His main objection to the 10% myth seems to be that it's used by people whose ideas he disagrees with. The evidience to support his own claim that we use 100% of our brains boils down to two points
> 1. Pet scans show that...over the course of a whole day, > just about all of the brain is used at one time or another.
If I have 200Meg of ram in my machine and I use pretty much all of it over the course of a day, but never more than 20% at a time, does that prove it is all necessary ?
> 2. The myth presupposes an extreme localization of functions > in the brain. If the "used" or "necessary" parts of the brain were scattered all around the organ, that would imply that much of the brain is in fact necessary.
No, it doesn't. You might as well claim that since apples are scattered around the apple tree, the whole tree is edible.
I am aware that neither of my objectictions demonstrates that we don't use/need all our brain. I am just irritated by sloppy thinking and flawed arguments from such an intellectually self-righteous writer.
Personally I suspect that most people use most of their brains. Nature is efficient: brains require a lot of energy to keep going so we would be unlikely to evolve larger brains than we need.
Doesn't the federal government have some sort of exemption form class action lawsuits ? I would be surprised if they don't.
Nice idea though. There has to be some campaign for changing the appalling state of patent law. It's not just the US either, but it's probably worst there, and the US pressures other countries into going along with their take on IP issues. The problem is that these stupid laws hurt everybody. Furthermore, it's not at all obvious how badly they hurt people. Silly patents like this make everybodies lives worse in a million minor ways. Oh, so now I have to enter credit information every time I buy something from a particular company. It's not the kind of thing people get passionate about. Nobody can estimate the true cost of inappropriate patents.
However, when laws hurt a small number of powerful companies they form an effective lobby to bribe congressmen with campaign contributions in no time. Any attempt to change patent laws will recieve an effective counter attack from the few large companies that actually benefit from them.
About design patterns. Read the GoF book by all means, it's a classic, and it provides a new level of vocab for talking about software design...
BUT:
it's kinda boring - especially if you have enough experience for the patterns to be familiar. Worse, it encourages people to think of software design as a kind of mix and match mechanical process. That is, I believe people should focus on coming up with elegent solutions for their particular problem, and if at the end they say - oh I've used a singleton here and a factory there... then great. The problem arises when people start thinking, well i've got a design problem here - what category of pattern fits best ? Er, lets see, well I think I should use a "composite object" here, now what group of classes are used to implement that...
Next thing you know someone has churned out a whole shitload of classes, with only the vaguest notion of what they are for, AND THAT'S THE DESIGNER, but with a preposterous belief that he has a good design because he "followed the rules of OO design". And don't even get me started on that Rational Rose/UML crap.
Don't get me wrong. I believe understanding patterns is a good thing. But, it's often used as one more straw to clutch for people who don't really know what they're doing. Like every other technique in software - someone smart creates something to make life easier, but quite a few people will then take that and use it as an excuse to think less.
Software has gone through this process frequently (assemblers, compilers, procedural software, OO, patterns) at each stage the bar is lowered you make it possible for idiots to create more complex software. It's dangerous. This is why many people prefer C to C++ (not me). Show some piece of MFC based crap to an experienced C programmer and they will recoil in horror, I don't blame them. The more powerful the tools, the bigger piece of vaguely working rubbish an idiot come up with. There is no antidote to sloppy thinking.
Powerful tools make life easier, but paradoxically require more, not less experience to use well. Having seen what utter crap can be produced when lousy programmers get their hands on OO languages, I shudder to think what monstrosities will be around in a few years time as a result of patterns.
Basically, I'm saying anyone who designs/writes OO software should be aware of patterns, (and UML etc), but then ignore them and concentrate on the actual specific problem. If you're good you will find your own patterns.
To summarise: if you let children play with power tools, you end up with a big fucking mess.
the process of updating your internal representation of yourself.
BTW, how is Edinburgh CS department these days. I hear that 1st years still cheats like crazy. I certainly did when I was there - it's a valuable skill to learn.
We're not going to a race of cyborgs. We're going to evolve into one cyborg. Once people start putting hardware in their heads they will use that hardware to improve communication, and once communication reaches a decent speed it no longer makes sense to think of a network as a collection processors - its is a single, massively parallel machine. As for this crap about all of humanity must be literate in order to compete with these hyper intelligent machines - how the heck does he figure that out. If computers can be smarter than humans they can be 1000s of times smarter than humans. The fact that a greater proportion of humans can read and write or have a vauge understanding of technology would be irrelevent. But, we're far closer to fusing hardware with wetware than we are to creating smart machines. It's easier and has more immediate benefit. Bring on the implants, preferably before I go senile.
Nothing new... Yes it is. Nobody has done this before. If you think it's easy, try it yourself. This is creating LIFE (if it works). Bacterium are living organisims. Making live organisms from dead ones is pretty impressive in my book.
I used to live in US and I honestly couldn't understand why there was not FAR MORE terrorist activity than there is.
It's not as though the US has not made a few enemies over the years. For instance the IRA may cause the British government a few problems, but Irish catholics have received wonderful treatment over the years compared to American Indians.
Then there are large sections of Muslims, various central american countries, not to mention miscellaneous crackpots who don't really have a legitimate excuse, but would just like to kill Americans out of some misguided jealousy.
I came to the conclusion that there must be a massively higher amount of surveillance going on than was generally acknowledged. Since hearing about Echelon and various other NSA activities I think this conclusion is correct.
Don't misunderstand - I'm not saying that all this surveillance is a good thing, but don't write off terrorists as a threat just because you don't often get bothered by them. There are plenty of potential terrorists out there.
Every programmer that I know who has tried using java for a commerical product (except servlets) would rather eat cockraches than go through that again.
Slashdot has large student/youngster bias. Yes- it has fewer pitfalls than C/C++, but so what. It's the power of the tools, not ease of use that matters in the long run. Java's libraries are so fucking buggy - etc, while the langauge itself is a SUBSET of C++ (well, almost).
Some gullible medical types are swooning over Pascals new "Penny-sillum" treatment. The increasingly popular treatment for various diseases has produced high expectations amongst various new fangled doctors.
But mention Penicillum to Phillipe Caas, product manager for "Leaches-R-us", and he seems almost on the verge of stifling a yawn. 'When you look at the hype versus the reality today,' says Caas, 'there is a big disconnect.
Maybe he is whistling in the dark. But then, Leeches-R-us has been here before. Remember Electricity? That was the radical new treatment of four years ago, an upstart product that threatened to smash Leaches-R-us once and for all.
Born in the labs of Monsieur Volt this product raised high expectations, but people who tried using it to cure syphillus just ended up with smoking testicles. Most surgeons agree that leaches are the best modern treatment for STDs.
And yet, Electricity is far from dead. On the contrary, there are still thousands of developers cobbling together Electricity products, but for purposes quite different than the overthrow of Leaches-R-us.
Talk to Leaches-R-us executives and they'll tell you that a similar fate awaits penicillum. Charles Whoever, director of business development in Leaches-R-us STD development unit, says the penicillum hype has already peaked. ''Cold hard reality is coming to bear,'' he says.
Some gullible medical types are swooning over Pascals new "Penny-sillum" treatment. The increasingly popular treatment for various diseases has produced high expectations amongst various new fangled doctors.
But mention Penicillum to Phillipe Caas, product manager for "Leaches-R-us", and he seems almost on the verge of stifling a yawn. 'When you look at the hype versus the reality today,' says Caas, 'there is a big disconnect.
Maybe he is whistling in the dark. But then, Leeches-R-us has been here before.
Remember Electricity? That was the radical new treatment of four years ago, an upstart product that threatened to smash Leaches-R-us once and for all.
Born in the labs of Monsieur Volt this product raised high expectations, but people who tried using it to cure syphillus just ended up with smoking testicles. Most surgeons agree that leaches are the best modern treatment for STDs.
And yet, Electricity is far from dead.
On the contrary, there are still thousands of developers cobbling together Electricity products, but for purposes quite different than the overthrow of Leaches-R-us.
Talk to Leaches-R-us executives and they'll tell you that a similar fate awaits penicillum. Charles Whoever, director of business development in Leaches-R-us STD development unit, says the penicillum hype has already peaked. ''Cold hard reality is coming to bear,'' he says.
as a way of discouraging competition. I truly believe MS encourages developers to use MFC in order to prevent other companies producing worthwhile software.
MS does not use MFC internally - they're not stupid. Even the designer of MFC knew it was terrible - it was created for people who wanted to be C++ programmers but couldn't handle it.
C++ is NOT right for beginners. Even professional C developers with 3+ years experience shouldn't be allowed to touch a C++ compiler until they've read Effective C++ (Scott Meyers)- there are many pitfalls.
I also think programmers should all be taught assembly at some stage.
Faster, more stable, just plain better. Some sites don't work well with KFM, so I email them telling them to fix the site;-)
And how many lines of source code is KFM...1/100 the size of mozilla. If you don't like something, you really can fix it without having to wade through reams and reams of mediocre eighties code.
I for one would like to be able to remember everything I see hear or touch. At least with 2.3TB I could remember everything I hear for about 4 years, (add speech recognition to make it searchable).
Imagine how great that would be in arguments: "You said this software would be completed on time" "No, I said it was a complete waste of time - I'll play it back for you"
And it would be really handy with the wife too...
Bring on the implants.. preferably before I go senile.
While the Xanadu system sounds technically and conceptually brilliant, the fact that twenty years down the line it still doesn't work is worrying. Is there a limit at which complex ideas become unimplementable, or has Xanadu simply been unlucky?
Complicated software that works pretty much always derives from something simple that works, not something complicated that doesn't work.
OOP is overrated, it's useful, but PHBs tend to forget that it's just even easier to write complete crap using OOP. There's this laugable notion: OOP == Good, !OOP == !good. I believe that the problem with the smalltalk version is that it is Xerox parc smalltalk which is not a commonly supported development environment. The C++ is auto-generated, so unlikely to be very readable.
You just don't know what generic programming means do you.
> what else are you using templates for ?
All sorts of stuff, check out STL or blitz (matrix/math library with fortran performance). Shit, if you're really sick you can write sin functions with templates [that are evaluated during compile time] float x =sin and have it replaced with float x = 0.39073... at compile time
I did. I hear this argument all the time, but have these people actually written a lot of Java or just heard and understood the arguments. I don't know anyone who has tried to use java in an industrial strength application and not become VERY annyed with java. The language is fine, but the libraries are buggy pieces of shit, and the garbage collection DOESN'T $@#!ing WORK. C++ has definite flaws but the lack of elegance is made up for by the fact that you can do anything, and IT WORKS.
1. Generic programming is NOT necessary, but it is a really nice feature.
True, but you could say that about anything. Human readable programming languages are not necessary, but they are a really nice feature.
> Templates are overglorified macros True in the same way that programmers are just overglorified VDU operators (ie plausible, but basically bullshit). They are the mechanism for generic programming in C++ and the main reason why for algorithmic code the C++ version will take 1000 lines compared to 4000 lines for java, while still being faster and more type safe.
> Not saying Java is the perfect OO language, but that C++ DEFINATELY isn't.
True, C++ is not a perfect OO langauge, but OO is not a perfect programming paradigm. C++ lets you chose the paradigm appropriate for the task.
After 10 years of C++ and 4 years of Java I am certain: object orientation is best kept to a minimum. Object-based (there is a difference) programming is the way to write robust maintainable code. Deep class hierachy's end up increasing interdependence between separate areas, and are mostly generated by mediocre programmers.
Perl - it's a great script language, so what, we're talking about programming here.
I did. I hear this argument all the time, but have these people actually written a lot of Java or just heard and understood the arguments. I don't know anyone who has tried to use java in an industrial strength application and not become VERY annyed with java. The language is fine, but the libraries are buggy pieces of shit, and the garbage collection DOESN'T $@#!ing WORK. C++ has definite flaws but the lack of elegance is made up for by the fact that you can do anything, and IT WORKS.
1. Generic programming is NOT necessary, but it is a really nice feature.
True, but you could say that about anything. Human readable programming languages are not necessary, but they are a really nice feature.
> Templates are overglorified macros True in the same way that programmers are just overglorified VDU operators (ie plausible, but basically bullshit). They are the mechanism for generic programming in C++ and the main reason why for algorithmic code the C++ version will take 1000 lines compared to 4000 lines for java, while still being faster and more type safe.
> Not saying Java is the perfect OO language, but that C++ DEFINATELY isn't.
True, C++ is not a perfect OO langauge, but OO is not a perfect programming paradigm. C++ lets you chose the paradigm appropriate for the task.
After 10 years of C++ and 4 years of Java I am certain: object orientation is best kept to a minimum. Object-based (there is a difference) programming is the way to write robust maintainable code. Deep class hierachy's end up increasing interdependence between separate areas, and are mostly generated by mediocre programmers.
PC world (uk) has Athlon based systems for about £1400 +vat You can pick one up today if you really want ( I'm tempted but, I'm waiting for a biathlon)
You're on the right track, but the main thing I take issue with is "Of course they would only hate it until they came to recognise that they were getting more done. The incurable tweaks and games players would quit. (yet another improvement in overall productivity!) " that would be assuming that people enjoy being productive. While that may true for some, I sincerely doubt ti's true for the majority. The main (99%) reason people go to work is that they need the money. Sure it helps if you have an interesting job and enjoy being productive, but basically work for most people is a drag and they couldn't care less how productive they are as long as they don't get sacked.
Tools which genuinely enhance productivity are generally used to let people work less hard rather than to achieve more.
The happier people are at work, the less they mind working, so preventing them from pissing time away by playing with their environment will just piss them off. Of course having good tools does make people happier so it can have a benefit aside from the obvious.
Productivity is mostly a social thing, MS software might make people less productive but I reckon that is more to do with increasing level of frustration making people think "are fuck it, why bother" rather than the obvious effect of lost work.
I wish SGI would release the source to this. They don't seem to realise that they're toast otherwise. OpenInventor is still the best library for developing interactive 3d apps (apps, not games). SGI is best at making 3D hardware, but if all 3d apps are developed for MS platforms (using farenheit or whatever) then SGI will have to compete in the MS-hardware compatable world. They just admitted that they can't do so. What gives ??
Their only chance is for developers to create 3D apps under Linux. Until SGI releases the source code to OpenInventor, very few people are going to do so. Still, until SGI figure out where their interest lies, we have to consider the alternatives.
TGS Inventor- great if you can afford it, but there are run-time fees involved, which eliminates certain possibilities.
Apprentice - "free for non-commerical use", it's inventor like, but hardly a replacement doesn't look like much is happening on this front yet.
Pyran - LGPL as far as I can tell no development has occured on this for a year or so there are a few things to fix if you're using a recent egcs.
Scene - LGPL looks promising, current version is 0.13, and it's problematic to compile currently. Probably your best bet.
Coin - QPL, only available through CVS at the moment, and as it's neither a free software project, nor past the alpha stage, I can't be bothered.
While I applaud the way they handled that situation, Apple are hardly people who should complain about bogus IP claims.
These are the people who sued MS for using an approach to HCI that apple had copied from Xerox . They are probably the worst offenders in terms of dubious IP claims of any high tech company, and that's against some some pretty tough competition".
Trademarking plain language makes no sense to me. If you want a trademark, invent a word.
Although, maybe MS should be entitled to "Where are we going to take you today"
Redhat's stock value stands a good chance of getting virtually WIPED out when the net stock bubble bursts dragging all tech stocks into the dust. Amazon, ebay etc cannot continue to be valued this highly with no profits forever. When they go down they'll go down hard - a realistic valuation of most net stocks is roughly 1/30 of current value. This could well happen shortly after the Redhat IPO.
If you really want to buy RH, I suggest buying some then, when it's nice and cheap.
It's only worth doing on classes that provide interface between major modules (eg between one programmers area of responsibility and anothers) - but worth it since implementation changes never require recompilation of client modules.
And you're really contributing to the level of respect people would have for the "us" whoever the hell "us" is supposed to be. Who are you ?? He can spend his time however he pleases, and are you seriously suggesting people shouldn't say whatever they feel, how is that advocating freedom ?
The article was funny. RMS does seem awfully self righteous and it's irritating. Re-definition of words is a powerful technique (1984) and is also my main problem with the FSF. Parody is just what's needed.
Slashdot Mirror
His main objection to the 10% myth seems to be that it's used by people whose ideas he disagrees with. The evidience to support his own claim that we use 100% of our brains boils down to two points
> 1. Pet scans show that...over the course of a whole day,
> just about all of the brain is used at one time or another.
If I have 200Meg of ram in my machine and I use pretty much all of it over the course of a day, but never more than 20% at a time, does that prove it is all necessary ?
> 2. The myth presupposes an extreme localization of functions
> in the brain. If the "used" or "necessary" parts of the brain were scattered all around the organ, that would imply that much of the brain is in fact necessary.
No, it doesn't. You might as well claim that since apples are scattered around the apple tree, the whole tree is edible.
I am aware that neither of my objectictions demonstrates that we don't use/need all our brain. I am just irritated by sloppy thinking and flawed arguments from such an intellectually self-righteous writer.
Personally I suspect that most people use most of their brains. Nature is efficient: brains require a lot of energy to keep going so we would be unlikely to evolve larger brains than we need.
Doesn't the federal government have some sort of exemption form class action lawsuits ? I would be surprised if they don't.
Nice idea though. There has to be some campaign for changing the appalling state of patent law. It's not just the US either, but it's probably worst there, and the US pressures other countries into going along with their take on IP issues. The problem is that these stupid laws hurt everybody. Furthermore, it's not at all obvious how badly they hurt people. Silly patents like this make everybodies lives worse in a million minor ways. Oh, so now I have to enter credit information every time I buy something from a particular company. It's not the kind of thing people get passionate about. Nobody can estimate the true cost of inappropriate patents.
However, when laws hurt a small number of powerful companies they form an effective lobby to bribe congressmen with campaign contributions in no time. Any attempt to change patent laws will recieve an effective counter attack from the few large companies that actually benefit from them.
"Democracy is coming to the USA" - Leonard Cohen
About design patterns. Read the GoF book by all means, it's a classic, and it provides a new level of vocab for talking about software design...
BUT:
it's kinda boring - especially if you have enough experience for the patterns to be familiar.
Worse, it encourages people to think of software design as a kind of mix and match mechanical process. That is, I believe people should focus on coming up with elegent solutions for their particular problem, and if at the end they say - oh I've used a singleton here and a factory there... then great. The problem arises when people start thinking, well i've got a design
problem here - what category of pattern fits best ? Er, lets see, well I think I should use a "composite object" here, now what group of classes are used to implement that...
Next thing you know someone has churned out a whole shitload of classes, with only the vaguest notion of what they are for, AND THAT'S THE DESIGNER, but with a preposterous belief that he has a good design because he "followed the rules of OO design". And don't even get me started on that Rational Rose/UML crap.
Don't get me wrong. I believe understanding patterns is a good thing. But, it's often used as one more straw to clutch for people who don't really know what they're doing. Like every other technique in software - someone smart creates
something to make life easier, but quite a few people will then take that and use it as an excuse to think less.
Software has gone through this process frequently (assemblers, compilers, procedural software, OO, patterns) at each stage the bar is lowered you make it possible for idiots to create more complex software. It's dangerous. This is why many people prefer C to C++ (not me). Show some piece of MFC based crap to an experienced C programmer and they will recoil in horror, I don't blame them. The more powerful the tools, the bigger piece of vaguely working rubbish an idiot come up with. There is no antidote to sloppy thinking.
Powerful tools make life easier, but paradoxically
require more, not less experience to use well. Having seen what utter crap can be produced when lousy programmers get their hands on OO languages, I shudder to think what monstrosities will be around in a few years time as a result of patterns.
Basically, I'm saying anyone who designs/writes OO software should be aware of patterns, (and UML etc), but then ignore them and concentrate on the actual specific problem. If you're good you will find your own patterns.
To summarise: if you let children play with power tools, you end up with a big fucking mess.
the process of updating your internal representation of yourself.
BTW, how is Edinburgh CS department these days.
I hear that 1st years still cheats like crazy.
I certainly did when I was there - it's a valuable
skill to learn.
We're not going to a race of cyborgs. We're going to evolve into one cyborg. Once people start putting hardware in their heads they will use that hardware to improve communication, and once communication reaches a decent speed it no longer makes sense to think of a network as a collection processors - its is a single, massively parallel machine. As for this crap about all of humanity must be literate in order to compete with these hyper intelligent machines - how the heck does he figure that out. If computers can be smarter than humans they can be 1000s of times smarter than humans. The fact that a greater proportion of humans can read and write or have a vauge understanding of technology would be irrelevent. But, we're far closer to fusing hardware with wetware than we are to creating smart machines. It's easier and has more immediate benefit. Bring on the implants, preferably before I go senile.
Nothing new... Yes it is. Nobody has done this before. If you think it's easy, try it yourself. This is creating LIFE (if it works). Bacterium are living organisims. Making live organisms from dead ones is pretty impressive in my book.
I used to live in US and I honestly couldn't understand
why there was not FAR MORE terrorist activity than there is.
It's not as though the US has not made a few enemies over the
years. For instance the IRA may cause the British government
a few problems, but Irish catholics have received wonderful treatment
over the years compared to American Indians.
Then there are large sections of Muslims, various central american
countries, not to mention miscellaneous crackpots who don't
really have a legitimate excuse, but would just like to kill Americans
out of some misguided jealousy.
I came to the conclusion that there must be a massively higher
amount of surveillance going on than was generally acknowledged.
Since hearing about Echelon and various other NSA activities I think
this conclusion is correct.
Don't misunderstand - I'm not saying that all this surveillance is
a good thing, but don't write off terrorists as a threat just because
you don't often get bothered by them. There are plenty of
potential terrorists out there.
Every programmer that I know
who has tried using java for a commerical product (except servlets) would rather eat cockraches than go through that again.
Slashdot has large student/youngster bias. Yes- it has fewer pitfalls than C/C++, but so what. It's the power of the tools, not ease of use that matters in the long run. Java's libraries are so fucking buggy - etc, while the langauge itself is a SUBSET of C++ (well, almost).
Some gullible medical types are swooning over Pascals new "Penny-sillum" treatment. The increasingly popular treatment for various diseases has produced high expectations amongst various new fangled doctors.
But mention Penicillum to Phillipe Caas, product manager for "Leaches-R-us", and he seems almost on the verge of stifling a yawn. 'When you look at the hype versus the reality today,' says Caas, 'there is a big disconnect.
Maybe he is whistling in the dark. But then, Leeches-R-us has been here before. Remember Electricity? That was the radical new treatment of four years ago, an upstart product that threatened to smash Leaches-R-us once and for all.
Born in the labs of Monsieur Volt this product raised high expectations, but people who tried using it to cure syphillus just ended up with smoking testicles. Most surgeons agree that leaches are the best modern treatment for STDs.
And yet, Electricity is far from dead. On the contrary, there are still thousands of developers cobbling together Electricity products, but for purposes quite different than the overthrow of Leaches-R-us.
Talk to Leaches-R-us executives and they'll tell you that a similar fate awaits penicillum. Charles Whoever, director of business development in Leaches-R-us STD development unit, says the penicillum hype has already peaked. ''Cold hard reality is coming to bear,'' he says.
Some gullible medical types are swooning over Pascals new "Penny-sillum" treatment. The increasingly popular treatment for various diseases has produced high expectations amongst
various new fangled doctors.
But mention Penicillum to Phillipe Caas, product manager for "Leaches-R-us", and he seems almost on the verge of stifling a yawn. 'When you look at the hype versus the reality today,' says Caas, 'there is a big disconnect.
Maybe he is whistling in the dark. But then, Leeches-R-us has been here before.
Remember Electricity? That was the radical new treatment of four years ago, an upstart product that threatened to smash Leaches-R-us once and for all.
Born in the labs of Monsieur Volt this product raised high expectations, but people who tried using it to cure syphillus just ended up with smoking testicles. Most surgeons agree that leaches are the best modern treatment for STDs.
And yet, Electricity is far from dead.
On the contrary, there are still thousands of developers cobbling together Electricity products, but for purposes quite different than the overthrow of Leaches-R-us.
Talk to Leaches-R-us executives and they'll tell you that a similar fate awaits penicillum. Charles Whoever, director of business development in Leaches-R-us STD development unit, says the penicillum hype has already peaked. ''Cold hard reality is coming to bear,'' he says.
as a way of discouraging competition. I truly believe MS
encourages developers to use MFC in order to prevent other
companies producing worthwhile software.
MS does not use MFC internally - they're not stupid. Even the designer
of MFC knew it was terrible - it was created for people who
wanted to be C++ programmers but couldn't handle it.
C++ is NOT right for beginners. Even professional C developers with
3+ years experience shouldn't be allowed to touch a C++ compiler until they've
read Effective C++ (Scott Meyers)- there are many pitfalls.
I also think programmers should all be taught assembly at some stage.
Faster, more stable, just plain better. Some sites don't work ;-)
well with KFM, so I email them telling them to fix the site
And how many lines of source code is KFM...1/100 the size
of mozilla. If you don't like something, you really can fix it without
having to wade through reams and reams of mediocre
eighties code.
I for one would like to be able to remember everything I see hear or touch.
At least with 2.3TB I could remember everything I hear for about 4 years,
(add speech recognition to make it searchable).
Imagine how great that would be in arguments:
"You said this software would be completed on time"
"No, I said it was a complete waste of time - I'll play it back for you"
And it would be really handy with the wife too...
Bring on the implants.. preferably before I go senile.
Nah, make -j2 is fastest for me :
make --version
GNU Make version 3.76.1
While the Xanadu system sounds technically and conceptually brilliant, the fact that twenty years down the line it still doesn't work is worrying. Is there a limit at which complex ideas become unimplementable, or has Xanadu simply been unlucky?
Complicated software that works pretty much always derives from something
simple that works, not something complicated that doesn't work.
OOP is overrated, it's useful, but PHBs tend to forget that it's just even easier to write
complete crap using OOP. There's this laugable notion: OOP == Good, !OOP == !good.
I believe that the problem with the smalltalk version
is that it is Xerox parc smalltalk which is not a commonly
supported development environment. The C++ is auto-generated, so unlikely
to be very readable.
You just don't know what generic programming means do you.
/math library
> what else are you using templates for ?
All sorts of stuff, check out STL or blitz (matrix
with fortran performance). Shit, if you're really sick you can
write sin functions with templates [that are evaluated during
compile time]
float x =sin and have it replaced with
float x = 0.39073... at compile time
I did. I hear this argument all the time, but have these people
actually written a lot of Java or just heard and understood the
arguments. I don't know anyone who has tried to use java in an
industrial strength application and not become VERY annyed with
java. The language is fine, but the libraries are buggy pieces of shit,
and the garbage collection DOESN'T $@#!ing WORK. C++ has definite
flaws but the lack of elegance is made up for by the fact that you
can do anything, and IT WORKS.
1. Generic programming is NOT necessary, but it is a really nice feature.
True, but you could say that about anything. Human readable programming
languages are not necessary, but they are a really nice feature.
> Templates are overglorified macros
True in the same way that programmers are just overglorified
VDU operators (ie plausible, but basically bullshit).
They are the mechanism for generic programming in C++
and the main reason why for algorithmic code the C++ version
will take 1000 lines compared to 4000 lines for java, while
still being faster and more type safe.
> Not saying Java is the perfect OO language, but that C++ DEFINATELY isn't.
True, C++ is not a perfect OO langauge, but OO is not a perfect
programming paradigm. C++ lets you chose the paradigm
appropriate for the task.
After 10 years of C++ and 4 years of Java I am certain:
object orientation is best kept to a minimum. Object-based
(there is a difference) programming is the way to write robust maintainable code.
Deep class hierachy's end up increasing interdependence
between separate areas, and are mostly generated by
mediocre programmers.
Perl - it's a great script language, so what, we're talking about
programming here.
I did. I hear this argument all the time, but have these people
actually written a lot of Java or just heard and understood the
arguments. I don't know anyone who has tried to use java in an
industrial strength application and not become VERY annyed with
java. The language is fine, but the libraries are buggy pieces of shit,
and the garbage collection DOESN'T $@#!ing WORK. C++ has definite
flaws but the lack of elegance is made up for by the fact that you
can do anything, and IT WORKS.
1. Generic programming is NOT necessary, but it is a really nice feature.
True, but you could say that about anything. Human readable programming
languages are not necessary, but they are a really nice feature.
> Templates are overglorified macros
True in the same way that programmers are just overglorified
VDU operators (ie plausible, but basically bullshit).
They are the mechanism for generic programming in C++
and the main reason why for algorithmic code the C++ version
will take 1000 lines compared to 4000 lines for java, while
still being faster and more type safe.
> Not saying Java is the perfect OO language, but that C++ DEFINATELY isn't.
True, C++ is not a perfect OO langauge, but OO is not a perfect
programming paradigm. C++ lets you chose the paradigm
appropriate for the task.
After 10 years of C++ and 4 years of Java I am certain:
object orientation is best kept to a minimum. Object-based
(there is a difference) programming is the way to write robust maintainable code.
Deep class hierachy's end up increasing interdependence
between separate areas, and are mostly generated by
mediocre programmers.
PC world (uk) has Athlon based systems for about £1400 +vat
You can pick one up today if you really want
( I'm tempted but, I'm waiting for a biathlon)
You're on the right track, but the main thing I take issue with is
"Of course they would only hate it until they came to recognise that they were getting more done. The incurable tweaks and games players would quit. (yet another improvement in overall productivity!) "
that would be assuming that people enjoy being productive.
While that may true for some, I sincerely doubt ti's true for the majority.
The main (99%) reason people go to work is that they need the money. Sure it helps if you have an interesting job and enjoy being productive, but basically work for most people is a drag and they couldn't care less how productive they are as long as they don't get sacked.
Tools which genuinely enhance productivity are generally used to let people work less hard rather than to achieve more.
The happier people are at work, the less they mind working, so preventing them from pissing time away by playing with their
environment will just piss them off. Of course having good tools does make people happier so it
can have a benefit aside from the obvious.
Productivity is mostly a social thing, MS software might make people less productive
but I reckon that is more to do with increasing level of frustration
making people think "are fuck it, why bother" rather
than the obvious effect of lost work.
I wish SGI would release the source to this. They don't
seem to realise that they're toast otherwise. OpenInventor
is still the best library for developing interactive 3d apps (apps, not games).
SGI is best at making 3D hardware, but if all 3d apps are
developed for MS platforms (using farenheit or whatever)
then SGI will have to compete in the MS-hardware compatable
world. They just admitted that they can't do so. What gives ??
Their only chance is for developers to create 3D apps under Linux.
Until SGI releases the source code to OpenInventor, very few people
are going to do so. Still, until SGI figure out where their
interest lies, we have to consider the alternatives.
TGS Inventor- great if you can afford it, but there are run-time fees
involved, which eliminates certain possibilities.
Apprentice - "free for non-commerical use", it's inventor like, but hardly a replacement
doesn't look like much is happening on this front yet.
Pyran - LGPL as far as I can tell no development has occured on this for a year or so
there are a few things to fix if you're using a recent egcs.
Scene - LGPL looks promising, current version is 0.13, and it's problematic to
compile currently. Probably your best bet.
Coin - QPL, only available through CVS at the moment, and as it's neither
a free software project, nor past the alpha stage, I can't be bothered.
While I applaud the way they handled that situation, Apple are
hardly people who should complain about bogus IP claims.
These are the people who sued MS for using an approach to HCI
that apple had copied from Xerox . They are probably the
worst offenders in terms of dubious IP claims of any high tech
company, and that's against some some pretty tough competition".
Trademarking plain language makes no sense to me. If you want a
trademark, invent a word.
Although, maybe MS should be entitled to
"Where are we going to take you today"
The Redhat IPO is FAR from a sure thing.
Redhat's stock value stands a good chance of getting virtually WIPED out when the net stock bubble bursts dragging all tech stocks into the
dust. Amazon, ebay etc cannot continue to be valued this highly with no profits forever.
When they go down they'll go down hard - a realistic valuation of most net stocks is roughly 1/30 of current value. This could well happen shortly after the Redhat IPO.
If you really want to buy RH, I suggest buying some then, when it's nice and cheap.
It's a hack, but worth it on large, multiuser projects - .h file looks like this
class SomeImplementation;
class SomeInterface
{
SomeImplementation *imp_ptr;
public:
void method1();
etc...
};
It's a slight pain because programmer needs to instantiate SomeImplementation in SomeInterface constructor and then pass all methods onwards:
void SomeInterface::method1()
{
imp_ptr->method1();
}
etc
It's only worth doing on classes that provide interface between major modules (eg between one programmers area of responsibility and anothers) - but worth it since implementation changes never require recompilation of client modules.
And you're really contributing to the level of respect people would have for the "us" whoever the hell "us" is supposed to be. Who are you ?? He can spend his time however he pleases, and are you seriously suggesting people shouldn't say whatever they feel, how is that advocating freedom ?
The article was funny. RMS does seem awfully self righteous and it's irritating. Re-definition of words is a powerful technique (1984) and is also my main problem with the FSF. Parody is just what's needed.