Rephrased: "Certificates of skill are worthless, what matters is your actual skill."
Well obviously they're no substitute for skill, and obviously if you have the skill but not the certificate that's just as good, but.. I'm not really sure what your point is.
I'm not saying the all things I brought up are useless. Far from it; many of them were the units I enjoyed most of all, but it is such a wide field that you need to make cutbacks to make room for some more practical aspects.
As much as I would regret not having them the article is all about CS grads being ill-equipped for the real world, and I think that's a valid criticism (especially when some are ill-equipped but have the knowledge to get up to speed in whatever sub-field no time, while others simply bluffed their way through because there were so few concrete tests of ability. Without the ability to distinguish the two the purpose of a CS degree is undermined, and articles like these are a consequence).
You can argue that CS is an abstract course and shouldn't be based on the current trends in computing, but as your XMPP example earlier shows, the stuff you learn in CS does need to be as practical to the wide field of computing as possible. (I've not been taught about XML by the way, surely something as important to XMPP as knowing TCP/IP)
I admit computer communications wasn't the best example, but as much as I enjoyed it I still think a case could be made for having different priorities. TCP/IP isn't going to change, the OSI model is pretty irrelevant really, and you don't usually need to know about it (certainly not at the level we're taught it).
The transistor level of computing, and to a lesser extend transport-layer network protocols, just isn't where stuff is happening right now (SCTP is interesting, but I think TCP/IP will be staying for now). Not many people end up writing multi-threaded linux servers or making chips (certainly not with transistors), yet semesters are spent on those and no time is spent on, say, the browser environment.
You can also argue that knowing these things about transistor logic and the specifics of protocols helps you do higher level things more efficiently, and I agree to an extent, but it does seem like a wasteful way to teach fairly basic good habits. Also that seems like a pragmatic consideration that's not in-line with CS as an abstract course.
I went into CS basically because I wanted to learn what makes computers tick, and that any future work I do would be based on that fundamental knowledge. But when we're being taught about an arcane linux call to open several read/write/error sockets at once to prevent locking issues due to a TCP/IP relic, or about csh or Perl's basic syntax, or about how the simplest components of a microprocessor work, I think the line between abstract computer science and useless trivia is being blurred, when there are so many really important things to teach.
Also "hit the ground running" is a bit of a vague statement; I'm really referring to the complete incompetence of an unfortunate subset of CS graduates to be of any use in the real world without substantial further training.
I'm not suggesting that CS students should leave uni, necessarily, with in-depth knowledge of RoR development using git, or whatever, rather that they shouldn't be quite so hopeless.
Anyway it's a complicated problem, but I don't think it's constructive to say "I used xyz in my career so we should keep it", or comparing what we learned here or there. I think it's a huge field, CS needs to cover its underpinnings and it's an impossible task, but that it could be doing a better job. I'm not calling any knowledge within the field of computing useless, just relative usefulness
(Sorry that this post is a bit rambly, it's late and you raise some interesting points)
I'm an undergraduate of computer science in the final year. My observations as concise as I can:
Most students don't really care about computer science. This means they do what they need to pass at worst, or get a decent mark at best. This is a problem because:
-- The units do not prepare you to hit the ground running when they reach the real world. (A few examples below)
-- Getting a pass, and often even a decent mark, depends more on friends who already did the unit, knowing the lecturer and his style of examining, and has very little to do with getting a true understanding.
-- If you are legitimately interested in the subject and go out of your way to really get to grips with it you are putting yourself at a serious disadvantage in terms of grades.
The units do not prepare you for the real world. Examples:
-- Computer communications; all about a topic which is all about abstracting away complexity so you don't need to know more than you need to use it, with constantly changing protocols. Fascinating, but not applicable.
-- Computer graphics; all about the basic primitives of 3D computer graphics; raytracing, rasterization, etc. How many people are really going to go on to use this stuff, that won't need to relearn it with regard to some specific API anyway e.g. OpenGL or DirectX? Again fascinating, but rarely applicable.
-- Hardware fundamentals; transistors, flip-flops, integrated circuits. How many of us will really be going on to deal with this sort of thing, when a washing machine can run Java for all it matters. It's a pretty specialized field. Really interesting, but we don't need to know it.
-- Project Design and Management; I can't summarize concisely enough all the ways this unit was a complete and utter waste of time. Truly depressing.
-- Computer communications 2; perhaps learning about the protocols is worthwhile, but do we really need to spend 6 months learning to implement linux (specifically) based servers in a robust way? The way it works based on interrupts and things is just horrible, and there are so many better alternatives to using low-level linux calls (even on linux). How many of us are really going to go out there are start building high-load linux servers from absolute scratch? Very interesting, but I'm sure I'll never use it.
-- Software engineering; maybe I've just not been exposed to it but I cannot believe people out there actually use UML in meaningful ways to do useful things. That might be my ignorance, but they didn't sell it and I know little of it; it was either common sense or nonsense.
-- OS fundamentals; which was all about disk caching; although very few people actually need to deal with that stuff (and we went into much more detail than you'd need to understand swap caches etc), and disks are getting phased out for increasing memory and solid-state drives.
By contrast there was nothing on web technology, nothing on languages popularized in the last 10 years, nothing on internationalization, very little on IPv6, etc.
I realize it would be stupid to say "I want them to teach us HTML5, the id3 engine, C# 4.0, Visual Studio 2010, Ruby on Rails and Objective-C"; they need to teach computer science not about certain products, and the stuff I've found the most useful is the most abstract stuff (DFAs, CFGs, BFS, RB-trees, Turing machines, hash-tables, etc, etc), and I realize it's a fast moving area like no other where it's hard to keep up, but there really is a lot that's wrong with the way things are now and there has to be a better compromise.
I think it could be improved by:
-- More emphasis on open-source software (making, not using); incentivize people to be inventive and put useful code out there. Much more useful, impressive and practical than any assignment.
-- More emphasis on interacting with real-world cases, so CS students hit the ground runnin
It's the default filesystem in *BSD, so it's very well maintained etc. It has journalling (or does it call it "soft updates"?) auto-defrag, etc, etc. You fsck it if you power off without umount but otherwise you won't need to.
It's definitely a perfectly capable, full-featured, modern filesystem.
I realize that, and would be worried for anyone leaving uni without practical experience, but it's more complicated than a neat quip (it always is).
I would be equally worried for someone with 4 years of only writing code to respond to business/personal needs, you need a healthy dose of both of course.
Us graduates of today probably learn more or less the same stuff you learned. For better of for worse it's kept fairly theoretical; theory is related to a real-world example of the theory wherever possible, but more likely than not it'll be a fairly dated example.
Group e-mail is definitely a problem.. E-mail is a problem period; really I think it's safe to say when a company like Google recreates e-mail, and the result is a superset of e-mail, it's probably going to be at least a step forward.
I just don't think people should dismiss it; I understand people not using it, which is a chicken and egg problem Google need to figure out and probably won't figure out, but don't dismiss it out of hand.
Fair enough, but your post didn't really indicate that it was a harmless post about taking the effects of daily aspirin, rather something that family members would be ashamed of
I don't really see why if it was nothing to be ashamed of it would be a bother, but without a link to the post which I doubt you'd want to provide it's hard to make any further judgement
I wouldn't get too worked up about the definition of words.. The field of AI is all about algorithms which simulate intelligence, and playing chess is about practice more than intelligence anyway
I think Google knows which side its bread is buttered on; if/when a better privacy policy becomes beneficial for its bottom line it'll adopt one. Until then I doubt DuckDuckGo is a significant threat
And the Seinfeld Kramer actor will forever be remembered as the nigger-comedian-guy, it sucks but it's not Google's fault that the stupid things people do stick.
I'm not a shill, I'm just saying if you want a respectable blog you need a domain and Godaddy.com has the best deals with $10/domain, plus advanced privacy protection including proxy e-mail for a minor charge.
Fair enough, it turns out it was "Casual fallacy" as in casual-gamer not "Causal fallacy" as in (what sounds like it could be) the name of some obscure logical fallacy
The thought that someone is going to take a counter-argument and go and look up some fallacy, and that they'll consider that helpful and informative, is silly.
Non-trivial arguments about the real world aren't simple enough to apply logical rules to, and mistakes in logic are easier to recognize by pointing them out with respect to the specific case, rather than by the generalized case.
The idea that you can define a set of axioms and predicates and use rules of inference to prove that the 3DS is an attempt to "destroy" Sony, or something else in a real-world debate, is really crazy, so I don't think concepts from hard-nosed logic and proof are actually useful.
Also people all too often refer to "fallacy" like a fancy word for "mistake" (see the response before this one "a priori"), or to dismiss someone's argument in an intentionally inaccessible way, which comes across like condescending nonsense.
Mainly it's too often used to turn the vocab of logic and proof into an underhanded debating tactic, which seems like the opposite of what it's supposed to be for.
We have no idea who "Sean Malstrom" is, and no credentials are given
His blog is hosted on 50webs.com
Apparently his insight is that the 3DS represents an attack on a company which releases a competing product
"Frontal attack" / "destroy" / "demise" sound stupid when talking about companies
"Birdmen and the Causal Fallacy" is the most obnoxious title for a gaming opinion article I've ever heard. I hate it when people name logical fallacies to try and back up their opinion
How do you "bash" an entire type of gaming? Who the hell cares?
Just responding because your post is remarkably offputting if the intention was to refer us to this site.
We use CNET, it lets you simulate any layer of a network stack, but really its better for teaching low-level networking by getting you to recreate a the OSI stack, rather than teaching you how to configure this or that type of node/router.
If you want you can see frames and acks between this and that node as they travel across the various links needed to reach the final point, a thorough way to visualize what's going on, but not the fastest way to teach someone how to use ifconfig or configure a cisco router.
Reading your post actually made me shift my views slightly towards Apple's App Store policies.. Those 4 things all sound horrible, and if they got you to fix them great.
Slashdot Mirror
Isn't whirlpool an AES-based hashing system?
Rephrased: "Certificates of skill are worthless, what matters is your actual skill."
Well obviously they're no substitute for skill, and obviously if you have the skill but not the certificate that's just as good, but.. I'm not really sure what your point is.
I'm not saying the all things I brought up are useless. Far from it; many of them were the units I enjoyed most of all, but it is such a wide field that you need to make cutbacks to make room for some more practical aspects.
As much as I would regret not having them the article is all about CS grads being ill-equipped for the real world, and I think that's a valid criticism (especially when some are ill-equipped but have the knowledge to get up to speed in whatever sub-field no time, while others simply bluffed their way through because there were so few concrete tests of ability. Without the ability to distinguish the two the purpose of a CS degree is undermined, and articles like these are a consequence).
You can argue that CS is an abstract course and shouldn't be based on the current trends in computing, but as your XMPP example earlier shows, the stuff you learn in CS does need to be as practical to the wide field of computing as possible. (I've not been taught about XML by the way, surely something as important to XMPP as knowing TCP/IP)
I admit computer communications wasn't the best example, but as much as I enjoyed it I still think a case could be made for having different priorities. TCP/IP isn't going to change, the OSI model is pretty irrelevant really, and you don't usually need to know about it (certainly not at the level we're taught it).
The transistor level of computing, and to a lesser extend transport-layer network protocols, just isn't where stuff is happening right now (SCTP is interesting, but I think TCP/IP will be staying for now). Not many people end up writing multi-threaded linux servers or making chips (certainly not with transistors), yet semesters are spent on those and no time is spent on, say, the browser environment.
You can also argue that knowing these things about transistor logic and the specifics of protocols helps you do higher level things more efficiently, and I agree to an extent, but it does seem like a wasteful way to teach fairly basic good habits. Also that seems like a pragmatic consideration that's not in-line with CS as an abstract course.
I went into CS basically because I wanted to learn what makes computers tick, and that any future work I do would be based on that fundamental knowledge. But when we're being taught about an arcane linux call to open several read/write/error sockets at once to prevent locking issues due to a TCP/IP relic, or about csh or Perl's basic syntax, or about how the simplest components of a microprocessor work, I think the line between abstract computer science and useless trivia is being blurred, when there are so many really important things to teach.
Also "hit the ground running" is a bit of a vague statement; I'm really referring to the complete incompetence of an unfortunate subset of CS graduates to be of any use in the real world without substantial further training.
I'm not suggesting that CS students should leave uni, necessarily, with in-depth knowledge of RoR development using git, or whatever, rather that they shouldn't be quite so hopeless.
Anyway it's a complicated problem, but I don't think it's constructive to say "I used xyz in my career so we should keep it", or comparing what we learned here or there. I think it's a huge field, CS needs to cover its underpinnings and it's an impossible task, but that it could be doing a better job. I'm not calling any knowledge within the field of computing useless, just relative usefulness
(Sorry that this post is a bit rambly, it's late and you raise some interesting points)
-- The units do not prepare you to hit the ground running when they reach the real world. (A few examples below)
-- Getting a pass, and often even a decent mark, depends more on friends who already did the unit, knowing the lecturer and his style of examining, and has very little to do with getting a true understanding.
-- If you are legitimately interested in the subject and go out of your way to really get to grips with it you are putting yourself at a serious disadvantage in terms of grades.
-- Computer communications; all about a topic which is all about abstracting away complexity so you don't need to know more than you need to use it, with constantly changing protocols. Fascinating, but not applicable.
-- Computer graphics; all about the basic primitives of 3D computer graphics; raytracing, rasterization, etc. How many people are really going to go on to use this stuff, that won't need to relearn it with regard to some specific API anyway e.g. OpenGL or DirectX? Again fascinating, but rarely applicable.
-- Hardware fundamentals; transistors, flip-flops, integrated circuits. How many of us will really be going on to deal with this sort of thing, when a washing machine can run Java for all it matters. It's a pretty specialized field. Really interesting, but we don't need to know it.
-- Project Design and Management; I can't summarize concisely enough all the ways this unit was a complete and utter waste of time. Truly depressing.
-- Computer communications 2; perhaps learning about the protocols is worthwhile, but do we really need to spend 6 months learning to implement linux (specifically) based servers in a robust way? The way it works based on interrupts and things is just horrible, and there are so many better alternatives to using low-level linux calls (even on linux). How many of us are really going to go out there are start building high-load linux servers from absolute scratch? Very interesting, but I'm sure I'll never use it.
-- Software engineering; maybe I've just not been exposed to it but I cannot believe people out there actually use UML in meaningful ways to do useful things. That might be my ignorance, but they didn't sell it and I know little of it; it was either common sense or nonsense.
-- OS fundamentals; which was all about disk caching; although very few people actually need to deal with that stuff (and we went into much more detail than you'd need to understand swap caches etc), and disks are getting phased out for increasing memory and solid-state drives.
-- More emphasis on open-source software (making, not using); incentivize people to be inventive and put useful code out there. Much more useful, impressive and practical than any assignment.
-- More emphasis on interacting with real-world cases, so CS students hit the ground runnin
It's the default filesystem in *BSD, so it's very well maintained etc. It has journalling (or does it call it "soft updates"?) auto-defrag, etc, etc. You fsck it if you power off without umount but otherwise you won't need to.
It's definitely a perfectly capable, full-featured, modern filesystem.
I realize that, and would be worried for anyone leaving uni without practical experience, but it's more complicated than a neat quip (it always is).
I would be equally worried for someone with 4 years of only writing code to respond to business/personal needs, you need a healthy dose of both of course.
Try TP-Link, their stuff works very well for me.
The protocol is open, you can interface however you want.
Us graduates of today probably learn more or less the same stuff you learned. For better of for worse it's kept fairly theoretical; theory is related to a real-world example of the theory wherever possible, but more likely than not it'll be a fairly dated example.
Group e-mail is definitely a problem.. E-mail is a problem period; really I think it's safe to say when a company like Google recreates e-mail, and the result is a superset of e-mail, it's probably going to be at least a step forward.
I just don't think people should dismiss it; I understand people not using it, which is a chicken and egg problem Google need to figure out and probably won't figure out, but don't dismiss it out of hand.
I signed on and no-one else was on, and I couldn't think what to do with it, so it must be a really silly idea right?
Fair enough, but your post didn't really indicate that it was a harmless post about taking the effects of daily aspirin, rather something that family members would be ashamed of
I don't really see why if it was nothing to be ashamed of it would be a bother, but without a link to the post which I doubt you'd want to provide it's hard to make any further judgement
I wouldn't get too worked up about the definition of words.. The field of AI is all about algorithms which simulate intelligence, and playing chess is about practice more than intelligence anyway
I think Google knows which side its bread is buttered on; if/when a better privacy policy becomes beneficial for its bottom line it'll adopt one. Until then I doubt DuckDuckGo is a significant threat
And the Seinfeld Kramer actor will forever be remembered as the nigger-comedian-guy, it sucks but it's not Google's fault that the stupid things people do stick.
Because digital data is far easier to copy and maintain over time than it is to degrade it manually.
I'm not a shill, I'm just saying if you want a respectable blog you need a domain and Godaddy.com has the best deals with $10/domain, plus advanced privacy protection including proxy e-mail for a minor charge.
Fair enough, it turns out it was "Casual fallacy" as in casual-gamer not "Causal fallacy" as in (what sounds like it could be) the name of some obscure logical fallacy
So he's characterizing a dichotomy as being a fallacy, how odiously obtuse!
So he's saying the hardcore-casual gamer distinction is mistaken, how boring..
The thought that someone is going to take a counter-argument and go and look up some fallacy, and that they'll consider that helpful and informative, is silly.
Non-trivial arguments about the real world aren't simple enough to apply logical rules to, and mistakes in logic are easier to recognize by pointing them out with respect to the specific case, rather than by the generalized case.
The idea that you can define a set of axioms and predicates and use rules of inference to prove that the 3DS is an attempt to "destroy" Sony, or something else in a real-world debate, is really crazy, so I don't think concepts from hard-nosed logic and proof are actually useful.
Also people all too often refer to "fallacy" like a fancy word for "mistake" (see the response before this one "a priori"), or to dismiss someone's argument in an intentionally inaccessible way, which comes across like condescending nonsense.
Mainly it's too often used to turn the vocab of logic and proof into an underhanded debating tactic, which seems like the opposite of what it's supposed to be for.
True. Which is why you can't get away with attacking him. You need to look at his arguments.
No-one has the time to consider the arguments of everyone with a blog account and free subdomain at 50webs.tk .
There's a bit more to it than that.
If he has something interesting to say why spend three paragraphs saying "this guy says Nintendo is trying to out-compete Sony, check it out!" ?
So you read the title, and that's it?
My post listed the reasons you made me want to avoid the link, so no I didn't follow the link.
Apparently people at Nintendo are reading his blog, because Nintendo's strategy seems to go in the direction Malstrom points :)
I hope you're kidding.
If you don't like what he has to offer, feel free to go somewhere else.
I think opinion pieces about brands are pretty sad generally, but I do understand that I don't have to read his site
Just responding because your post is remarkably offputting if the intention was to refer us to this site.
We use CNET, it lets you simulate any layer of a network stack, but really its better for teaching low-level networking by getting you to recreate a the OSI stack, rather than teaching you how to configure this or that type of node/router.
If you want you can see frames and acks between this and that node as they travel across the various links needed to reach the final point, a thorough way to visualize what's going on, but not the fastest way to teach someone how to use ifconfig or configure a cisco router.
The old west was a lot more tame than Hollywood portrayed it
Reading your post actually made me shift my views slightly towards Apple's App Store policies.. Those 4 things all sound horrible, and if they got you to fix them great.