Pseudo-Code should be used as a teaching tool in programming courses
I was forced to take a required "programming logic" class my freshman year. It was entirely pseudo code, it was nothing but hand-tracing, and it was biggest fucking waste of time in the world. The instructor was an old fucking jackass who insisted that "in the real world" we would be required to debug our code on paper before we would ever be allowed to submit it for compilation. This was in 1991. We learned nothing in this class but how to write bad programs in pseudo code. Perhaps my issue was that I could already program in Pascal, Basic, C, and 6502 assembly (and had programming credit on a commercial C64 game title) and there was no option to "test out" of the mandatory introductory courses, but I don't think the beginners got much out of it either--- particularly when they were already all taking Pascal concurrently.
Pseudo code is a waste of time. Programming logic should be taught with actual computers, using an actual programming language.
I think C++ is way too complex for a first language, and the fact that it is really two languages (three if you count the preprocessor) in one is a big part of why.
I assert that the object-oriented portion of C++ isn't so much a second language as it is a tool for hiding complexity. The problem with learning languages like Java first is that they force you into OO behavior for everything. OO techniques are best implemented as a simplifying measure that allows you to build complex tools out of simple parts, and there are no simpler platform-agnostic parts than you get with C.
In my opinion you should start ingraining the OO paradigm as soon as possible
Hogwash. I say give them microcontrollers and teach them embedded programming in C first. Hiding the nuts and bolts of the underlying system behind multiple levels of object inheritance teaches a disregard for the limitations of the hardware. OO programming is better taught as an organizational tool for managing things beyond a certain level of complexity. OO programming excels at tasks that would become too byzantine if attempted in a straight procedural language, but is in my opinion an abomination when teaching "hello world". Throwing new students straight into OO land doesn't give them an appreciation for how the computer actually works.
maybe Python (though I personally detest it) or C# (which is a very nice language with very nice tools)
I'd have to second this one. I too find Python contemptible, mostly for its overly Basic-like syntax, but it is fairly easy to pick up. I think C# is good for two reasons: 1) free tools from microsoft, 2) teaches sensible C based syntax.
a good programmer should know all the chain, from Java/Python/Scheme/Whatever down to the machine code.
Yes, but you don't start with assembly language. You start with something conceptually simple, like Python. I started with Basic on the Commodore 64. Before a year was up, I was doing shit in 6502 assembly because interpreted Basic was too slow. Not a chance in hell I could've picked up assembly straight off without some understanding of a higher level language. Throwing assembly at someone is like throwing a pile of parts and fasteners at someone and telling them to build a combine harvester.
What a bloody nightmare. I learned Pascal in high school. Two years later when I learned C, I found myself wondering why the fuck anyone would inflict Pascal on people trying to learn programming. I later ended up maintaining a piece of business software written in Turbo Pascal (and later, stuff written in fucking Delphi) and found myself entertaining fantasies of building a time machine and choking Anders Hejlsberg to death in his bed in 1980 before he could inflict his monstrosity on the world.
That mindset is what brought us Logo, one of the most awful things ever to grace the curriculum of mandatory "computer literacy" classes. There's nothing wrong with the language per se, but it's almost never taught as a simple version of Lisp, but rather as a "tell dah turtle to drawr U a purty pitcher" scriptable graphics pen. Likewise, all graphics oriented, Basic syntax inspired languages do is give you quick access to a very limited and circumscribed programming space. It's like suggesting paint-by-numbers to someone who wants to be an artist. In order to do anything complex with 3D graphics, you really need a strong grounding in just plain nuts-and-bolts programming first. It's true that many 15 year olds aren't going to be very interested in learning about string handling or stacks or the like, but if they're not, then they're not actually interested in programming, are they.
What's up with this guy? Is he seriously suggesting that there's something in Vista that people like..... and they left it out of Windows 7? I don't know which assertion is more crazy, that they'd intentionally leave good stuff out, or that Vista had anything people will look upon fondly!
I've noticed that people who claim to have a "million dollar genius idea" universally don't. If you're not clever enough to know your idea has no major stumbling blocks, the idea is unlikely to be particularly clever itself. People with big-money ideas don't go around asking for advice on how to implement them. They already know. Coming up with a workable implementation is the part of the idea that makes it valuable.
Case in point:
Many years ago, a friend and I came up with an idea guaranteed to make us rich: the Nakedizer. It's box that you plug in between your TV and your [VCR|DVD|cable box] and at the touch of a button, it uses complex algorithms and best-guess placement of composites of actual skin already previously displayed to edit out the clothes of people on TV. We'd be rich, naturally, if only we could get someone to build the hardware and write the software to our nebulous specs. Guaranteed million dollar genius idea, there.... just waiting for us to figure out how to implement it.
This is exactly right. The ISS and SST are a pair of largely useless projects that for the last couple decades merely served to justify each others' existence. The trouble started when the DoD demanded the shuttle be large enough to haul a KH-11 satellite. The minimum thrust and payload volume specs kept having to be pushed up to match the KH-11 design, which incrementally grew heavier and larger during development. Then, when it became apparent that the expense of operating a shuttle launch facility at Vandenberg for polar orbit launches would be far greater than simply sticking with unmanned launch vehicles, the DoD largely withdrew from the shuttle program. In the end, NASA ended up with a white elephant that didn't do what they'd originally wanted, but was too far along in the process to change. When commercial satellite launches and repairs turned out to be a poor business plan, they started tub-thumping for a space station Space Station Freedom, later reborn as the ISS, initially came about as a project to give the shuttle something to do. No particular goal, just "let's try building a space station with the shuttle". Now we have a space station in too low an orbit to be truly useful, and a shuttle we have to keep around so we have something to haul part up to our useless space station. I say turn the SST's into museum pieces, drop the ISS into the Pacific, and start off fresh with no freakin' albatrosses around our collective space exploring neck.
Stop being an idiot. OK, I'll extend you the benefit of the doubt and assume you've never experienced the type of jackass he's talking about. He's not talking about reading web pages, or testing the fucking interface of what he's working on. He's talking about unnecessary mouse-centric foolishness within the IDE. It's about the kind of dumbass that repeatedly clicks four deep into a menu to recompile a unit, rather than just hitting ctrl-F12.
As well he should, whithin that context. No truly competent user of software environment (x) depends upon pedestrian mousing through menus to do things. Keyboard shortcuts are what competent people use. Even when X = "Photoshop", one hand is all over the keyboard, because menus are for amateurs.
No, he's right. You've linked to the DAGR, and ain't hardly anyone using those yet. In Afghanistan we were issued the PLGR (that big tan thing on the guy's belt that looks like a Fisher-Price calculator. It can probably be dropped 100' onto concrete, but it sucks ass as a GPS. Most of us had assorted Garmin units that ran off AA batteries and left the PLGRs in their packing.
My GPS gets traffic data and routes you according to it. Trouble is, they only collect data on major highways and sometimes the GPS thinks it's faster sending you along a surface street that's already full of people doing the same thing, but since it has no surface street traffic data, it assumes there is no traffic. Other times, it'll repeatedly try to get you onto the freeway because it has inexplicably calculated that a slow freeway is faster than a parallel deserted canyon road a mile east. Again, it's wrong. Traffic data is better than complete ignorance, but it really doesn't help much.
Last year, my friend I went to a cottage , I reached first thanks to my tomtom, he got lost and called me for directions, I said "I dunno, the GPS got me here!". After struggling for an hour, he stepped into a big box store along the way in a small town and picked up a GPS, got there in about an hour after that.
To be fair to you and your friend, neither of you had any local knowledge to destroy-- that's why you both needed the GPS. What was missing that we used to use was a set of turn by turn directions on paper which you could read to him over the phone when he called.
Really, the question is silly. People who rely on GPS don't have local knowledge to destroy. In situations where they do, they ignore the GPS and use it instead. I use my GPS daily to find work sites I've never been to, but the ones I have, I spend a lot of time ignoring the GPS's instructions. "Make a left here onto the most congested street in the city", it suggests, while I retort "no, you idiot, I'm going to parallel that street on a small side street where there's no traffic". My GPS is good at reading a map, but it's a complete moron when it comes to actual local knowledge. Where the GPS shines is at giving accurate turn-by-turn directions based on your current position, which is a hell of an improvement over the kind of human generated guidance we used to have to put up with: "turn down the street with all the trees along it and turn left where the old schoolhouse used to be; when you see the big oak tree, you've gone too far". GPS isn't destroying local knowledge. It's just destroying infuriatingly bad directions generated by people with no navigation skills.
GM should have been here a decade ago... they could have produced something practical out of the EV1 experiment.
Not really. The necessary battery technology wasn't there yet, and simply continuing to produce a mediocre electric car wasn't going to change that. Battery tech is hardly something that's under-researched, and GM throwing a few additional million dollars at the problem would have been a meanigless drop in the bucket. What the EV-1 represents is a state (California) thinking it could make technology advance by passing a law.WHen battery technology reaches the point where electric cars are truly useful, people will be all over them.
No, you need to parse his first sentence correctly. Set1 = "all scandanavian countries, inclusive of UK", Set2 = "other countries (in europe) except switzerland".
Gimme a fuckin' break. We don't play "social discourse" on Slashdot. We play "Fuck You, Fuck your Mother, and Fuck the Horse You Rode In On" here. This isn't real life. This is where we call each other "fucktard" and "shitstain", and generally belittle one another with as sharp a wit as we can manage for minor transgressions such as spelling and grammar errors, or political beliefs. People here often like to pretend everyone here is a sophisticated braniac, but it's just not the case.
This is not entirely accurate--I just switched from AT&T to Verizon...to a Blackberry Storm..It has a SIM card as well, and is capable of GSM roaming!
The Storm 9530 is nothing more than a GSM-only Storm 9500 with a Verizon CDMA module shoehorned in along with a Euro frequency only GSM module. You still do not have a SIM card you can swap to a different CDMA phone if your Storm craps out, as the SIM is just a jury-rig solution to solve the problem of CDMA phones being bricks in Europe. My original objection still stands: there are no Verizon phones that their Grand Viziers haven't given the royal seal of approval to. You can't (for example) mail order something like an HTC Diamond2-- a phone the US probably won't see until 2010-- or an HTC Magic-- the latest Google Android phone-- and stick your SIM in it and go. In fact, good luck finding anything for Verizon that's not that same insipid drool-proof "buy ringtones now!" Verizon OS other than a Crackberry or Palm.
and also, you've forgotten something, where do people typically live?
No, that' was entirely my point. The original assertion was that Australia has better coverage, when realistically it doesn't have any better coverage once you get a certain distance from civilization than the US does. The only difference is that in the US, the pockets of "nowhere" tend to be scattered, whereas the pocket of nowhere in Australia is essentially one big chunk there in the middle. Same thing, though. northern Nevada and southeastern California are about the equivalent of central Australia. It's all dirt and rocks and snakes.
Yeah, great. Let's hear it for an additional 500msec (that's half a second) signal delay while your communication makes a 44K mile round trip to geosync orbit and back.
AT&T was originally created and managed by the Federal Government
Come on. They regulated and allowed it to operate as a de facto monopoly, but there's no sane way you could twist the facts of history and claim they created AT&T. Prior to 1885, AT&T was called Bell Telephone, which was named after its founder Alexander Graham Bell. Perhaps you've heard of him.
Slashdot Mirror
Pseudo-Code should be used as a teaching tool in programming courses
I was forced to take a required "programming logic" class my freshman year. It was entirely pseudo code, it was nothing but hand-tracing, and it was biggest fucking waste of time in the world. The instructor was an old fucking jackass who insisted that "in the real world" we would be required to debug our code on paper before we would ever be allowed to submit it for compilation. This was in 1991. We learned nothing in this class but how to write bad programs in pseudo code. Perhaps my issue was that I could already program in Pascal, Basic, C, and 6502 assembly (and had programming credit on a commercial C64 game title) and there was no option to "test out" of the mandatory introductory courses, but I don't think the beginners got much out of it either--- particularly when they were already all taking Pascal concurrently.
Pseudo code is a waste of time. Programming logic should be taught with actual computers, using an actual programming language.
I think C++ is way too complex for a first language, and the fact that it is really two languages (three if you count the preprocessor) in one is a big part of why.
I assert that the object-oriented portion of C++ isn't so much a second language as it is a tool for hiding complexity. The problem with learning languages like Java first is that they force you into OO behavior for everything. OO techniques are best implemented as a simplifying measure that allows you to build complex tools out of simple parts, and there are no simpler platform-agnostic parts than you get with C.
In my opinion you should start ingraining the OO paradigm as soon as possible
Hogwash. I say give them microcontrollers and teach them embedded programming in C first. Hiding the nuts and bolts of the underlying system behind multiple levels of object inheritance teaches a disregard for the limitations of the hardware. OO programming is better taught as an organizational tool for managing things beyond a certain level of complexity. OO programming excels at tasks that would become too byzantine if attempted in a straight procedural language, but is in my opinion an abomination when teaching "hello world". Throwing new students straight into OO land doesn't give them an appreciation for how the computer actually works.
maybe Python (though I personally detest it) or C# (which is a very nice language with very nice tools)
I'd have to second this one. I too find Python contemptible, mostly for its overly Basic-like syntax, but it is fairly easy to pick up. I think C# is good for two reasons: 1) free tools from microsoft, 2) teaches sensible C based syntax.
a good programmer should know all the chain, from Java/Python/Scheme/Whatever down to the machine code.
Yes, but you don't start with assembly language. You start with something conceptually simple, like Python. I started with Basic on the Commodore 64. Before a year was up, I was doing shit in 6502 assembly because interpreted Basic was too slow. Not a chance in hell I could've picked up assembly straight off without some understanding of a higher level language. Throwing assembly at someone is like throwing a pile of parts and fasteners at someone and telling them to build a combine harvester.
Turbo Pascal
What a bloody nightmare. I learned Pascal in high school. Two years later when I learned C, I found myself wondering why the fuck anyone would inflict Pascal on people trying to learn programming. I later ended up maintaining a piece of business software written in Turbo Pascal (and later, stuff written in fucking Delphi) and found myself entertaining fantasies of building a time machine and choking Anders Hejlsberg to death in his bed in 1980 before he could inflict his monstrosity on the world.
That mindset is what brought us Logo, one of the most awful things ever to grace the curriculum of mandatory "computer literacy" classes. There's nothing wrong with the language per se, but it's almost never taught as a simple version of Lisp, but rather as a "tell dah turtle to drawr U a purty pitcher" scriptable graphics pen. Likewise, all graphics oriented, Basic syntax inspired languages do is give you quick access to a very limited and circumscribed programming space. It's like suggesting paint-by-numbers to someone who wants to be an artist. In order to do anything complex with 3D graphics, you really need a strong grounding in just plain nuts-and-bolts programming first. It's true that many 15 year olds aren't going to be very interested in learning about string handling or stacks or the like, but if they're not, then they're not actually interested in programming, are they.
What's up with this guy? Is he seriously suggesting that there's something in Vista that people like..... and they left it out of Windows 7? I don't know which assertion is more crazy, that they'd intentionally leave good stuff out, or that Vista had anything people will look upon fondly!
Unless you have filed a patent, just thinking of it first doesn't mean jack shit.
I dunno, winged airplane flight via lift was a pretty damned good idea circa 1903
It was also hardly a novel idea though, even in 1903. It was the implementations that were novel.
I've noticed that people who claim to have a "million dollar genius idea" universally don't. If you're not clever enough to know your idea has no major stumbling blocks, the idea is unlikely to be particularly clever itself. People with big-money ideas don't go around asking for advice on how to implement them. They already know. Coming up with a workable implementation is the part of the idea that makes it valuable.
Case in point:
Many years ago, a friend and I came up with an idea guaranteed to make us rich: the Nakedizer. It's box that you plug in between your TV and your [VCR|DVD|cable box] and at the touch of a button, it uses complex algorithms and best-guess placement of composites of actual skin already previously displayed to edit out the clothes of people on TV. We'd be rich, naturally, if only we could get someone to build the hardware and write the software to our nebulous specs. Guaranteed million dollar genius idea, there.... just waiting for us to figure out how to implement it.
This is exactly right. The ISS and SST are a pair of largely useless projects that for the last couple decades merely served to justify each others' existence. The trouble started when the DoD demanded the shuttle be large enough to haul a KH-11 satellite. The minimum thrust and payload volume specs kept having to be pushed up to match the KH-11 design, which incrementally grew heavier and larger during development. Then, when it became apparent that the expense of operating a shuttle launch facility at Vandenberg for polar orbit launches would be far greater than simply sticking with unmanned launch vehicles, the DoD largely withdrew from the shuttle program. In the end, NASA ended up with a white elephant that didn't do what they'd originally wanted, but was too far along in the process to change. When commercial satellite launches and repairs turned out to be a poor business plan, they started tub-thumping for a space station Space Station Freedom, later reborn as the ISS, initially came about as a project to give the shuttle something to do. No particular goal, just "let's try building a space station with the shuttle". Now we have a space station in too low an orbit to be truly useful, and a shuttle we have to keep around so we have something to haul part up to our useless space station. I say turn the SST's into museum pieces, drop the ISS into the Pacific, and start off fresh with no freakin' albatrosses around our collective space exploring neck.
Stop being an idiot. OK, I'll extend you the benefit of the doubt and assume you've never experienced the type of jackass he's talking about. He's not talking about reading web pages, or testing the fucking interface of what he's working on. He's talking about unnecessary mouse-centric foolishness within the IDE. It's about the kind of dumbass that repeatedly clicks four deep into a menu to recompile a unit, rather than just hitting ctrl-F12.
Someone sure thinks highly of themselves......
As well he should, whithin that context. No truly competent user of software environment (x) depends upon pedestrian mousing through menus to do things. Keyboard shortcuts are what competent people use. Even when X = "Photoshop", one hand is all over the keyboard, because menus are for amateurs.
No, he's right. You've linked to the DAGR, and ain't hardly anyone using those yet. In Afghanistan we were issued the PLGR (that big tan thing on the guy's belt that looks like a Fisher-Price calculator. It can probably be dropped 100' onto concrete, but it sucks ass as a GPS. Most of us had assorted Garmin units that ran off AA batteries and left the PLGRs in their packing.
My GPS gets traffic data and routes you according to it. Trouble is, they only collect data on major highways and sometimes the GPS thinks it's faster sending you along a surface street that's already full of people doing the same thing, but since it has no surface street traffic data, it assumes there is no traffic. Other times, it'll repeatedly try to get you onto the freeway because it has inexplicably calculated that a slow freeway is faster than a parallel deserted canyon road a mile east. Again, it's wrong. Traffic data is better than complete ignorance, but it really doesn't help much.
Last year, my friend I went to a cottage , I reached first thanks to my tomtom, he got lost and called me for directions, I said "I dunno, the GPS got me here!". After struggling for an hour, he stepped into a big box store along the way in a small town and picked up a GPS, got there in about an hour after that.
To be fair to you and your friend, neither of you had any local knowledge to destroy-- that's why you both needed the GPS. What was missing that we used to use was a set of turn by turn directions on paper which you could read to him over the phone when he called.
Really, the question is silly. People who rely on GPS don't have local knowledge to destroy. In situations where they do, they ignore the GPS and use it instead. I use my GPS daily to find work sites I've never been to, but the ones I have, I spend a lot of time ignoring the GPS's instructions. "Make a left here onto the most congested street in the city", it suggests, while I retort "no, you idiot, I'm going to parallel that street on a small side street where there's no traffic". My GPS is good at reading a map, but it's a complete moron when it comes to actual local knowledge. Where the GPS shines is at giving accurate turn-by-turn directions based on your current position, which is a hell of an improvement over the kind of human generated guidance we used to have to put up with: "turn down the street with all the trees along it and turn left where the old schoolhouse used to be; when you see the big oak tree, you've gone too far". GPS isn't destroying local knowledge. It's just destroying infuriatingly bad directions generated by people with no navigation skills.
GM should have been here a decade ago... they could have produced something practical out of the EV1 experiment.
Not really. The necessary battery technology wasn't there yet, and simply continuing to produce a mediocre electric car wasn't going to change that. Battery tech is hardly something that's under-researched, and GM throwing a few additional million dollars at the problem would have been a meanigless drop in the bucket. What the EV-1 represents is a state (California) thinking it could make technology advance by passing a law.WHen battery technology reaches the point where electric cars are truly useful, people will be all over them.
No, you need to parse his first sentence correctly. Set1 = "all scandanavian countries, inclusive of UK", Set2 = "other countries (in europe) except switzerland".
Gimme a fuckin' break. We don't play "social discourse" on Slashdot. We play "Fuck You, Fuck your Mother, and Fuck the Horse You Rode In On" here. This isn't real life. This is where we call each other "fucktard" and "shitstain", and generally belittle one another with as sharp a wit as we can manage for minor transgressions such as spelling and grammar errors, or political beliefs. People here often like to pretend everyone here is a sophisticated braniac, but it's just not the case.
This is not entirely accurate--I just switched from AT&T to Verizon...to a Blackberry Storm..It has a SIM card as well, and is capable of GSM roaming!
The Storm 9530 is nothing more than a GSM-only Storm 9500 with a Verizon CDMA module shoehorned in along with a Euro frequency only GSM module. You still do not have a SIM card you can swap to a different CDMA phone if your Storm craps out, as the SIM is just a jury-rig solution to solve the problem of CDMA phones being bricks in Europe. My original objection still stands: there are no Verizon phones that their Grand Viziers haven't given the royal seal of approval to. You can't (for example) mail order something like an HTC Diamond2-- a phone the US probably won't see until 2010-- or an HTC Magic-- the latest Google Android phone-- and stick your SIM in it and go. In fact, good luck finding anything for Verizon that's not that same insipid drool-proof "buy ringtones now!" Verizon OS other than a Crackberry or Palm.
and also, you've forgotten something, where do people typically live?
No, that' was entirely my point. The original assertion was that Australia has better coverage, when realistically it doesn't have any better coverage once you get a certain distance from civilization than the US does. The only difference is that in the US, the pockets of "nowhere" tend to be scattered, whereas the pocket of nowhere in Australia is essentially one big chunk there in the middle. Same thing, though. northern Nevada and southeastern California are about the equivalent of central Australia. It's all dirt and rocks and snakes.
Yeah, great. Let's hear it for an additional 500msec (that's half a second) signal delay while your communication makes a 44K mile round trip to geosync orbit and back.
AT&T was originally created and managed by the Federal Government
Come on. They regulated and allowed it to operate as a de facto monopoly, but there's no sane way you could twist the facts of history and claim they created AT&T. Prior to 1885, AT&T was called Bell Telephone, which was named after its founder Alexander Graham Bell. Perhaps you've heard of him.
The problem with NYC skewing the averages is that on average most people in the US aren't in NYC.