but that's nto the same thing as saying "Java is faster than C++". The benchmarks in question are bout execution speed, not development speed.
Some of the benchmarks could have been far better implmented with code no more complex than what was arleady used on the c++ side... that's the issue here.
"java is faster for lazy coders" is a fair statement.. but not really what a benchmark tries to asess.
Perhaps not for a specific compiler, but for a specific language, YES.. there are right and wrong ways to implement an algorithm, and differences in implmenetation can have drastic effects on speed.
You can't say "Just write it any way you want, and the compiler will figure out what you meant and optimize it".. it doesn't work that way.
- How equivalent were the benchmarks? Where they programmed in an optimum way for their respective compilers and libraries? I'm sure the java ones were.. what about the C++ ones? The author states he doesn't understand G++ very well.
G++ is also known to not produce the best results.
"I rant it with -O2"
My guess is many of the tests were not implemented properly in c++.
The main clue would be this... I can understand java having better than expected performance.. but there is no way I can accept that java is that much FASTER than properly done C++... it doesn't make any sense.
modern english recognizes "Flammable" as a word meaning the same thing as "inflammable". Pick any modern dictionary, you'll find it. I suppose crusty old english professors might shun it, and say it's not a word.. but let's face it, english evolves.
Flammable is actually preferred, as many people mistake the "in" for the negative latin prefix, when it is not. Using "Flammable" removes this ambiguity.
Every airship story goes back to the hindenberg because that's all most people know of airships. The airship age basically ended.
The CEO is not an employee.. he is a corporate officer, and an insider, and many other things, no matter WHAT his personal opinions are. The same goes for the rest of the executive.
To draw a parallel.. the stock market. CEOs are not permitted to make speculative statements about the company stock. Why? Because they have far too much knowledge and influence about everything involved.
I realize that election is not the same thing at all, but it seems like a perfect parallel. If diebold is contracted to run the election, they need to be neutral and focus on running the systems, not on who wins.
If you are hiring an engineer, generally, you want a certified engineer, a degree is not enough (or even strictly necessary). You want someone who has passed the local (regional, whatever) professional engineering exams, and is certified by the local professional engineers association. THAT is an engineer. Someone who just has a degree in engineering is someone who studies engineering, but not an engineer.
Similarly, if you want a lawyer, you want someone who has passed the local Bar exam, and is recognized by all the other lawyers (and the legal system) as a lawyer.. NOT simply someone who has a PhD in Law.
The same goes for Doctors, etc.
There is NO WAY a university can know that a person will, later in life, cheat. If the person is competent enough to get through the process at the university, then the university should stand by their original decision. The person's own record will speak for itself.
Nobody is suggesting switching back to assembly....just that programms should have some experience with it.
It's also not about optimizing instructions.. to truly optimize on many modern processors requires years of study of that particular architecture in the first place.. that kind of task is best left to he compiler nowadays.
The point is that by coding some stuff in assembly, you learn a lot more about how the machine really works, how the OS really interacts with the machine, and your code.
Again, it's not about optimizing instructions but more about having a critical understanding of what is really going on.. taking out the magic.
Understanding things like:
How are arguments passed? By register, or by stack? Which is faster? Why? How are shared libraries handled? How does an interrupt affect things? What is a context-switch?
I agree totally that hand-optimized assembly is no longer necessary.
What IS often necessary is that working in assembly on a given system teaches you a lot about how the system (including the OS, library loaders, etc) actually works.
Understanding things like context switches, why some sycalls are slower than others, interrupts, and all that jazz are things that you pretty much can't pick up from high level coding.. you have to get down and dirty.
To learn better programming, for sure, learn lisp, learn to code beautiful, functional, streamlined code....
but that's all for nothing if the application uses resources in such a way that the system in question bogs down.
So long as we are talking about making full applications work in assembly in the target system.
(ie: not so much about the code & algorithms, but abouut how that code exists in the system.)
Understanding the OS, how sockets are dealt with, what context switches are, how shared libraries are handled (ie: Windows and Linux are VERY different in this regard, and not understanding this can lead you to write really shitty applications) is very important.
I've had to use so many applications that just SUCK from a systems point of view, because they were written without consideration for the target system.
I think assembly still has a lot to teach us, but it's no longer simply in the area of optimized instructions... that is done almost as well, if not better overall, by compilers nowadays. The real trick is that working in assembly makes you learn how the OS *really* works, and moves data around.. and that's what you really need to know.
The bank may be liable for negligence.. but that doesnt' equate to you automatically owning money deposited in your account.
It just means the bank is liable to the original depositor.
Money that is not yours that ends up in your account by accident is not yours, period.. and to take it is to invite trouble.. though certainly the bank is to blame.
I think, though, the intent of that statement was more about ensuring that some other copyright held by Novell at the time would not interfere with SCO using the rights it had purchased in the agreement.
This makes sense.
The problem is, it is too wordy, and open to interpretation.
That is a big assumption, based on nothing. Minix is of a completely different design.
That he used minix as a host, a development platform isn't even really relevant.... he could have used windows, or DOS or.... whatever had a compiler and the tools he wanted.
Nobody has a shred of evidence that he copied anything, ever, from the minix source code.
When people are prepared to lie about something, like if they were out at the movies last night instead of at the office, where they told the wife they would be... asking them a direct question will get you a direct lie?
"Were you at the movies last night?" "No"
Asking them an indirect question, though, and they will most likely speak before realizing their mistake.
"What did you think of the movie?" "It was good.. errr.. what movie?"
or better
"Did you see that accident on 5th street last night?" "What accident?"
Unless you are a game freak, you don't need windows, and WILL be happier using a mac, regardless of what misgivings you have, even though a mac may be more expensive. you will not regret your investment.
These are not a new thing.. they have been out for years.
And to all the doomsayers:
These things use a hard, flat surface, with something to separate it from your lap so you don't feel the heat. you know, LIKE YOUR DESK. If your desk isn't going to fry your laptop, neither are these.
a) Laptop "cooling" holders are not a new thing; they have been out for several years.
b) They generally have a hard surface, you know, JUST LIKE YOUR DESK. If the cooling pad is somehow going to fry your laptop.. so is your desk.
c) People often use a pillow or a blanket or other things on their lap when their laptop gets hot.. this is much, much, much worse than one of these cooling pads.
We always say microsoft has not innovated. WHat does that mean?
It means that what they have chosen to release into the public, to market, has not been innovative.. and has often been mediocre. We don't really respect them for it at all, right?
That doesn't mean that internally there is no innovation.. microsoft has a LOT of good programmers, and developers, and so-on... not everyone at microsoft is an MCSE know-it-all.. many are very talented, learned people.
Given that, and given some examples that slip through (like Office for the mac.. it's actually quite a bit nicer than the windows version)... you can see that they are capable of producing good software that plays nice.
The question is whether, as a company, they will choose to market such software.
If most of their solid income is from corporate windows workstation & server licensing... a model that requires lock-in and a fairly closed minded development model to continue generating revenue from... then they will naturally persue that over, say, writing good mac software that everyone likes, yet making far less money.
The problem, in short, is that they make the most money from their sleaziest practices...
Now, I understand why they rave like morons all day.
OS X isn't good because it's got OSS in it.... it's good becuase it's good. The fact that some of it is OSS is just a bonus.
I personally couldn't give a shit if darwin was OSS or not.. I just know I like the fact that the command line tools are standard bsd things, not re-invented crap... and that if I want to port some BSD or linux apps over, it's easy.
Sun has open source, sure.. but open source doesn't make a shitty operating system less shitty.. it just makes it more open shit.
Yes.. but when an ISP wants to sell DSL access.. is the DSL gear at the telco end?
I recall a DSL provider in Calgary... sure, Telus owned the phone lines.. but the network setup & DSL belonged to the ISP... meaning the ISP had lines terminating at their own equipment, and had their owned leased lines.
In the US, what was commonly done was a company would set up DSL access points, and a big frame relay or ATM network or whatever, and then re-sell access to the DSL network to various ISPs....
In most countries, the ISP is the ADSL infrastructure provider.
THe setup the US had with regional ADSL providers who resell to ISPs and all that convoluted nonsense (which didn't work) hasn't been mirrored, well, anywhere as far as I know.
AAC and MP3 do not have DRM either.. the DRM layer was added for the iTunes music STORE, which is a recent addition.... people bought ipods LONG before the iTMS existed.....
Why do people insist on thinking that ipods and itunes are all just about the store? The majority of ipod owners DONT use the store.
Slashdot Mirror
but that's nto the same thing as saying "Java is faster than C++". The benchmarks in question are bout execution speed, not development speed.
Some of the benchmarks could have been far better implmented with code no more complex than what was arleady used on the c++ side... that's the issue here.
"java is faster for lazy coders" is a fair statement.. but not really what a benchmark tries to asess.
Perhaps not for a specific compiler, but for a specific language, YES.. there are right and wrong ways to implement an algorithm, and differences in implmenetation can have drastic effects on speed.
You can't say "Just write it any way you want, and the compiler will figure out what you meant and optimize it".. it doesn't work that way.
There seem to be some unanswered questions here..
- How equivalent were the benchmarks? Where they programmed in an optimum way for their respective compilers and libraries? I'm sure the java ones were.. what about the C++ ones? The author states he doesn't understand G++ very well.
G++ is also known to not produce the best results.
"I rant it with -O2"
My guess is many of the tests were not implemented properly in c++.
The main clue would be this... I can understand java having better than expected performance.. but there is no way I can accept that java is that much FASTER than properly done C++... it doesn't make any sense.
You don't have the newest firefox for sure... I switched to firefox a few months ago, and rarely use safari now.
There is no point in using Camino anymore. Firefox looks and works excellently.. and doesn't feel like a foreign app anymore.
modern english recognizes "Flammable" as a word meaning the same thing as "inflammable". Pick any modern dictionary, you'll find it. I suppose crusty old english professors might shun it, and say it's not a word.. but let's face it, english evolves.
Flammable is actually preferred, as many people mistake the "in" for the negative latin prefix, when it is not. Using "Flammable" removes this ambiguity.
Every airship story goes back to the hindenberg because that's all most people know of airships. The airship age basically ended.
The CEO is not an employee.. he is a corporate officer, and an insider, and many other things, no matter WHAT his personal opinions are. The same goes for the rest of the executive.
To draw a parallel.. the stock market. CEOs are not permitted to make speculative statements about the company stock. Why?
Because they have far too much knowledge and influence about everything involved.
I realize that election is not the same thing at all, but it seems like a perfect parallel. If diebold is contracted to run the election, they need to be neutral and focus on running the systems, not on who wins.
Not as true as you'd like though.
If you are hiring an engineer, generally, you want a certified engineer, a degree is not enough (or even strictly necessary). You want someone who has passed the local (regional, whatever) professional engineering exams, and is certified by the local professional engineers association. THAT is an engineer. Someone who just has a degree in engineering is someone who studies engineering, but not an engineer.
Similarly, if you want a lawyer, you want someone who has passed the local Bar exam, and is recognized by all the other lawyers (and the legal system) as a lawyer.. NOT simply someone who has a PhD in Law.
The same goes for Doctors, etc.
There is NO WAY a university can know that a person will, later in life, cheat. If the person is competent enough to get through the process at the university, then the university should stand by their original decision. The person's own record will speak for itself.
It used to be a capital Omicron.. nowadays it's just a capital O.
Wikipedia does it better.
Nobody is suggesting switching back to assembly... .just that programms should have some experience with it.
It's also not about optimizing instructions.. to truly optimize on many modern processors requires years of study of that particular architecture in the first place.. that kind of task is best left to he compiler nowadays.
The point is that by coding some stuff in assembly, you learn a lot more about how the machine really works, how the OS really interacts with the machine, and your code.
Again, it's not about optimizing instructions but more about having a critical understanding of what is really going on.. taking out the magic.
Understanding things like:
How are arguments passed? By register, or by stack? Which is faster? Why?
How are shared libraries handled? How does an interrupt affect things?
What is a context-switch?
I agree totally that hand-optimized assembly is no longer necessary.
What IS often necessary is that working in assembly on a given system teaches you a lot about how the system (including the OS, library loaders, etc) actually works.
Understanding things like context switches, why some sycalls are slower than others, interrupts, and all that jazz are things that you pretty much can't pick up from high level coding.. you have to get down and dirty.
To learn better programming, for sure, learn lisp, learn to code beautiful, functional, streamlined code....
but that's all for nothing if the application uses resources in such a way that the system in question bogs down.
So long as we are talking about making full applications work in assembly in the target system.
(ie: not so much about the code & algorithms, but abouut how that code exists in the system.)
Understanding the OS, how sockets are dealt with, what context switches are, how shared libraries are handled (ie: Windows and Linux are VERY different in this regard, and not understanding this can lead you to write really shitty applications) is very important.
I've had to use so many applications that just SUCK from a systems point of view, because they were written without consideration for the target system.
I think assembly still has a lot to teach us, but it's no longer simply in the area of optimized instructions... that is done almost as well, if not better overall, by compilers nowadays. The real trick is that working in assembly makes you learn how the OS *really* works, and moves data around.. and that's what you really need to know.
Can you back that up?
The bank may be liable for negligence.. but that doesnt' equate to you automatically owning money deposited in your account.
It just means the bank is liable to the original depositor.
Money that is not yours that ends up in your account by accident is not yours, period.. and to take it is to invite trouble.. though certainly the bank is to blame.
I think, though, the intent of that statement was more about ensuring that some other copyright held by Novell at the time would not interfere with SCO using the rights it had purchased in the agreement.
This makes sense.
The problem is, it is too wordy, and open to interpretation.
That is a big assumption, based on nothing.
Minix is of a completely different design.
That he used minix as a host, a development platform isn't even really relevant.... he could have used windows, or DOS or.... whatever had a compiler and the tools he wanted.
Nobody has a shred of evidence that he copied anything, ever, from the minix source code.
This is a common interrogation technique.
When people are prepared to lie about something, like if they were out at the movies last night instead of at the office, where they told the wife they would be... asking them a direct question will get you a direct lie?
"Were you at the movies last night?" "No"
Asking them an indirect question, though, and they will most likely speak before realizing their mistake.
"What did you think of the movie?" "It was good.. errr.. what movie?"
or better
"Did you see that accident on 5th street last night?" "What accident?"
I bought a mac.
Unless you are a game freak, you don't need windows, and WILL be happier using a mac, regardless of what misgivings you have, even though a mac may be more expensive. you will not regret your investment.
These are not a new thing.. they have been out for years.
And to all the doomsayers:
These things use a hard, flat surface, with something to separate it from your lap so you don't feel the heat. you know, LIKE YOUR DESK. If your desk isn't going to fry your laptop, neither are these.
Err, okay.
I dont' get all the doomsayers here.
a) Laptop "cooling" holders are not a new thing; they have been out for several years.
b) They generally have a hard surface, you know, JUST LIKE YOUR DESK. If the cooling pad is somehow going to fry your laptop.. so is your desk.
c) People often use a pillow or a blanket or other things on their lap when their laptop gets hot.. this is much, much, much worse than one of these cooling pads.
Great.. enjoy your little XP box.
We always say microsoft has not innovated. WHat does that mean?
It means that what they have chosen to release into the public, to market, has not been innovative.. and has often been mediocre. We don't really respect them for it at all, right?
That doesn't mean that internally there is no innovation.. microsoft has a LOT of good programmers, and developers, and so-on... not everyone at microsoft is an MCSE know-it-all.. many are very talented, learned people.
Given that, and given some examples that slip through (like Office for the mac.. it's actually quite a bit nicer than the windows version)... you can see that they are capable of producing good software that plays nice.
The question is whether, as a company, they will choose to market such software.
If most of their solid income is from corporate windows workstation & server licensing... a model that requires lock-in and a fairly closed minded development model to continue generating revenue from... then they will naturally persue that over, say, writing good mac software that everyone likes, yet making far less money.
The problem, in short, is that they make the most money from their sleaziest practices...
I used to talk juts like you.
Then I got a mac.
Now, I understand why they rave like morons all day.
OS X isn't good because it's got OSS in it.... it's good becuase it's good. The fact that some of it is OSS is just a bonus.
I personally couldn't give a shit if darwin was OSS or not.. I just know I like the fact that the command line tools are standard bsd things, not re-invented crap... and that if I want to port some BSD or linux apps over, it's easy.
Sun has open source, sure.. but open source doesn't make a shitty operating system less shitty.. it just makes it more open shit.
Yes.. but when an ISP wants to sell DSL access.. is the DSL gear at the telco end?
I recall a DSL provider in Calgary... sure, Telus owned the phone lines.. but the network setup & DSL belonged to the ISP... meaning the ISP had lines terminating at their own equipment, and had their owned leased lines.
In the US, what was commonly done was a company would set up DSL access points, and a big frame relay or ATM network or whatever, and then re-sell access to the DSL network to various ISPs....
I don't belive telus does this?
In most countries, the ISP is the ADSL infrastructure provider.
THe setup the US had with regional ADSL providers who resell to ISPs and all that convoluted nonsense (which didn't work) hasn't been mirrored, well, anywhere as far as I know.
AAC and MP3 do not have DRM either.. the DRM layer was added for the iTunes music STORE, which is a recent addition.... people bought ipods LONG before the iTMS existed.....
Why do people insist on thinking that ipods and itunes are all just about the store? The majority of ipod owners DONT use the store.
Tried Irish Broadband? (IBB?)
http://www.irishbroadband.ie/
Both their residential service and business service are solid so far.