from age 10. The education system is there to develop people as human beings and cultivate their skills and interests. There will be plenty of time to learn job skills on the job; save the education system for more important things.
At the moment I live in a city, but a lot of the people I know are in shooting clubs. So technically, what I said isn't entirely true. I do come from out in the sticks originally, where most people *do* have various rifles and shotguns, for pest control, hunting and defence.
I know a few people who used to shoot for sport, too, and they basically had to give up post-Dunblane because the new regulations got in the way. Obviously removing the guns stored securely at the gun club and used by members who were law-abiding members of society has done a huge amount to curb gun crime.;-)
One man with a handgun = someone better able to defend himself against a larger opponent attacking him.
You watch too much television.
No, I've spent most of my life training in various aspects of martial arts/combat sports/self defence. Aside from my own experience of the field as an amateur, I've also met a lot of interesting characters along the way, including military types, police firearms officers, professional security guys (bodyguards), and people who have fired their gun in self-defence in other countries where it is legal to do so.
Now, aside from my irrelevant television-watching habits, are you seriously telling me that someone with a weapon isn't better able to defend themselves against a larger opponent than someone without?
Well, your government hasn't tried particularly hard to remove the civil rights of Americans, but seems to have been completely unopposed. [...] OTOH, the UK government has been trying *really* hard to restrict our civil rights, and look how far that's got them, laughable bunch of fools that they are.
You do realise that I live in the UK, right?
Anyway, our government is in the process of ramming both detention by political will and ID cards/backing database through parliament, despite massive public opposition. We're also the most spied upon nation in the world, although neither CCTV nor speed cameras have an overall beneficial effect on crime levels unless used in appropriate locations. Those are just the first examples that come to mind.
Look how far that's got them, you say? We're 90% of the way down the slope to a police state, and most people haven't even noticed.
Is there actually a method of directly using these definitions?
Where you actually need to use them directly, sure.
To give a real world example of how the standards work in practice... I used to write software for a company in the metrology (high precision measurement) business. They made machines that are used, for example, in quality control at the end of production lines. The gauges on the most popular machines gave accurate readings with resolutions of say 1-10m.
Those machines were calibrated from reference artifacts. These were themselves checked for accuracy on still higher precision equipment. (How they actually manufacture something so close to physical perfection is an interesting area in itself...)
Ultimately, there were white room areas with very careful decontamination procedures in place that were used almost exclusively for calibrating the company's most precise equipment and checking their reference artifacts.
From there, you were one step removed from the national standards laboratories. At that level the formal scientific definitions are just fine.
In other words, you work from major standards labs that can use the precise definitions effectively, and propagate the information (with some less, but little enough to be acceptable for the application in question) to more widely distributed testing facilities. A more trendy application of the same basic idea is the use of Internet-based real time clock services.
Keep in mind that quicksort will always have a worst case performance of O(n^2); however, on average it runs about 10% faster than the fastest possible heapsort and mergesort algorithms, which are both O(log n). The main reason for using quicksort is the fact that it's 10% faster *on average* than the pure O(log n) algorithms.
I'd imagine quicksort is probably also rather more correct than any O(log n) sorting algorithm...
Please, please go and read the more informed posts about NULL above. In fact, please take the scarily large number of other posters in this thread who are horribly ill-informed with you.
NULL certainly isn't a void* in C++, and indeed it would be illegal to define it as such since this would prohibit many common pointer conversions.
NULL is a macro, and could never be typedef'd since it's not a type.
There is no requirement for NULL to be 32 bits normally or otherwise.
Your opinion about using NULL vs. using 0 is entirely subjective, and since NULL is typically #defined to be 0 anyway, your argument about letting the compiler do type conversions for you has no merit.
In the general case (as opposed to the specific NULL case where you can just omit the explicit comparison entirely) most modern compilers will issue a warning message if you use assignment for equality where it is suspicious to do so (e.g., in a test condition for a control statement). Since the defensive coding style is also hideously hard to read even if you're fairly used to it, and is now mostly redundant, it's gone the way of most good ideas that have had their time.
A more-concise, and potentially more clear solution would be: for_each(foo.begin(), foo.end(),...).
Which would be lovely, if C++ also supported first order functions so that the STL algorithms requiring functors could be written in a useful way. As it is, you have to mess around creating small helper functions well away from where they are used, or worse start involving function objects, to achieve what should be a trivial effect. The closest you can get on an enlightened project using a modern compiler is one of the lambda-style libraries, and even those are a long way short of the basics of a serious functional programming language.
I think C++ has a lot of practical merits, and I'm all for using the right tool for the right job, I just find that a simple for-loop is almost invariably a cleaner solution than any standard library algorithm that requires a functor. for_each is a particularly bad example, because even the literature isn't clear on what it's intended to do, and whether things like mutating effects are allowed (and will continue to be so in future).
My rule is if the programmer can't figure it out, they shouldn't be working on my code.
I've spent roughly a year working on my current assignment. The first six weeks of that were developing the feature that motivated the work; that was almost exactly what we estimated. The next ten months were fixing all the regressions, essentially clearing up the mess made by people like you who left special cases and witty comments like your example lying all over the code. I have figured it out, and in the process I've fixed a very large number of bugs, but if the original developers had done a competent job in the first place I wouldn't have had to waste nearly a year cleaning up after them.
It's perfectly valid in that it will compile without requiring a diagnostic, but in any language I program that would parse it, it yields undefined behaviour. If you run that code and the object foo points to really is constant, that code is allowed to e-mail those dodgy pictures of your girlfriend to a lads' mag, reformat your hard drive, and then nuke the computer to destroy the evidence.
Yet it still dates from the mid-90s, is two (soon to be three) major revisions behind the times, and was mostly written before the C++ standard was even finished. Optimisation technology has improved a little in the intervening decade or so, y'know!
I live in the UK. I don't personally own a gun, but most of the people I know do.
There aren't many ways that could happen legally, unless perhaps you live in the country and you're talking about neighbours with shotguns for pest control?
WTF do you need a handgun for *anyway*? They're shit.
One man with a handgun = someone better able to defend himself against a larger opponent attacking him.
15 million men with handguns = a government who can't forcibly remove the civil rights of the people no matter how hard it tries.
Which is more useful is left as an exercise to the reader, but in today's climate either has merit.
This is already the case. There is some strange misconception among USians that people in the UK are not allowed guns.
That's because you're wrong and we're not, as various friends and colleagues of mine who used to shoot for sport before the post-Dunblane mess can testify. We're also not allowed to carry much else that would be useful as a weapon without restriction under OWA and friends.
That's just because people in the UK love to steal! [...] I guess that's what happens when your country has restrictive gun ownership laws.
Wow, that's the most impressive non sequitur I've seen on Slashdot in days!
Personally, I'm coming round to the view that the UK should legalise the ownership/carrying/use of some weapons, including firearms, for any member of society in good standing who demonstrates competence in their storage, handling and use. However, my reasons for doing so do not include using the threat of shooting Little Johnny as he walks out of the sweet shop forgetting to pay for his Mars bar to deter the crimewave you seem to think we're experiencing.
I read several articles about how theft is extremely common in the UK.
I read somewhere that you shouldn't confuse correlation with causality while trying to make a logical argument.:-)
Perhaps this is why it's less common to see this in the UK; over here there are also certain reasons for which a non-police officer can lawfully arrest you, but on the flip-side you are entitled to use reasonable force to resist a wrongful arrest. Depending on the circumstances, that could involve anything up to taking the guy out... permanently.:-/
This is probably why UK store detectives (generally a pretty professional bunch, from what I've seen) tend to steer well clear of the one that got away, and stick to pulling guys they've just watched making a poor effort to conceal nicking something that's found on their person within moments of leaving the store?
Nothing in this post is legal advice, and I'm not a lawyer anyway, but hell, don't let that stop you exercising your legal rights as entertainingly as you see fit if anyone ever tries this on you. }:-)
Personally I'm not sure I could live in a world without eBay.
Now that is a scary thought.
No, not a world without eBay, but a world where someone thinks they can't live without eBay. You really need to switch off your broadband connection for a week, and remember that most of the good things in life do not occur in cyberspace.
As you should have noticed by now, XPSP2 is not just a security patch, it has performance improvements and assorted non-security bug fixes and it's not all that surprising that a workaround for a poorly implemented or configured BIOS should fail in a new version.
Perhaps, but in that case, the argument that everyone should install it to get the security updates is rather flawed, no? Microsoft would have been more responsible to release the security updates independently, and minimise the risk to working systems from additional changes.
First: I have an IBM laptop with XPSP2 on it. The problem is drivers, and/or BIOS. Have you updated your BIOS yet?
Updating a BIOS is an inherently dangerous thing to do, and on many systems can easily render the system permanently useless. If a "security patch" requires updating the BIOS to work, I'd bet a large sum of money the problem is the security patch.
I run a pretty busy PC shop and every customers machine that comes in gets SP2 and a full Windowsupdate whether they ask for it or not.
In which case you desperately need a lawyer, a PR guy, and a new line of work, probably in that order.
I've heard loads of technically competent people say they've installed SP2/latest updates/etc. and not had a problem. For each two of them, I've heard at least one equally technically competent person tell me about at least a major software failure, and frequently the whole system being rendered inoperable and requiring a reinstallation. Just last week, my whole company backed out an official MS patch for WinXP that came in through automatic updates after we traced the sudden breaking of our Samba servers to that patch.
A slightly more secure system that doesn't run the tools I need to run is not an improvement over a slightly less secure system that runs the tools I need to run. Please understand this.
Slashdot Mirror
I think I disagree with you on one fundamental point here: you seem to imply that the level of material taught in an A-level is equivalent to what is taught in the first year of an undergraduate degree course. While I agree that a lot of the focus at sixth-form level is preparation for university courses, the level is completely different.
For example, while A-level maths covers several of the same subjects as first-year university work, the university level presentation will tend to be significantly more generalised and open-ended. The first year of my undergraduate maths degree was mostly spent "levelling the playing field" since many people had come in with different levels of experience in different fields within mathematics. However, even those who'd studied an unusually large amount of, say, mechanics or statistics at A-level hadn't covered everything taught in the first year courses at university.
To give a common contrast, at A-level you might learn about matrices and simple transformations in the Euclidean plane. Covering the same material at undergraduate level, you would be encouraged to think of those transformations as linear maps over vector spaces. The matrix techniques are still used for computation, of course, but the more advanced work is all based on the properties of those linear maps and vector spaces.
No, but if you can just wait a second while I check the random number generator...
Yep, it's a number between 0 and INT_MAX, so Slashdot's definition gonna dupe this one. See, random numbers can predict the future!
I guess that depends on what you think the goal of an education system should be. Is it about more than turning everyone into a productive worker? I think so.
I live in the UK, so my education was fairly general in secondary school (aged 11-16), more specialised in the sixth form (16-18) and then entirely specialised at university (18-21 in my case).
During the secondary stage, I studied not only the fields of maths and science in which I'd later choose to specialise, but also history, English, modern languages, Latin (actually one of the most useful classes I took, notwithstanding the subject matter being "a little outdated"), art, craft, music, and more. This gave me a level of general background knowledge about the world, and an appreciation of what my peers were studying later on. I've found speaking several languages to at least a basic conversational level useful on any number of occasions since, so it's hardly redundant knowledge, either.
During the sixth form, I started focussing on maths, physics and chemistry. This level is the hardest to categorise in the UK; much of the material is beyond what an everyday person would need to know of, say, maths, and the focus is more on preparation for studying a related degree than anything else. It's interesting in its own right as well, of course, even if I never use the knowledge of chemistry I gained there in a job.
Once I got to university, I specialised in maths, and later CS. This was obviously very academic, yet is directly relevant to my chosen profession. Even then, though, it's important to separate this academic training from vocational training. A university course shouldn't be teaching specific tools and today's buzzword techniques, it should be teaching (a) the general knowledge needed to appreciate those tools and techniques, and (b) how to study independently, so you can learn the details of specific areas by yourself later.
It's often said around here that a good programmer can learn a new programming language in a few days, and there's at least an element of truth in that. More importantly, in ten years' time, someone with a good background in the theory and the drive and ability to study independently will still be keeping up with new tools and new buzzwords, while the Java McDegree holders will be wondering what this new language is for, and waiting to be spoon fed over-priced training materials by the commercial entities behind it.
Given what the IT industry has been doing to CS courses in recent years, essentially reducing them to vocational qualifications in buzzword subjects, they are clearly interested in propagating the use of newbie programmers at cheap rates for a few years, then trading them in (firing them) when they get too expense and hiring more cheap newbies instead. From a business perspective, this makes for a pretty good "software construction line", but you're losing the essential higher level of quality, both by neglecting proper training and by giving up your more experienced assets. Ultimately, that sort of behaviour leads to inefficient development processes (one skilled and experienced developer can easily be more productive than three newbies who each cost a quarter as much) and loss of quality (witness the declining performance and security of many modern software projects for obvious examples).
So thanks Bill, but I'd rather you didn't try to convert the rest of secondary education into vocational training from age 10. The education system is there to develop people as human beings and cultivate their skills and interests. There will be plenty of time to learn job skills on the job; save the education system for more important things.
I know a few people who used to shoot for sport, too, and they basically had to give up post-Dunblane because the new regulations got in the way. Obviously removing the guns stored securely at the gun club and used by members who were law-abiding members of society has done a huge amount to curb gun crime. ;-)
No, I've spent most of my life training in various aspects of martial arts/combat sports/self defence. Aside from my own experience of the field as an amateur, I've also met a lot of interesting characters along the way, including military types, police firearms officers, professional security guys (bodyguards), and people who have fired their gun in self-defence in other countries where it is legal to do so.
Now, aside from my irrelevant television-watching habits, are you seriously telling me that someone with a weapon isn't better able to defend themselves against a larger opponent than someone without?
You do realise that I live in the UK, right?
Anyway, our government is in the process of ramming both detention by political will and ID cards/backing database through parliament, despite massive public opposition. We're also the most spied upon nation in the world, although neither CCTV nor speed cameras have an overall beneficial effect on crime levels unless used in appropriate locations. Those are just the first examples that come to mind.
Look how far that's got them, you say? We're 90% of the way down the slope to a police state, and most people haven't even noticed.
We've been trying to tell you young upstarts the same thing ever since you forgot how to use the letter "u". :-)
Looks like we lost a mu in there somewhere: the resolutions for the popular machines were around 1-10 micrometres.
Where you actually need to use them directly, sure.
To give a real world example of how the standards work in practice... I used to write software for a company in the metrology (high precision measurement) business. They made machines that are used, for example, in quality control at the end of production lines. The gauges on the most popular machines gave accurate readings with resolutions of say 1-10m.
Those machines were calibrated from reference artifacts. These were themselves checked for accuracy on still higher precision equipment. (How they actually manufacture something so close to physical perfection is an interesting area in itself...)
Ultimately, there were white room areas with very careful decontamination procedures in place that were used almost exclusively for calibrating the company's most precise equipment and checking their reference artifacts.
From there, you were one step removed from the national standards laboratories. At that level the formal scientific definitions are just fine.
In other words, you work from major standards labs that can use the precise definitions effectively, and propagate the information (with some less, but little enough to be acceptable for the application in question) to more widely distributed testing facilities. A more trendy application of the same basic idea is the use of Internet-based real time clock services.
I'd imagine quicksort is probably also rather more correct than any O(log n) sorting algorithm...
Please, please go and read the more informed posts about NULL above. In fact, please take the scarily large number of other posters in this thread who are horribly ill-informed with you.
YHBT. HAND.
In the general case (as opposed to the specific NULL case where you can just omit the explicit comparison entirely) most modern compilers will issue a warning message if you use assignment for equality where it is suspicious to do so (e.g., in a test condition for a control statement). Since the defensive coding style is also hideously hard to read even if you're fairly used to it, and is now mostly redundant, it's gone the way of most good ideas that have had their time.
Which would be lovely, if C++ also supported first order functions so that the STL algorithms requiring functors could be written in a useful way. As it is, you have to mess around creating small helper functions well away from where they are used, or worse start involving function objects, to achieve what should be a trivial effect. The closest you can get on an enlightened project using a modern compiler is one of the lambda-style libraries, and even those are a long way short of the basics of a serious functional programming language.
I think C++ has a lot of practical merits, and I'm all for using the right tool for the right job, I just find that a simple for-loop is almost invariably a cleaner solution than any standard library algorithm that requires a functor. for_each is a particularly bad example, because even the literature isn't clear on what it's intended to do, and whether things like mutating effects are allowed (and will continue to be so in future).
I've spent roughly a year working on my current assignment. The first six weeks of that were developing the feature that motivated the work; that was almost exactly what we estimated. The next ten months were fixing all the regressions, essentially clearing up the mess made by people like you who left special cases and witty comments like your example lying all over the code. I have figured it out, and in the process I've fixed a very large number of bugs, but if the original developers had done a competent job in the first place I wouldn't have had to waste nearly a year cleaning up after them.
It's perfectly valid in that it will compile without requiring a diagnostic, but in any language I program that would parse it, it yields undefined behaviour. If you run that code and the object foo points to really is constant, that code is allowed to e-mail those dodgy pictures of your girlfriend to a lads' mag, reformat your hard drive, and then nuke the computer to destroy the evidence.
Yet it still dates from the mid-90s, is two (soon to be three) major revisions behind the times, and was mostly written before the C++ standard was even finished. Optimisation technology has improved a little in the intervening decade or so, y'know!
There aren't many ways that could happen legally, unless perhaps you live in the country and you're talking about neighbours with shotguns for pest control?
One man with a handgun = someone better able to defend himself against a larger opponent attacking him.
15 million men with handguns = a government who can't forcibly remove the civil rights of the people no matter how hard it tries.
Which is more useful is left as an exercise to the reader, but in today's climate either has merit.
That's because you're wrong and we're not, as various friends and colleagues of mine who used to shoot for sport before the post-Dunblane mess can testify. We're also not allowed to carry much else that would be useful as a weapon without restriction under OWA and friends.
Wow, that's the most impressive non sequitur I've seen on Slashdot in days!
Personally, I'm coming round to the view that the UK should legalise the ownership/carrying/use of some weapons, including firearms, for any member of society in good standing who demonstrates competence in their storage, handling and use. However, my reasons for doing so do not include using the threat of shooting Little Johnny as he walks out of the sweet shop forgetting to pay for his Mars bar to deter the crimewave you seem to think we're experiencing.
I read somewhere that you shouldn't confuse correlation with causality while trying to make a logical argument. :-)
Perhaps this is why it's less common to see this in the UK; over here there are also certain reasons for which a non-police officer can lawfully arrest you, but on the flip-side you are entitled to use reasonable force to resist a wrongful arrest. Depending on the circumstances, that could involve anything up to taking the guy out... permanently. :-/
This is probably why UK store detectives (generally a pretty professional bunch, from what I've seen) tend to steer well clear of the one that got away, and stick to pulling guys they've just watched making a poor effort to conceal nicking something that's found on their person within moments of leaving the store?
Nothing in this post is legal advice, and I'm not a lawyer anyway, but hell, don't let that stop you exercising your legal rights as entertainingly as you see fit if anyone ever tries this on you. }:-)
Ah, finally the answer to Slashdot's "only 5 mod points" system -- just add an many +'s as you want in one mod!
This post will now immediately become (Score: 9,765,264, Overrated)...
Is that where you shoot everyone else who might bid, and then just bid the reserve yourself? Worked for me...
Marked at: NOTABUG :-)
Now that is a scary thought.
No, not a world without eBay, but a world where someone thinks they can't live without eBay. You really need to switch off your broadband connection for a week, and remember that most of the good things in life do not occur in cyberspace.
Perhaps, but in that case, the argument that everyone should install it to get the security updates is rather flawed, no? Microsoft would have been more responsible to release the security updates independently, and minimise the risk to working systems from additional changes.
Updating a BIOS is an inherently dangerous thing to do, and on many systems can easily render the system permanently useless. If a "security patch" requires updating the BIOS to work, I'd bet a large sum of money the problem is the security patch.
In which case you desperately need a lawyer, a PR guy, and a new line of work, probably in that order.
I've heard loads of technically competent people say they've installed SP2/latest updates/etc. and not had a problem. For each two of them, I've heard at least one equally technically competent person tell me about at least a major software failure, and frequently the whole system being rendered inoperable and requiring a reinstallation. Just last week, my whole company backed out an official MS patch for WinXP that came in through automatic updates after we traced the sudden breaking of our Samba servers to that patch.
A slightly more secure system that doesn't run the tools I need to run is not an improvement over a slightly less secure system that runs the tools I need to run. Please understand this.