I've actually been working on this off-and-on for a while, I'm hoping we can release some beta code soon. Currently developing it on Linux, but planning to release OSX and Windows versions, too. We're caching reads and writes, and only the blocks that are most frequently used, plus various other SSD-relevant optimisations. The block allocation logic is pretty complex (and I'm too busy with work), which is why it's been taking so long.
You're right, of course. I should have said I don't run an antivirus application for continuous scanning in the background, which is what actually causes the slowdown. I do have a scanner installed and let it do a system scan occasionally. I only actually use Windows once a month or so, and it generally doesn't talk to the outside world much. I'm prepared to take that small risk.
Interesting stuff about the mp3s though, I didn't know there were any mp3 viruses 'in the wild'. (not that I listen to music in Windows, or download MP3s)
Well, permissions will prevent anything running as an unprivileged user from overwriting system files. That means, anything that isn't in \Documents and Settings under Windows, or/home under UNIX. Buffer overruns will only allow privilege escalation if the vulnerable code is running as root, or even happens in the kernel. Buffer overruns in an app running as an unprivieged user will still only be able to do stuff that user can do. Sure, it can probably still run a program that'll send e-mail to replicate, etc. but at least it doesn't infect the whole system. You could log in as a different user and tidy up without having to fear about infecting the rest of the system.
Backups are good. I make backups. But it'd still be a pain in the arse and a waste of time to have to reinstall the OS even if my personal data was safe.
What I continually fail to understand is why everyone I know logs in as an Administrator under Windows, even after falling victim to a virus, spyware, etc. I don't necessarily mean the account with that name, having a personal user in that group amounts to the same thing.
I'm a fulltime Linux user (4 years on the desktop, 7 years otherwise, so no veteran, and no newbie either) and I'd never even consider using logging in as root for any activities that aren't associated with system administration. (guess where "Administrator" comes from) Typing in the root password to install software isn't something I'd call a nuisance or even mildly irritating.
The same thing is of course possible under Windows: Make your main login a 'Power User', or if you feel that's not safe enough, put it in a group with the same policies as the 'Users' group and slowly increase its permissions until you can work productively. (there are problems with debugging code and other niggles by default) Recent versions of Windows will prompt you for an Admin password for stuff your user isn't allowed to touch, although in some cases you have to explicitly right-click the link/executable and select 'run as'. I think there even are some utilities around to make the process even less painful.
If you're doing extensive admin stuff, you can also log in as an Admin explicitly of course, and since XP you can switch between users quite easily without logging out.
It always astounds me how incredibly adverse peoples' reactions are to this suggestion. Sure, it doesn't provide absolute security (ActiveX springs to mind) but that, together with frequent Windows Updates, an enabled WinXP SP2 firewall, and not using IE, I can't imagine you'll have a problem. You might be able to lose some data if you catch a virus, but you're very, very unlikely to bone your system. I do occasionally boot into Windows to play games (Cedega doesn't really work on ATI graphics cards) and I've never caught a virus or spyware, and I don't have an antivirus program installed, as they slow the system down to an infuriating degree IMO.
The problem is, once you've seen their code, you're no longer allowed to re-implement it, as Microsoft surely would be quick to "point out" (i.e. lawsuit) that you've infringed their copyright. So what would need to be done is to have a team of people inspect the code, write up a specification of what it is supposed to do, pass that specification to another team, who can then actually implement it.
This error-prone, time- and resource-consuming and thus won't help much to re-enable fair competition. It's not that much better than doing the same thing with a packet sniffer instead of the source.
Microsoft most likely do have accurate specs, (or at least more accurate than what a bunch of reverse engineers could come up with) they're just trying to get out of, or at least delay publishing them.
Yes, changing the clock speed in Linux is no problem (where supported) and many distros even have automatic throttling and such activated. However, the article and GP are talking about changing the CPU's core voltage. Can Cpufrequtils change that? If not, is there another way of doing it? I suspect a very low-level hardware call needs to be made for this to be done.
If you equate the government with 'a gang of criminals', then you can argue a lot of things, sure. That doesn't automatically make it true though.
But to the best of my knowledge, this force comes from ONLY ONE PLACE: governments. Well, the competitors who have been unfairly forced out of the market by the monopolist can hardly be expected to use force.
If you have a problem with the definition of what is unfair, prompt the governing body to change the law, don't just break it.
So the government "convicting" someone of being a monopolist is a bit like the local thieves gang condemning one of it's own members and "making things right" with the local neighborhood by asking for protection money. No. Thank. You. One of its own members? Last time I checked, Microsoft wasn't a member of the European Parliament. Asking protection money from the local neighbourhood? I'm really confused as to how this is even an analogy at all.
And I don't follow your argumentation of requiring the source code. Documenting all features and how to utilise them should be sufficient, surely? There's the problem of finding out how you know that they really are all features, but there are plenty of ways to check these things: For example, looking at the exported symbols of the libraries, putting a checking layer between two interoperating pieces of Microsoft software and intercepting all calls and making sure that it's all documented. Sure, it's hardly trivial, but Microsoft currently are stalling it completely, rather than causing problems in the details.
Well, if it really came to that, and I doubt it will, there are plenty of companies that would support legacy Windows installations. Of course, alternative operating system vendors would have a field day and offer all sorts of special migration deals. Like in a healthy market that isn't dominated by a monopoly.
You, just as plenty of other posters in this thread, seem to be forgetting a couple of facts: - They are convicted of abusing their monopoly in one market to force their way into other markets, and thus are subject to stricter rules than companies that are not monopolies. - The terms of 'opening up' cover API documentation and guaranteeing interoperability with other proprietary vendors as well as Free Software. NOT opening the source code.
Ergo: this doesn't impact proprietary software at all. Non-monopolists are perfectly allowed not to document their APIs. Microsoft can keep their sacred source code closed.
What they are asked to do is pay up until they open their specs.
While this is probably a troll, I probably should point out that Microsoft are doing trade within the EU. As in, they operate a business here, just as they do in the US. Are you aware that you're suggesting that European businesses operating in the US should be able to break the law and go unpunished? Interesting point of view.
Well, they operate in the EU, or else this wouldn't even be an issue. Enforcement is easy: shut down Microsoft's European offices, ban any Microsoft product or service imports. Essentially stop any money flowing out of Europe into Microsoft's pockets. They'd notice that.
If I'm on holiday in the US and I commit what is a crime there, but not back home, does that mean I'm innocent? No. Same thing with doing business in other countries.
People were downloading music from Napster before DMCA and equivalent copyright law amendments. Your point is what? The law changes, for better or for worse, depending on your point of view. Besides, each of the EU member states had their own anti-trust laws before joining, it just takes the responsibility off each individual country to enforce them, which makes sense for a common trade area.
From what I gathered, "Open Source" was a buzzword to begin with. The various early proponents of it (ESR, Bruce Perens, etc.) came up with it when they figured that "Free Software" would not catch on with PHBs and investors. The term was designed to generate buzz. Before it was associated with Free Software, it didn't have a meaning, and thus took, and still takes on whatever meaning people attribute to it. If that doesn't fit the definition of a buzzword, I don't know what does.
At least this is the way ESR portrays it in one (or more) of his essays, and the naming process is also mentioned in the film RevolutionOS.
Note: I don't agree with most of ESR's views, I'm just attempting to state facts. I personally use the term "Free Software".
It shouldn't be illegal provided that it's operated as a 'tip jar' should it?
You're right, it shouldn't, morally anyway, but you forget that it's usually the labels that hold the rights to the music, not the artists. The artists aren't able to publish free samples on their website, even if they want to, without permission from their label.
Sorry to be pedantic, but the comment syntax for HTML is actually , as it is in XML, and also SGML, since that is the 'big daddy' of both markup languages. SGML and HTML actually have more than just that, but:
The Wikipedia Article on comments gives you a list of the comment syntax of quite a few programming (and markup) languages, so I won't make an inferior attempt at doing so here.
Too far fetched? Probably. There are a few slightly different ways of loading libraries. You have languages that don't compile to native code. Either way, for automatic detection, you'd really have to wait until runtime, unless the binary format had a detailed list of dependencies. Why bloat up the binaries if you can let the packaging system take care of this stuff?
I'm not really sure what you're trying to say with your post. Are you trying to imply that that's what I'm trying to say? Either way, it's not. I'm only suggesting going a step further than specifying what dependencies a package has, and also specify where to find those dependencies, be it in a big repository, or on a random website.
I've thought about this occasionally. I'd agree that a monolithic repository system is not the way to go. It still doesn't solve the problem of being able to just download a piece of software and just being able to install it without problems. Debian fares no better than RPM-based distros in this respect, and installing core packages is no problem on either of the existing distros' architectures.
A solution that seems to make the most sense to me, which nobody seems to have tried yet, is the following:
Don't rely on one big repository (e.g. debian, gentoo, etc.) but also don't make the whole thing file-based like in OSX. Do keep repositories if you want, but in addition to having a bunch of basic repositories, (e.g. Ubuntu vs. Debian Unstable) you also put information not only on what other packages are required, but also how to get those other packages into each package.
For example, this could be done by pointing to a bunch of mirrored URLs that point to some XML data describing the package at that mirror. The installer could pick the most recent version, choose the fastest mirror, whatever.
Additionally, some sort of 'compound packages' would be useful. That way, you can ship rare libraries directly with the application. They may or may not be installed once downloaded, depending if you've already got the same or a newer version of them on your system. This could be especially helpful for systems that don't have internet connectivity. (gasp!)
Sure, it's not perfect, but it beats RPMs (I use SUSE so I experience this myself) and the debian system any day, because you can just go and download packages off the internet and install them, without having to go and hunt for the dependencies yourself. Most likely whoever made the package actually had the necessary libraries installed (and the package system can remember where he got them from!) so all that is needed is to convey that information to the user's system.
The case where it breaks down is of course when all the mirrors eventually die, for example if a package ends up becoming unmaintained. But if it's not been updated for that long, it and its dependencies could be added to the various monolithic repositories. I'm sure organisations would pop up that would keep 'dead' packages around for people to use. The way to combat this would be to have as much redundancy as possible, of course.
I don't know. It might just work better than what we've got at the moment?
While this is quite likely true, university degree programmes these days are focussed on passing exams more than anything else. Just because you're able to pass exams well does not mean you're able to innovate, and, even more importantly, vice versa. Quite frankly, if you're the kind of person who is prone to coming up with original ideas, I doubt you need exams as stimulation. Lectures? Definitely.
I'm aware of that, but try convincing semi-bad programmers to use 'size_t' rather than 'int' when appropriate, or making use of the 'sizeof' operator instead of hardcoding the values. (I don't let people like that anywhere near my source code repositories;)) Besides, if you don't have access to the source, you're still stuck, no matter how easy it would be to port if you did have the source. I think this is the issue that MS (and Sun for that matter) wanted to address with.NET and Java.
Independence of which hardware platform you're running Windows on. I think the.NET project was started when it was still believed that Itanium might make it. Then, there's the whole Issue of porting all those 32-bit x86 apps to the AMD64 architecture. Moving Windows to a new architecture is by no means as easy as it is to move GNU and Linux, or most other Free Software to another architecture. It's a chicken-and-egg problem, and as such, it's much more pronounced on proprietary systems, you have to get all your software vendors to port their software for you. AMD64 would have been stillborn if there wasn't a way of running 32-bit x86 applications on it, much like it was the case for 16-bit applications on the 386 and later.
Oh yes, definitely. Some people have no affinity to this kind of stuff, which isn't really a problem in itself, there are plenty of other fulfilling and interesting things to do that don't require maths. She seems to have found something that interests her. I know intelligent biology students, fresh out of school, who couldn't solve a linear equation if their life depended on it.
The point I was trying to make was that there are people who do have a natural talent for maths and physical sciences, but are not exposed to maths at an early enough stage in their life, when their minds are still much more fertile. This is especially problematic if they're interested in science, or engineering, but aren't especially excited about maths itself, and then are scared off by the amount of maths required to do anything useful.
Actually, I think, speaking as a Computational Physics student, the problem of lack of interest does exist, and lies a lot earlier than when it comes to choosing a subject to study at university.
At least in Europe, there doesn't seem to be enough mathematical education in primary school and early secondary school to allow later science courses at school to even spark interest. For most of the people on my course that I've talked to, university physics is nothing like the physics they did at school, primarily because they completely lacked the mathematical foundations. I'm not saying teach calculus to 10-year olds, but some 18-year olds who have been at school all their life can't multiply fractions, not because of lack of intelligence (there would not be any point recruiting those cases for university anyway, sorry) but simply because they were never taught it. Of course they're not going to want to study a maths-heavy science, because it seems incredibly daunting.
I don't think this is specific to one or two countries, let alone particular political factions, but (at the very least) something that the entire Western society faces. Educational institutions and teachers are slowly doing the jobs that used to be the parents' responsibilities. No wonder they have no time for real teaching.
Increases in investment into science probably would help, I don't disagree with you on that point. However, I think attacking the root of the problem, early education, would be a much more effective use of resources. If there is enough inherent interest, additional funding for science can go into research rather than recruitment.
By the way, I don't claim to have a practical solution to the problem, but I think the problem hasn't even been identified properly yet by the scientific community. As a member of the Institute of Physics, I constantly get information about this and that campaign to raise science awareness among teenagers and young adults. I believe these attempts at solving the problem are fundamentally flawed in their short-sightedness.
Let's face it, science is never going to be 'cool' to the masses. Instead, let's encourage those who have a natural interest in the subject.
Well, if you need an int of a particular size, you need to typedef yourself a compiler/platform specific one anyway; this has always been that way. I believe 'int' used to be 2 bytes on most 16-bit DOS compilers. If you're talking about pointer arithmetic and array indexing, you should have been using size_t for those uses all along.
By the way, the GCC sizes for i386 and x86_64 are: int: 4 bytes, 4 bytes long: 4 bytes, 8 bytes long long: 8 bytes, 16 bytes size_t: 4 bytes, 8 bytes (I believe it's typedef'ed as an unsigned long)
Slashdot Mirror
I've actually been working on this off-and-on for a while, I'm hoping we can release some beta code soon. Currently developing it on Linux, but planning to release OSX and Windows versions, too. We're caching reads and writes, and only the blocks that are most frequently used, plus various other SSD-relevant optimisations. The block allocation logic is pretty complex (and I'm too busy with work), which is why it's been taking so long.
You're right, of course. I should have said I don't run an antivirus application for continuous scanning in the background, which is what actually causes the slowdown. I do have a scanner installed and let it do a system scan occasionally. I only actually use Windows once a month or so, and it generally doesn't talk to the outside world much. I'm prepared to take that small risk.
Interesting stuff about the mp3s though, I didn't know there were any mp3 viruses 'in the wild'. (not that I listen to music in Windows, or download MP3s)
~phil
Well, permissions will prevent anything running as an unprivileged user from overwriting system files. That means, anything that isn't in \Documents and Settings under Windows, or /home under UNIX. Buffer overruns will only allow privilege escalation if the vulnerable code is running as root, or even happens in the kernel. Buffer overruns in an app running as an unprivieged user will still only be able to do stuff that user can do. Sure, it can probably still run a program that'll send e-mail to replicate, etc. but at least it doesn't infect the whole system. You could log in as a different user and tidy up without having to fear about infecting the rest of the system.
Backups are good. I make backups. But it'd still be a pain in the arse and a waste of time to have to reinstall the OS even if my personal data was safe.
~phil
What I continually fail to understand is why everyone I know logs in as an Administrator under Windows, even after falling victim to a virus, spyware, etc. I don't necessarily mean the account with that name, having a personal user in that group amounts to the same thing.
I'm a fulltime Linux user (4 years on the desktop, 7 years otherwise, so no veteran, and no newbie either) and I'd never even consider using logging in as root for any activities that aren't associated with system administration. (guess where "Administrator" comes from) Typing in the root password to install software isn't something I'd call a nuisance or even mildly irritating.
The same thing is of course possible under Windows: Make your main login a 'Power User', or if you feel that's not safe enough, put it in a group with the same policies as the 'Users' group and slowly increase its permissions until you can work productively. (there are problems with debugging code and other niggles by default) Recent versions of Windows will prompt you for an Admin password for stuff your user isn't allowed to touch, although in some cases you have to explicitly right-click the link/executable and select 'run as'. I think there even are some utilities around to make the process even less painful.
If you're doing extensive admin stuff, you can also log in as an Admin explicitly of course, and since XP you can switch between users quite easily without logging out.
It always astounds me how incredibly adverse peoples' reactions are to this suggestion. Sure, it doesn't provide absolute security (ActiveX springs to mind) but that, together with frequent Windows Updates, an enabled WinXP SP2 firewall, and not using IE, I can't imagine you'll have a problem. You might be able to lose some data if you catch a virus, but you're very, very unlikely to bone your system. I do occasionally boot into Windows to play games (Cedega doesn't really work on ATI graphics cards) and I've never caught a virus or spyware, and I don't have an antivirus program installed, as they slow the system down to an infuriating degree IMO.
~phil
The problem is, once you've seen their code, you're no longer allowed to re-implement it, as Microsoft surely would be quick to "point out" (i.e. lawsuit) that you've infringed their copyright. So what would need to be done is to have a team of people inspect the code, write up a specification of what it is supposed to do, pass that specification to another team, who can then actually implement it.
This error-prone, time- and resource-consuming and thus won't help much to re-enable fair competition. It's not that much better than doing the same thing with a packet sniffer instead of the source.
Microsoft most likely do have accurate specs, (or at least more accurate than what a bunch of reverse engineers could come up with) they're just trying to get out of, or at least delay publishing them.
~phil
Nonono, not 'XBox portable'! Surely they'll call it the YBoy!
Now, don't ask me what the hell a ZBoz would be.
~phil
Yes, changing the clock speed in Linux is no problem (where supported) and many distros even have automatic throttling and such activated. However, the article and GP are talking about changing the CPU's core voltage. Can Cpufrequtils change that? If not, is there another way of doing it? I suspect a very low-level hardware call needs to be made for this to be done.
~phil
If you equate the government with 'a gang of criminals', then you can argue a lot of things, sure. That doesn't automatically make it true though.
But to the best of my knowledge, this force comes from ONLY ONE PLACE: governments.
Well, the competitors who have been unfairly forced out of the market by the monopolist can hardly be expected to use force.
If you have a problem with the definition of what is unfair, prompt the governing body to change the law, don't just break it.
So the government "convicting" someone of being a monopolist is a bit like the local thieves gang condemning one of it's own members and "making things right" with the local neighborhood by asking for protection money. No. Thank. You.
One of its own members? Last time I checked, Microsoft wasn't a member of the European Parliament. Asking protection money from the local neighbourhood? I'm really confused as to how this is even an analogy at all.
And I don't follow your argumentation of requiring the source code. Documenting all features and how to utilise them should be sufficient, surely? There's the problem of finding out how you know that they really are all features, but there are plenty of ways to check these things: For example, looking at the exported symbols of the libraries, putting a checking layer between two interoperating pieces of Microsoft software and intercepting all calls and making sure that it's all documented. Sure, it's hardly trivial, but Microsoft currently are stalling it completely, rather than causing problems in the details.
~phil
Well, if it really came to that, and I doubt it will, there are plenty of companies that would support legacy Windows installations. Of course, alternative operating system vendors would have a field day and offer all sorts of special migration deals. Like in a healthy market that isn't dominated by a monopoly.
~phil
You, just as plenty of other posters in this thread, seem to be forgetting a couple of facts:
- They are convicted of abusing their monopoly in one market to force their way into other markets, and thus are subject to stricter rules than companies that are not monopolies.
- The terms of 'opening up' cover API documentation and guaranteeing interoperability with other proprietary vendors as well as Free Software. NOT opening the source code.
Ergo: this doesn't impact proprietary software at all. Non-monopolists are perfectly allowed not to document their APIs. Microsoft can keep their sacred source code closed.
What they are asked to do is pay up until they open their specs.
~phil
While this is probably a troll, I probably should point out that Microsoft are doing trade within the EU. As in, they operate a business here, just as they do in the US. Are you aware that you're suggesting that European businesses operating in the US should be able to break the law and go unpunished? Interesting point of view.
~phil
Well, they operate in the EU, or else this wouldn't even be an issue. Enforcement is easy: shut down Microsoft's European offices, ban any Microsoft product or service imports. Essentially stop any money flowing out of Europe into Microsoft's pockets. They'd notice that.
If I'm on holiday in the US and I commit what is a crime there, but not back home, does that mean I'm innocent? No. Same thing with doing business in other countries.
~phil
People were downloading music from Napster before DMCA and equivalent copyright law amendments. Your point is what? The law changes, for better or for worse, depending on your point of view. Besides, each of the EU member states had their own anti-trust laws before joining, it just takes the responsibility off each individual country to enforce them, which makes sense for a common trade area.
~phil
From what I gathered, "Open Source" was a buzzword to begin with. The various early proponents of it (ESR, Bruce Perens, etc.) came up with it when they figured that "Free Software" would not catch on with PHBs and investors. The term was designed to generate buzz. Before it was associated with Free Software, it didn't have a meaning, and thus took, and still takes on whatever meaning people attribute to it. If that doesn't fit the definition of a buzzword, I don't know what does.
At least this is the way ESR portrays it in one (or more) of his essays, and the naming process is also mentioned in the film RevolutionOS.
Note: I don't agree with most of ESR's views, I'm just attempting to state facts. I personally use the term "Free Software".
~phil
It shouldn't be illegal provided that it's operated as a 'tip jar' should it?
You're right, it shouldn't, morally anyway, but you forget that it's usually the labels that hold the rights to the music, not the artists. The artists aren't able to publish free samples on their website, even if they want to, without permission from their label.
~phil
Google are always recruiting, from what I hear, although I guess that doesn't necessarily apply to execs.
~phil
Sorry to be pedantic, but the comment syntax for HTML is actually , as it is in XML, and also SGML, since that is the 'big daddy' of both markup languages. SGML and HTML actually have more than just that, but:
The Wikipedia Article on comments gives you a list of the comment syntax of quite a few programming (and markup) languages, so I won't make an inferior attempt at doing so here.
~phil
Too far fetched? Probably. There are a few slightly different ways of loading libraries. You have languages that don't compile to native code. Either way, for automatic detection, you'd really have to wait until runtime, unless the binary format had a detailed list of dependencies. Why bloat up the binaries if you can let the packaging system take care of this stuff?
I'm not really sure what you're trying to say with your post. Are you trying to imply that that's what I'm trying to say? Either way, it's not. I'm only suggesting going a step further than specifying what dependencies a package has, and also specify where to find those dependencies, be it in a big repository, or on a random website.
~phil
I've thought about this occasionally. I'd agree that a monolithic repository system is not the way to go. It still doesn't solve the problem of being able to just download a piece of software and just being able to install it without problems. Debian fares no better than RPM-based distros in this respect, and installing core packages is no problem on either of the existing distros' architectures.
A solution that seems to make the most sense to me, which nobody seems to have tried yet, is the following:
Don't rely on one big repository (e.g. debian, gentoo, etc.) but also don't make the whole thing file-based like in OSX. Do keep repositories if you want, but in addition to having a bunch of basic repositories, (e.g. Ubuntu vs. Debian Unstable) you also put information not only on what other packages are required, but also how to get those other packages into each package.
For example, this could be done by pointing to a bunch of mirrored URLs that point to some XML data describing the package at that mirror. The installer could pick the most recent version, choose the fastest mirror, whatever.
Additionally, some sort of 'compound packages' would be useful. That way, you can ship rare libraries directly with the application. They may or may not be installed once downloaded, depending if you've already got the same or a newer version of them on your system. This could be especially helpful for systems that don't have internet connectivity. (gasp!)
Sure, it's not perfect, but it beats RPMs (I use SUSE so I experience this myself) and the debian system any day, because you can just go and download packages off the internet and install them, without having to go and hunt for the dependencies yourself. Most likely whoever made the package actually had the necessary libraries installed (and the package system can remember where he got them from!) so all that is needed is to convey that information to the user's system.
The case where it breaks down is of course when all the mirrors eventually die, for example if a package ends up becoming unmaintained. But if it's not been updated for that long, it and its dependencies could be added to the various monolithic repositories. I'm sure organisations would pop up that would keep 'dead' packages around for people to use. The way to combat this would be to have as much redundancy as possible, of course.
I don't know. It might just work better than what we've got at the moment?
~phil
While this is quite likely true, university degree programmes these days are focussed on passing exams more than anything else. Just because you're able to pass exams well does not mean you're able to innovate, and, even more importantly, vice versa. Quite frankly, if you're the kind of person who is prone to coming up with original ideas, I doubt you need exams as stimulation. Lectures? Definitely.
~phil
I'm aware of that, but try convincing semi-bad programmers to use 'size_t' rather than 'int' when appropriate, or making use of the 'sizeof' operator instead of hardcoding the values. (I don't let people like that anywhere near my source code repositories ;)) Besides, if you don't have access to the source, you're still stuck, no matter how easy it would be to port if you did have the source. I think this is the issue that MS (and Sun for that matter) wanted to address with .NET and Java.
~phil
Independence of which hardware platform you're running Windows on. I think the .NET project was started when it was still believed that Itanium might make it. Then, there's the whole Issue of porting all those 32-bit x86 apps to the AMD64 architecture. Moving Windows to a new architecture is by no means as easy as it is to move GNU and Linux, or most other Free Software to another architecture. It's a chicken-and-egg problem, and as such, it's much more pronounced on proprietary systems, you have to get all your software vendors to port their software for you. AMD64 would have been stillborn if there wasn't a way of running 32-bit x86 applications on it, much like it was the case for 16-bit applications on the 386 and later.
~phil
Oh yes, definitely. Some people have no affinity to this kind of stuff, which isn't really a problem in itself, there are plenty of other fulfilling and interesting things to do that don't require maths. She seems to have found something that interests her. I know intelligent biology students, fresh out of school, who couldn't solve a linear equation if their life depended on it.
The point I was trying to make was that there are people who do have a natural talent for maths and physical sciences, but are not exposed to maths at an early enough stage in their life, when their minds are still much more fertile. This is especially problematic if they're interested in science, or engineering, but aren't especially excited about maths itself, and then are scared off by the amount of maths required to do anything useful.
~phil
Actually, I think, speaking as a Computational Physics student, the problem of lack of interest does exist, and lies a lot earlier than when it comes to choosing a subject to study at university.
At least in Europe, there doesn't seem to be enough mathematical education in primary school and early secondary school to allow later science courses at school to even spark interest. For most of the people on my course that I've talked to, university physics is nothing like the physics they did at school, primarily because they completely lacked the mathematical foundations. I'm not saying teach calculus to 10-year olds, but some 18-year olds who have been at school all their life can't multiply fractions, not because of lack of intelligence (there would not be any point recruiting those cases for university anyway, sorry) but simply because they were never taught it. Of course they're not going to want to study a maths-heavy science, because it seems incredibly daunting.
I don't think this is specific to one or two countries, let alone particular political factions, but (at the very least) something that the entire Western society faces. Educational institutions and teachers are slowly doing the jobs that used to be the parents' responsibilities. No wonder they have no time for real teaching.
Increases in investment into science probably would help, I don't disagree with you on that point. However, I think attacking the root of the problem, early education, would be a much more effective use of resources. If there is enough inherent interest, additional funding for science can go into research rather than recruitment.
By the way, I don't claim to have a practical solution to the problem, but I think the problem hasn't even been identified properly yet by the scientific community. As a member of the Institute of Physics, I constantly get information about this and that campaign to raise science awareness among teenagers and young adults. I believe these attempts at solving the problem are fundamentally flawed in their short-sightedness.
Let's face it, science is never going to be 'cool' to the masses. Instead, let's encourage those who have a natural interest in the subject.
~phil
Well, if you need an int of a particular size, you need to typedef yourself a compiler/platform specific one anyway; this has always been that way. I believe 'int' used to be 2 bytes on most 16-bit DOS compilers. If you're talking about pointer arithmetic and array indexing, you should have been using size_t for those uses all along.
By the way, the GCC sizes for i386 and x86_64 are:
int: 4 bytes, 4 bytes
long: 4 bytes, 8 bytes
long long: 8 bytes, 16 bytes
size_t: 4 bytes, 8 bytes (I believe it's typedef'ed as an unsigned long)
Note that long long isn't ANSI/ISO compatible.
~phil