(Does anyone else get the impression that almost the only maps available online are all stolen from the CIA World Factbook? Information Superhighway my ass...)
I drew the line there because that's when I started using Linux and so I have no idea what previous versions were like. >:)
In general my point is that students should be learning about fairly advanced features in OS design and therefore need to be looking at fairly recent operating systems. I'm not exactly sure which features are implemented in which versions, but I'm sure we can look that up before the first day of class.
The problem with writing an OS from scratch is that you'll never get into most of the features of a modern OS. In today's age of maintenance and consulting, it's more important for students to be able to read code and plug together reusable components than make a couple of basic components from scratch.
Ideally we'd be able to approach the study of operating systems from both sides, but as OSes are no longer a central part of CompSci curriculums, there just isn't time.
Seems to me that with that many holes in the case, G5s should collect dust pretty fast. And that amount of space would be difficult to cover with filters.
Um, but when you download them from a P2P source, you did pay for it? Well not you exactly, but all the future iPod purchasers and people who are currently burning to CD.
Our government is, in their infinite wisdom, publicizing (and therefore socializing) the music industry. If you don't approve, please contact your member of parliament. In the meanwhile, download without guilt.
(What I'm trying to figure out is whether, if I can borrow a CD from the library and make a copy of it, can the library puts MP3s on its web page?)
A multi-party system would work fine for Congress and the Senate, the problem is that the Presidency is winner-takes-all and therefore they need to represent as many of the voters as possible. If the other branches of government were more heterogeneous then maybe there'd actually be some checks and balances on the President.
In addition to a strong multi-party system, runoff voting in each district creates even more dynamic, representative houses. Just whatever you do, stay away from proportional representation!
(Personally, I think you guys should switch to the Westminister system -- it seems to be working pretty well up here.)
If you need employees to plug in RAM, then maybe you should have hired college students instead of grad students? I'll admit though, that most grad students I work with are also horrible programmers; but they're in grad school because of their poor programming ability -- they know they couldn't hold a developer position so they're training to do things more abstract.
The point of university degrees is that the students are getting theoretical fundamentals and learning how to learn. Unfortunately, most students aren't actually interested enough in their discipline to understand how to apply the fundamentals and have no interest in learning on their own time.
You can't seriously expect schools to teach things like platform-specific hardware and system administration -- the platforms will have changed by the time they graduate! And fitting more than one major programming project into a curriculum is very difficult (that's what co-op is for!). Unfortunately, the private sector has just enough influence on universities that they're teaching students skills which are just applied enough that they become obsolete.
Ideally, universities would teach students programming on a very theoretical level. Then give them an idea of how to convert that theory into practice in all sorts of ways. And finally ensure that they get good co-op or summer jobs where they can try all of this out.
Some really important paradigms like object oriented and functional* require machines that are at least younger than the students. It's also arguable that humans should still be programming assembler. And it's difficult to say whether students would learn more writing their own basic operating system from scratch or modifying the byzantian cathedral that is Linux. Finally, its a waste of time to have students doing networking on anything other than TCP/IP.
So in general I'd say Windows 98 or Linux 2.0.x is great for CompSci, but anything older than that might be pushing it.
* Even if you don't think they're good paradigms, they're still significant enough to be worth learning.
Well that's why I included the prion joke: because there's still the possibility that we don't understand life well enough to sterilize something properly.
But seriously, I don't believe it's present-centric to think that we can do a better job than the New World explorers. Science does make progress, and the universe is not infinitely complex (well, maybe on a subatomic level...).
Also, it's important to note that New World diseases did so much damage because once the effect of diseases on the native population was observed, the colonialists tried to use it to their advantage (eg: smallpox blankets). Consider how things would have been if colonization had been a bit slower, or even better if someone bothered to tell the natives about how to use smallpox scabs for vaccination!
The problem with outlawing human cloning is that, religious nuts aside, the legislation is obliqulely targetting the actual concern.
Identical twins are not bad (although they can be kind of spooky), therefore clones are not bad. However, people believe that clones are the first step towards genetic modification of humans, forced organ donation, armies of bounty hunters, etc. Therefore, we should outlaw cloning on a slippery-slope basis.
But if clones are the first step to at least one beneficial thing, wouldn't it make more sense to outlaw what we're actually trying to avoid? Otherwise the mad scientists of the world can go right on modifying genetically unique zygotes, crazy parents can go on forcing siblings to donate organs, and George Lucas can go on making his shitty movies while the world confidently sleeps underneith its anti-cloning laws.
Just about the only significant cost to society of straight cloning that I've heard of is the risk of monoculture. Well why not require that only n humans can be produced using the same DNA or have a rising tax or something?
I'm getting really tired of legislators that paint over liberties with a thick, sloppy brush because they don't have the cognitive powers to figure out what they're actually trying to accomplish.
I've always wondered about the "V1 rocket" considering the modern convention that "missiles are rockets designed to blow up". If this wasn't the convention during WW2, then can't we still retroactively rename it? Or is it because we're doing the most direct translation from the Deutsche overloaded word "rakete"?
Some people claim the intention of the US's 2nd Amendment really is to enable militia to overthrow the government. One of the arguments against this interpretation is that it is unreasonable to expect militia to fare as well against the US Army as they would have historically. Therefore cruise missles should be more legal than handguns -- harder to use in crimes against individuals and much more effective at usurping.
Presumably they stuck with 802.11b because they wanted their customers to use off-the-shelf receivers?
There's nothing technologically novel about sending digital data over radio waves, the reason that its so popular right now is that it's standardized which has lead to it becoming very cheap. So anything that doesn't follow that standard is not benefitting from economies of scale.
That's because NASA didn't bother to sterilize Galileo either because it wasn't practice at the time or because its mission suggested that such extra cost was unnecessary (you'll get one of those two reasons depending on which news report you read). Space agencies are of the opinion that they are now capable of producing probes with a reasonably low risk of contamination (the Beagle 2 is being manufacturered in a clean room).
I just hope they're happy when sentient organisms evolved from prions send their ships to invade Earth...
Well maybe all the jobs are getting outsourced because our education system is totally unrealistic, then?
I was kind of hoping that people would pretend that diploma->technologist, Master's->manager long enough to see my point, but you're right that treating things somewhat idealistically. On the other hand,/. seems to have an irrational bias against post-graduate education (and management in general), so I hope you were at least half joking.
Well then the developed Commonwealth universities had better start working on their PR!
I think the University of California model is a good one. UC* adds name-recognition to places like University of California at Riverside. It wouldn't be a far stretch for the (world-class) University of Toronto to take over administration of all the universities in Ontario -- I'm sure the rest of the regions could be similarly amalgamated.
You may be right, and the outsourcing to India certainly suggests that they have the concrete skills covered. However before I'd be willing to accept that the universities are all-round as good I'd want to see some demonstration of abstract skills, such as by winning a programming contest.
The question is: where do we get the project managers of the future?
As someone just about to leave university with a Master's in CS, I think I can say with some confidence that very few companies won't make their PMs start out as developers. Problem is, if there are no coders there's nowhere for PMs to cut their teeth. Clearly if the outsourcing of programming is the future, we need a radically different culture and probably a different education system for software professionals -- maybe in a few decades time the universities will figure that out?
Worst case scenario: the developed Commonwealth countries (the US and Ireland are lapsed members) export no products but post-secondary education. Many otherwise developed countries like Japan have demonstrated an inability to provide competitive education and the developing countries getting our jobs are still decades away from providing more than college-level skills. Plus as English is the language of business, wouldn't you want to get your education where people speak Business as their native tongue!
Lets consider a world population of 10 billion with average life expectancy of 70. If the average person spends 6 years in university (things are getting more complex), then we're looking at 850 million post-secondary students worldwide at any given time. Google suggests that 1 academic staff for every 10 students is not an unreasonable number, so that's 85 million jobs -- that's almost the entire US workforce right now! Add all the support staff to provide services to the academic staff and run the surrounding infrastructure and you've got yourself an economy!
The problem is that most people don't realise that some items should be crafted and some should be produced.
If it's needed in mass quantities, doesn't require intricate design, and price is important, then it should be produced. If it's one-of-a-kind, complex and difficult, and price is not an object, then it should be crafted.
The paradoxical thing about software, is that since it can be duplicated for free, the commodity items are the ones that should be crafted. So every in-house database front-end should be made in a production-line environment by technician-class workers (these can be outsourced). But operating systems and major applications should be designed with care.
For example: the reason Linux is better than Windows is that Microsoft develops software on a production line while open source uses the craft approach. When a big consulting company like IBM outsources their coding they won't have a similar quality drop because they're producing a bunch of simple products.
Slashdot Mirror
So man redundant links and not a one has a picture of where the tunnel will be located. :(
Can anyone find one? This is the best detail I could find.
(Does anyone else get the impression that almost the only maps available online are all stolen from the CIA World Factbook? Information Superhighway my ass...)
I drew the line there because that's when I started using Linux and so I have no idea what previous versions were like. >:)
In general my point is that students should be learning about fairly advanced features in OS design and therefore need to be looking at fairly recent operating systems. I'm not exactly sure which features are implemented in which versions, but I'm sure we can look that up before the first day of class.
The problem with writing an OS from scratch is that you'll never get into most of the features of a modern OS. In today's age of maintenance and consulting, it's more important for students to be able to read code and plug together reusable components than make a couple of basic components from scratch.
Ideally we'd be able to approach the study of operating systems from both sides, but as OSes are no longer a central part of CompSci curriculums, there just isn't time.
Even better: if you live anywhere West of Ontario you can find out who wins before you even go to vote!
Seems to me that with that many holes in the case, G5s should collect dust pretty fast. And that amount of space would be difficult to cover with filters.
Has anyone opened theirs up lately?
Um, but when you download them from a P2P source, you did pay for it? Well not you exactly, but all the future iPod purchasers and people who are currently burning to CD.
Our government is, in their infinite wisdom, publicizing (and therefore socializing) the music industry. If you don't approve, please contact your member of parliament. In the meanwhile, download without guilt.
(What I'm trying to figure out is whether, if I can borrow a CD from the library and make a copy of it, can the library puts MP3s on its web page?)
A multi-party system would work fine for Congress and the Senate, the problem is that the Presidency is winner-takes-all and therefore they need to represent as many of the voters as possible. If the other branches of government were more heterogeneous then maybe there'd actually be some checks and balances on the President.
In addition to a strong multi-party system, runoff voting in each district creates even more dynamic, representative houses. Just whatever you do, stay away from proportional representation!
(Personally, I think you guys should switch to the Westminister system -- it seems to be working pretty well up here.)
If you need employees to plug in RAM, then maybe you should have hired college students instead of grad students? I'll admit though, that most grad students I work with are also horrible programmers; but they're in grad school because of their poor programming ability -- they know they couldn't hold a developer position so they're training to do things more abstract.
The point of university degrees is that the students are getting theoretical fundamentals and learning how to learn. Unfortunately, most students aren't actually interested enough in their discipline to understand how to apply the fundamentals and have no interest in learning on their own time.
You can't seriously expect schools to teach things like platform-specific hardware and system administration -- the platforms will have changed by the time they graduate! And fitting more than one major programming project into a curriculum is very difficult (that's what co-op is for!). Unfortunately, the private sector has just enough influence on universities that they're teaching students skills which are just applied enough that they become obsolete.
Ideally, universities would teach students programming on a very theoretical level. Then give them an idea of how to convert that theory into practice in all sorts of ways. And finally ensure that they get good co-op or summer jobs where they can try all of this out.
Some really important paradigms like object oriented and functional* require machines that are at least younger than the students. It's also arguable that humans should still be programming assembler. And it's difficult to say whether students would learn more writing their own basic operating system from scratch or modifying the byzantian cathedral that is Linux. Finally, its a waste of time to have students doing networking on anything other than TCP/IP.
So in general I'd say Windows 98 or Linux 2.0.x is great for CompSci, but anything older than that might be pushing it.
* Even if you don't think they're good paradigms, they're still significant enough to be worth learning.
Well that's why I included the prion joke: because there's still the possibility that we don't understand life well enough to sterilize something properly.
But seriously, I don't believe it's present-centric to think that we can do a better job than the New World explorers. Science does make progress, and the universe is not infinitely complex (well, maybe on a subatomic level...).
Also, it's important to note that New World diseases did so much damage because once the effect of diseases on the native population was observed, the colonialists tried to use it to their advantage (eg: smallpox blankets). Consider how things would have been if colonization had been a bit slower, or even better if someone bothered to tell the natives about how to use smallpox scabs for vaccination!
The problem with outlawing human cloning is that, religious nuts aside, the legislation is obliqulely targetting the actual concern.
Identical twins are not bad (although they can be kind of spooky), therefore clones are not bad. However, people believe that clones are the first step towards genetic modification of humans, forced organ donation, armies of bounty hunters, etc. Therefore, we should outlaw cloning on a slippery-slope basis.
But if clones are the first step to at least one beneficial thing, wouldn't it make more sense to outlaw what we're actually trying to avoid? Otherwise the mad scientists of the world can go right on modifying genetically unique zygotes, crazy parents can go on forcing siblings to donate organs, and George Lucas can go on making his shitty movies while the world confidently sleeps underneith its anti-cloning laws.
Just about the only significant cost to society of straight cloning that I've heard of is the risk of monoculture. Well why not require that only n humans can be produced using the same DNA or have a rising tax or something?
I'm getting really tired of legislators that paint over liberties with a thick, sloppy brush because they don't have the cognitive powers to figure out what they're actually trying to accomplish.
Thankfully they have a site for her, too.
I've always wondered about the "V1 rocket" considering the modern convention that "missiles are rockets designed to blow up". If this wasn't the convention during WW2, then can't we still retroactively rename it? Or is it because we're doing the most direct translation from the Deutsche overloaded word "rakete"?
Some people claim the intention of the US's 2nd Amendment really is to enable militia to overthrow the government. One of the arguments against this interpretation is that it is unreasonable to expect militia to fare as well against the US Army as they would have historically. Therefore cruise missles should be more legal than handguns -- harder to use in crimes against individuals and much more effective at usurping.
Presumably they stuck with 802.11b because they wanted their customers to use off-the-shelf receivers?
There's nothing technologically novel about sending digital data over radio waves, the reason that its so popular right now is that it's standardized which has lead to it becoming very cheap. So anything that doesn't follow that standard is not benefitting from economies of scale.
That's because NASA didn't bother to sterilize Galileo either because it wasn't practice at the time or because its mission suggested that such extra cost was unnecessary (you'll get one of those two reasons depending on which news report you read). Space agencies are of the opinion that they are now capable of producing probes with a reasonably low risk of contamination (the Beagle 2 is being manufacturered in a clean room).
I just hope they're happy when sentient organisms evolved from prions send their ships to invade Earth...
Wow, that's awesome evidence. You're right: we're fucked.
Well maybe all the jobs are getting outsourced because our education system is totally unrealistic, then?
/. seems to have an irrational bias against post-graduate education (and management in general), so I hope you were at least half joking.
I was kind of hoping that people would pretend that diploma->technologist, Master's->manager long enough to see my point, but you're right that treating things somewhat idealistically. On the other hand,
Yeah, it's pretty nuts. My favourite part is how the guy is collecting donations to buy him a satellite dish -- at least he'll be working for it!
Well then the developed Commonwealth universities had better start working on their PR!
I think the University of California model is a good one. UC* adds name-recognition to places like University of California at Riverside. It wouldn't be a far stretch for the (world-class) University of Toronto to take over administration of all the universities in Ontario -- I'm sure the rest of the regions could be similarly amalgamated.
Would that be Heather Tesch you're talking about?
You may be right, and the outsourcing to India certainly suggests that they have the concrete skills covered. However before I'd be willing to accept that the universities are all-round as good I'd want to see some demonstration of abstract skills, such as by winning a programming contest.
The question is: where do we get the project managers of the future?
As someone just about to leave university with a Master's in CS, I think I can say with some confidence that very few companies won't make their PMs start out as developers. Problem is, if there are no coders there's nowhere for PMs to cut their teeth. Clearly if the outsourcing of programming is the future, we need a radically different culture and probably a different education system for software professionals -- maybe in a few decades time the universities will figure that out?
Lets consider a world population of 10 billion with average life expectancy of 70. If the average person spends 6 years in university (things are getting more complex), then we're looking at 850 million post-secondary students worldwide at any given time. Google suggests that 1 academic staff for every 10 students is not an unreasonable number, so that's 85 million jobs -- that's almost the entire US workforce right now! Add all the support staff to provide services to the academic staff and run the surrounding infrastructure and you've got yourself an economy!
The problem is that most people don't realise that some items should be crafted and some should be produced.
If it's needed in mass quantities, doesn't require intricate design, and price is important, then it should be produced. If it's one-of-a-kind, complex and difficult, and price is not an object, then it should be crafted.
The paradoxical thing about software, is that since it can be duplicated for free, the commodity items are the ones that should be crafted. So every in-house database front-end should be made in a production-line environment by technician-class workers (these can be outsourced). But operating systems and major applications should be designed with care.
For example: the reason Linux is better than Windows is that Microsoft develops software on a production line while open source uses the craft approach. When a big consulting company like IBM outsources their coding they won't have a similar quality drop because they're producing a bunch of simple products.