I think his basic point was that rather than design something correctly alot of people have the notion of "just throw more horsepower at it" I know many many programmers that the solution to poor design is add more memory and a second processor (especially java). I think that just seems silly. I'd rather see a nice tight design that takes a little more time than a quick sloppy one that can work if you have a monster of a machine (the reason being its easier to scale up when your not bloated).
Now I'm not saying it has to be this way I just I have seen more abuse of this idea from OO programmers than the hardcore C fans (I'm biased being a C nut:).
How do you know? Have you seen the prototype? Why can't people just let them do what they do best, design good product with HMI's in mind. If it comes out and sucks then you can start second guessing them but its just stupid to do it based off the patent filling.
Could be done but the whole point of altivec was for optimization (i.e. speed increases)... Do you think you'll still have the speed increase when you start transalting it?
the recording industry doesn't make money off the band touring. They make it off selling cd's. Bands like the greatful dead and phish made/make most of there money off of concert tickets.
Not that I disagree but have you used the Free clients compared to the transarc/IBM clients... they aren't there yet. And don't even talk about the servers... there is a real reason IBM charges what they do for an AFS instillation. Are there any free DFS clients?
But their margin of profit is up which is actually a good sign. The fact of the matter is that their machines have slipped in the speed department and its directly effecting the bottom line. I think if they can get the IBM 970 processor out the door in six months things will turn around pretty quickly... any longer than that and I think it will be a slower up take. You can't stay out of the game to long or the rest of the pack really passes you by.
Ahhh but to some managers and customers OO is a buzzword. If your whole system is written in C and works correctly you don't go ripping out parts that work just fine and replacing them with C++ just for the sake of having OO components. As for training new engineers it all about having good design and documentation and you can do that with C++ or C.
i always looked it as you can code for particular case easier than some writing a compilier. I mean general case VS specific. And the fact of the matter is C compiliers have been around alot longer than C++ copiliers (if they aren't just pre-processors for a C compilier).
Now mind you if you try and code alot of C++ feature in you are right it will turn out slower but for simple models and uses it will be faster.
I think his point was with a little bit of elbow grease and some function pointer you can do everything you need without the overhead of C++. It doesn't do everything but for some people and job its the right fit. I wrote a C struct/class loader that built its behavior off of XML files. It actually worked quite well and was really fun to write. I use it for the base of a few of my side projects.
Its an arguement against reimplementing something for the sake of using OO. There are to many people out there that believe things are outdated or obsolete just because they aren't OO.
But your right it isn't a reason not to use OO to solve a problem. Just a warning to any that think OO is the solution to everything. I guess it comes down to having good engineers and a good design (which can use OO but doesn't have to;-)
Don't let it out but I'm a huge fan of Objective C
That is very true C++ is a superset of C but we weren't talking about using C++ as if it where C. We where talking about the OO paradigm which is inherintely slower (because of ironicly enough inheritence... as well as a few other things).
I actually don't dislike C++ I just think that OO can and is used to much or incorrectly by ALOT of coders. I've seen things reimplemented in OO just for the sake of being buzzword complient and trust me they failed misserably.
But thats your oppinion of it being a chainsaw. For alot of things its not... its just a heavier axes.
I'll grant you that for something it is easier to approach with OO but at the same time you have a whole new set of problems to deal with.
I've just seen way to many attempts to write object oriented code that just turned out horrible. Just like every other language out there you can write some very bad code in C++.
OO is not the magic buillet but rather another tool to use when the time is right. The trick is finding a good coder to use the tool belt.
The point I was going after was the paragigm of inheritence was not need for a language to be useful which is where he was going.
Oh and my crappy little company is one of the top 3 defense contractors in the US i.e. 80,000+ employes. What crappy "little" company do you work for...
Oh and FYI C is and will continue (for a good long time) to be a faster language than C++ or java... So I wouldn't say I drive a pinto but rather a 911 with a single CD player. C++ is more like a 944 with the kitchen sink sitting in the passenger seat waying you down.
Some of us can't write crap code and expect it to run well on embedded processors...
Do you really believe that C has no value? I mean I can understand if your affraid to alloc your own memory and keep track of it but please don't assume everyone is inept!
He said no such thing I think you need to read before you post.
He's saying that just because it is a programming language it doesn't have to have inheritance to be a vaible language. There are many good languages out there that don't have inheritance. Trust me its not the end all be all in programming languages... give me C any day...
Slashdot Mirror
I think you have to move to NH ...
... its "Live, Freeze , and die"
And FYI
I think his basic point was that rather than design something correctly alot of people have the notion of "just throw more horsepower at it" I know many many programmers that the solution to poor design is add more memory and a second processor (especially java). I think that just seems silly. I'd rather see a nice tight design that takes a little more time than a quick sloppy one that can work if you have a monster of a machine (the reason being its easier to scale up when your not bloated).
:).
Now I'm not saying it has to be this way I just I have seen more abuse of this idea from OO programmers than the hardcore C fans (I'm biased being a C nut
Just my 2 cents.
Hmmm I always thought linux was a kernal I mean I could be wrong so how is it linux if it has a mach kernal?
Linux is a kernal so why would you run a kernal on top of a kernal?
If you'd get off your ass and bought yourself a nice piece of hardware you too could run a nice OS ;-)
How do you know? Have you seen the prototype? Why can't people just let them do what they do best, design good product with HMI's in mind. If it comes out and sucks then you can start second guessing them but its just stupid to do it based off the patent filling.
Could be done but the whole point of altivec was for optimization (i.e. speed increases) ... Do you think you'll still have the speed increase when you start transalting it?
the recording industry doesn't make money off the band touring. They make it off selling cd's. Bands like the greatful dead and phish made/make most of there money off of concert tickets.
FYI beirut (beer pong) is the game of champions ... See they even have a national tourney ... :)
Smoke crack for baby jesus? ... I want my files now not next week when the JVM decides its done cleaning up after itself!
Not that I disagree but have you used the Free clients compared to the transarc/IBM clients ... they aren't there yet. And don't even talk about the servers ... there is a real reason IBM charges what they do for an AFS instillation. Are there any free DFS clients?
But their margin of profit is up which is actually a good sign. The fact of the matter is that their machines have slipped in the speed department and its directly effecting the bottom line. I think if they can get the IBM 970 processor out the door in six months things will turn around pretty quickly ... any longer than that and I think it will be a slower up take. You can't stay out of the game to long or the rest of the pack really passes you by.
Ahhh but to some managers and customers OO is a buzzword. If your whole system is written in C and works correctly you don't go ripping out parts that work just fine and replacing them with C++ just for the sake of having OO components. As for training new engineers it all about having good design and documentation and you can do that with C++ or C.
I think you missunderstand sarcasm .... its ok you didn't really get the rest of the thread either.
i always looked it as you can code for particular case easier than some writing a compilier. I mean general case VS specific. And the fact of the matter is C compiliers have been around alot longer than C++ copiliers (if they aren't just pre-processors for a C compilier).
Now mind you if you try and code alot of C++ feature in you are right it will turn out slower but for simple models and uses it will be faster.
I think his point was with a little bit of elbow grease and some function pointer you can do everything you need without the overhead of C++. It doesn't do everything but for some people and job its the right fit. I wrote a C struct/class loader that built its behavior off of XML files. It actually worked quite well and was really fun to write. I use it for the base of a few of my side projects.
:)
But different strokes for different folks
Its an arguement against reimplementing something for the sake of using OO. There are to many people out there that believe things are outdated or obsolete just because they aren't OO.
;-)
But your right it isn't a reason not to use OO to solve a problem. Just a warning to any that think OO is the solution to everything. I guess it comes down to having good engineers and a good design (which can use OO but doesn't have to
Don't let it out but I'm a huge fan of Objective C
That is very true C++ is a superset of C but we weren't talking about using C++ as if it where C. We where talking about the OO paradigm which is inherintely slower (because of ironicly enough inheritence ... as well as a few other things).
I actually don't dislike C++ I just think that OO can and is used to much or incorrectly by ALOT of coders. I've seen things reimplemented in OO just for the sake of being buzzword complient and trust me they failed misserably.
But thats your oppinion of it being a chainsaw. For alot of things its not ... its just a heavier axes.
I'll grant you that for something it is easier to approach with OO but at the same time you have a whole new set of problems to deal with.
I've just seen way to many attempts to write object oriented code that just turned out horrible. Just like every other language out there you can write some very bad code in C++.
OO is not the magic buillet but rather another tool to use when the time is right. The trick is finding a good coder to use the tool belt.
No I'm actually not.
...
The point I was going after was the paragigm of inheritence was not need for a language to be useful which is where he was going.
Oh and my crappy little company is one of the top 3 defense contractors in the US i.e. 80,000+ employes. What crappy "little" company do you work for
Oh and FYI C is and will continue (for a good long time) to be a faster language than C++ or java ... So I wouldn't say I drive a pinto but rather a 911 with a single CD player. C++ is more like a 944 with the kitchen sink sitting in the passenger seat waying you down.
Some of us can't write crap code and expect it to run well on embedded processors ...
Do you really believe that C has no value? I mean I can understand if your affraid to alloc your own memory and keep track of it but please don't assume everyone is inept!
He said no such thing I think you need to read before you post.
... give me C any day ...
He's saying that just because it is a programming language it doesn't have to have inheritance to be a vaible language. There are many good languages out there that don't have inheritance. Trust me its not the end all be all in programming languages
No thats just plain f'kin stupid ....
that's so funny ... about half way down the page is a quote from Randy.Witlicki@hanover.valley.net.
.. lets put it this way ... there is a real reason he was working at a NON-PROFIT ISP ....
I used to work at valley.net for a time