C++ Supremacy Over Java or MS Supremacy over Borland
-
Hey guys... I've been assigned to write a technical paper on any particular subject concerning C++ for one of my college courses. I haven't really decided what the topic of the paper should be so I'm wondering if anyone here has any suggestions. (I really don't want to end up writing anything about the history of C++, so that narrows things down a little.) My instructor seems to be anti-Microsoft and I think favors Java over C++. I probably can't change his mind, but I've been thinking about writing something that'll make him think of these two things in a different light. Unfortunately I am incredibly biased in this area and therefore tend to know a great deal about C++ and MS and very little about other compilers and Java. (I have looked at Java to some degree and ended up with a bad taste in my mouth.) Could anyone direct me to some reference material? Also, I'd like to hear everyone's opinion on C++, Java, MS, Borland, etc. I appreciate your help on this... -Michael Long Live The Code Project!
Sparticus wrote: My instructor seems to be anti-Microsoft and I think favors Java over C++. You won't convince your teacher. My bet is that he's coming from the Visual Basic department and couldn't write software without garbage collectors even if his life depended on it. Once upon a time I sort of liked Java. That was of course before I was writing software professionally. Now that I'm stuck in Java hell, all I can say is that Java is a crap language and environment which forces you as much as possible to follow some twits wet OO-dreams. OO is fine and dandy most of the time, but this is too damn much. My general questions to the Sun team are:
- Where are my friggin types? Object has no place in containers. I store strings, ints, my own classes, etc in my containers. I don't store no stinking objects!
- What's wrong with operator overloading? (If I have to implement Comparable/Comparator again before using Map I'm gonna shoot somebody).
- What's wrong with free functions? Hey, just shove that communism OO down my throat - I have no free will - I cannot take responsibility for my own actions.
- What's wrong with decent performance? Spare me the lectures about "it's all about the algorithms". A O(n2) language doesn't make my efforts worthwhile.
- What's up with that slow and UGLY GUI?
- Why do you keep two semi-interdependent GUI toolkits - one which blows and one which blows even more. AWT is obsolote - so why not just dump the crap and refer to JDK 1.1 or something?
- Why is the garbage collector retarded? I mean hello?? Do you HAVE to wait until all 512 megs are allocated and THEN start to reclaim memory with a realtime thread??
Gah. I woke up grumpy today, so take my ramblings with a pinch of salt. -- Please state the nature of your medical emergency.
-
I'll give you a hint: If you don't want to flunk the course, agree with whatever the instructor believes! In the real world, you try and choose the appropriate tool for the job. For some jobs, Java will be the right choice. For others, C++ is it. Unfortunately, most college professors have never worked in the real world. As a result, they tend to have this black-and-white view of things, where they say "C++ is useless", or "all software should be written in Java", or some other crap. They are perfectly safe making those kinds of blanket statements sitting in their ivory tower, where the only consequence is the possible scorn of their peers. The feel perfectly free in enforcing whatever cockamamie view they have, regardless of its implications in an actual application. Enforcement, in this case, means the student that disagrees with them flunks.
"Think of it as evolution in action." - 'Oath of Fealty' by Larry Niven and Jerry Pournelle
Gary Wheeler wrote: Enforcement, in this case, means the student that disagrees with them flunks. ...which is exactly why I had to learn Algol-68 in my second year at Uni (despite already being proficient in Pascal and C!). X| Anna :rose: "Be yourself - not what others think you should be"
- Marcia Graesch -
Gary Wheeler wrote: I'll give you a hint: If you don't want to flunk the course, agree with whatever the instructor believes! Oh, man, you beat me to it! Didn't they teach that in Survival Skills 101? :-) Chistopher Duncan Author - The Career Programmer: Guerilla Tactics for an Imperfect World (Apress)
Christopher Duncan wrote: Didn't they teach that in Survival Skills 101 Remembering my old school days, "teacher-rubbing" wasn't a good survival technic, unless it was not important to eat quietly in university canteen :-D Ami entends-tu le vol noir des corbeaux sur nos plaînes Ami entends-tu les cris sourds du pays qu'on enchaîne Friend, do you hear the black flight of the corbels on our plains Friend, do you hear the deaf cries of the country which one chains
-
I'll give you a hint: If you don't want to flunk the course, agree with whatever the instructor believes! In the real world, you try and choose the appropriate tool for the job. For some jobs, Java will be the right choice. For others, C++ is it. Unfortunately, most college professors have never worked in the real world. As a result, they tend to have this black-and-white view of things, where they say "C++ is useless", or "all software should be written in Java", or some other crap. They are perfectly safe making those kinds of blanket statements sitting in their ivory tower, where the only consequence is the possible scorn of their peers. The feel perfectly free in enforcing whatever cockamamie view they have, regardless of its implications in an actual application. Enforcement, in this case, means the student that disagrees with them flunks.
"Think of it as evolution in action." - 'Oath of Fealty' by Larry Niven and Jerry Pournelle
Screw that. Argue whatever you want to argue, but back it up with solid evidence. If you disagree with the professor and you have nothing to backup your opinion, he'll eat you alive. I say, "Question authority, but do it with hard facts". And, yes, I did graduate with an A- average. ------------------------------------------ "Isn't it funny how people say they'll never grow up to be their parents, then one day they look in the mirror and they're moving aircraft carriers into the Gulf region?" - The Onion
-
Hey guys... I've been assigned to write a technical paper on any particular subject concerning C++ for one of my college courses. I haven't really decided what the topic of the paper should be so I'm wondering if anyone here has any suggestions. (I really don't want to end up writing anything about the history of C++, so that narrows things down a little.) My instructor seems to be anti-Microsoft and I think favors Java over C++. I probably can't change his mind, but I've been thinking about writing something that'll make him think of these two things in a different light. Unfortunately I am incredibly biased in this area and therefore tend to know a great deal about C++ and MS and very little about other compilers and Java. (I have looked at Java to some degree and ended up with a bad taste in my mouth.) Could anyone direct me to some reference material? Also, I'd like to hear everyone's opinion on C++, Java, MS, Borland, etc. I appreciate your help on this... -Michael Long Live The Code Project!
My instructor seems to be anti-Microsoft and I think favors Java over C++. I probably can't change his mind, but I've been thinking about writing something that'll make him think of these two things in a different light. Unfortunately I am incredibly biased in this area and therefore tend to know a great deal about C++ and MS and very little about other compilers and Java. (I have looked at Java to some degree and ended up with a bad taste in my mouth.) Yawn. You shouldn't accuse people of bias when you admit to not knowing the facts. You talk about changing your professor's mind (as if he was the student in this situation), but don't know enough about both sides to actually have a decent opinion of your own. You go to school to get an education afterall, so new is probably a good time to learn something new. Telling your professor (or any other person) that "Java gave you a bad taste in your mouth" is not going to convince him that C++ is better than Java. The fact of the matter is that people stick with what's familiar, and dislike what their not familiar with. Sounds like you're doing the same. ------------------------------------------ "Isn't it funny how people say they'll never grow up to be their parents, then one day they look in the mirror and they're moving aircraft carriers into the Gulf region?" - The Onion
-
Sun Java Pet Store is an educational tool. It was NEVER meant to run quickly. It's total BS that Microsoft even compares .NET to Java Pet Store. ------------------------------------------ "Isn't it funny how people say they'll never grow up to be their parents, then one day they look in the mirror and they're moving aircraft carriers into the Gulf region?" - The Onion
-
Sun Java Pet Store is an educational tool. It was NEVER meant to run quickly. It's total BS that Microsoft even compares .NET to Java Pet Store. ------------------------------------------ "Isn't it funny how people say they'll never grow up to be their parents, then one day they look in the mirror and they're moving aircraft carriers into the Gulf region?" - The Onion
All the evidence is to the contrary. Pet Store is a J2EE application and was picked precisely because Sun and Oracle brag about how great it is. Since the white paper, Sun and Oracle have spent a lot of resources trying to tune it's performance and failed miserably. Are you seriously suggesting a company would intentionally create a lame, hamstrung product as it's hallmark demonstration application? What education value is that? (Here's a bad peforming application, if you want something that runs decently, you're on your own?) If you took time to read the papers, I think you'd be persuaded otherwise.
-
I like it already... If all you do is look for evidence that you want to find, you'll always be deeply biased and frequently wrong. But, at least you'll always be satisfied in never having to change your opinion. BTW, here's another article by the same author: Java: Good and Bad[^] At least you can get some good and bad in this one. ------------------------------------------ "Isn't it funny how people say they'll never grow up to be their parents, then one day they look in the mirror and they're moving aircraft carriers into the Gulf region?" - The Onion
-
I like it already... If all you do is look for evidence that you want to find, you'll always be deeply biased and frequently wrong. But, at least you'll always be satisfied in never having to change your opinion. BTW, here's another article by the same author: Java: Good and Bad[^] At least you can get some good and bad in this one. ------------------------------------------ "Isn't it funny how people say they'll never grow up to be their parents, then one day they look in the mirror and they're moving aircraft carriers into the Gulf region?" - The Onion
-
My instructor seems to be anti-Microsoft and I think favors Java over C++. I probably can't change his mind, but I've been thinking about writing something that'll make him think of these two things in a different light. Unfortunately I am incredibly biased in this area and therefore tend to know a great deal about C++ and MS and very little about other compilers and Java. (I have looked at Java to some degree and ended up with a bad taste in my mouth.) Yawn. You shouldn't accuse people of bias when you admit to not knowing the facts. You talk about changing your professor's mind (as if he was the student in this situation), but don't know enough about both sides to actually have a decent opinion of your own. You go to school to get an education afterall, so new is probably a good time to learn something new. Telling your professor (or any other person) that "Java gave you a bad taste in your mouth" is not going to convince him that C++ is better than Java. The fact of the matter is that people stick with what's familiar, and dislike what their not familiar with. Sounds like you're doing the same. ------------------------------------------ "Isn't it funny how people say they'll never grow up to be their parents, then one day they look in the mirror and they're moving aircraft carriers into the Gulf region?" - The Onion
Again, you're right. Perhaps I should revise my question: I need to write a technical paper on anything concerning the C++ language. So far I've ruled out the history of C++ as well as 'showing' that C++ is superior to Java even if I find myself leaning in that direction due to the material provided (minus the issue of using the right tool for the right job). Does anyone have any suggestions? Is there any particular facet of C++ that seems completely fascinating to you? Maybe I should start a new thread.
-Michael Anderson-
完全な円 -
Hey guys... I've been assigned to write a technical paper on any particular subject concerning C++ for one of my college courses. I haven't really decided what the topic of the paper should be so I'm wondering if anyone here has any suggestions. (I really don't want to end up writing anything about the history of C++, so that narrows things down a little.) My instructor seems to be anti-Microsoft and I think favors Java over C++. I probably can't change his mind, but I've been thinking about writing something that'll make him think of these two things in a different light. Unfortunately I am incredibly biased in this area and therefore tend to know a great deal about C++ and MS and very little about other compilers and Java. (I have looked at Java to some degree and ended up with a bad taste in my mouth.) Could anyone direct me to some reference material? Also, I'd like to hear everyone's opinion on C++, Java, MS, Borland, etc. I appreciate your help on this... -Michael Long Live The Code Project!
And what, pray tell is Java written in? Jim Wuerch www.miwasoft.com Quote from my readme files: "This is BETA software, and as such may completely destroy your computer, change the alignment of the planets and invert the structure of the universe."
-
All the evidence is to the contrary. Pet Store is a J2EE application and was picked precisely because Sun and Oracle brag about how great it is. Since the white paper, Sun and Oracle have spent a lot of resources trying to tune it's performance and failed miserably. Are you seriously suggesting a company would intentionally create a lame, hamstrung product as it's hallmark demonstration application? What education value is that? (Here's a bad peforming application, if you want something that runs decently, you're on your own?) If you took time to read the papers, I think you'd be persuaded otherwise.
Are you seriously suggesting a company would intentionally create a lame, hamstrung product as it's hallmark demonstration application? What education value is that? (Here's a bad peforming application, if you want something that runs decently, you're on your own?) Have you ever written code as en educational tool? Have you ever optimized code? That's right: if you optimize code, you end up making it difficult to understand! I've written enough 3D graphics applications to know that making the code completely plain and understandable makes for poor performance. If you don't believe me, then go buy a beginner-level or intemediate-level 3D programming book and then ask yourself why you can't get your 3D engine to run as fast as anyone else's. What Macromedia has to say about the comparison: Soon after Sun shipped the Java Pet Store application in May of 2001, Microsoft shipped the .NET Pet Shop and made a number of claims about .NET being faster than J2EE using the respective applications as a point of comparison. The community responded (see Catfight in a Pet Store) by pointing out a few weaknesses in the Microsoft analysis. They argued that the comparison was apples and oranges. There were different test beds (hardware, operating systems, databases) and different design goals (Java Pet Store was designed to be a learning tool). There were also different architectures: the .NET Pet Shop used stored procedures with SQL; Java Pet Store was designed to be ported across application servers and databases. All of these issues clouded the analysis. http://www.macromedia.com/desdev/mx/blueprint/articles/performance.html[^] And a few interesting charts from Macromedia: .NET generally consumes more bandwidth: http://www.macromedia.com/desdev/mx/blueprint/articles/performance/chart1.gif .NET isn't any faster than Java Pet Store: http://www.macromedia.com/desdev/mx/blueprint/articles/performance/chart2.gif What Oracle said about the benchmarks: In response to Microsoft's benchmark claims, Oracle responded with a Java Pet Store-based application that reflected optimizations similar to those Microso
-
Are you seriously suggesting a company would intentionally create a lame, hamstrung product as it's hallmark demonstration application? What education value is that? (Here's a bad peforming application, if you want something that runs decently, you're on your own?) Have you ever written code as en educational tool? Have you ever optimized code? That's right: if you optimize code, you end up making it difficult to understand! I've written enough 3D graphics applications to know that making the code completely plain and understandable makes for poor performance. If you don't believe me, then go buy a beginner-level or intemediate-level 3D programming book and then ask yourself why you can't get your 3D engine to run as fast as anyone else's. What Macromedia has to say about the comparison: Soon after Sun shipped the Java Pet Store application in May of 2001, Microsoft shipped the .NET Pet Shop and made a number of claims about .NET being faster than J2EE using the respective applications as a point of comparison. The community responded (see Catfight in a Pet Store) by pointing out a few weaknesses in the Microsoft analysis. They argued that the comparison was apples and oranges. There were different test beds (hardware, operating systems, databases) and different design goals (Java Pet Store was designed to be a learning tool). There were also different architectures: the .NET Pet Shop used stored procedures with SQL; Java Pet Store was designed to be ported across application servers and databases. All of these issues clouded the analysis. http://www.macromedia.com/desdev/mx/blueprint/articles/performance.html[^] And a few interesting charts from Macromedia: .NET generally consumes more bandwidth: http://www.macromedia.com/desdev/mx/blueprint/articles/performance/chart1.gif .NET isn't any faster than Java Pet Store: http://www.macromedia.com/desdev/mx/blueprint/articles/performance/chart2.gif What Oracle said about the benchmarks: In response to Microsoft's benchmark claims, Oracle responded with a Java Pet Store-based application that reflected optimizations similar to those Microso
Brit wrote: Have you ever written code as en educational tool? None that was hamstrung. Ever. If you are going to show off a technology, in this case J2EE, even for educational purposes, you are not going to create shit. Oracle and Sun have staked a lot in J2EE and were upset about this test. They've had ample opportunity to optimize and modify it, yet these optimizations didn't hold up to a third party examination. I reference again: http://www.gotdotnet.com/team/compare/veritest.aspx[^] Given that the stock for both companies is tanking, perhaps I am expecting too much of them.:) As for the claim of apples and oranges, I completely disagree. Sun's J2EE white papers staked a claim that they can provide a total solution superior to Microsoft's--this included the hardware, operating system, database etc.--it is how real purchase evaluations are done in the real world. (Several years ago, Apple claimed that they made supercomputers. To prove this, Adobe tweaked the crap out of Photoshop for them. The results were impressive and I think totally fair because what matters is the solution and the users benefited from what Adobe had done--now had Adobe refused to release this version, then it would have been disingenuous, but still fine with me.) Both Oracle and Sun make a big deal about making claims of the superiority of their products and produce a plethora of white papers arguing the theoretical merits of their solutions against not just Microsoft, but IBM, SAP, etc. But when a third party, for example, published data showing DB2 to be faster than Oracle 9, Oracle throws a fit and starts changing the argument. They've been playing this shell game for years and are still at it. If J2EE is so great, prove it. Oracle and Sun haven't yet. Their results cannot be verified by an independent third party. And they have yet to answer some of the serious issues raised not just by the white paper in question, but other papers and articles as well. (Remember when Ellison claimed Oracle 9 was unbreakable. And it was broken. Nuf said.)
-
Hey guys... I've been assigned to write a technical paper on any particular subject concerning C++ for one of my college courses. I haven't really decided what the topic of the paper should be so I'm wondering if anyone here has any suggestions. (I really don't want to end up writing anything about the history of C++, so that narrows things down a little.) My instructor seems to be anti-Microsoft and I think favors Java over C++. I probably can't change his mind, but I've been thinking about writing something that'll make him think of these two things in a different light. Unfortunately I am incredibly biased in this area and therefore tend to know a great deal about C++ and MS and very little about other compilers and Java. (I have looked at Java to some degree and ended up with a bad taste in my mouth.) Could anyone direct me to some reference material? Also, I'd like to hear everyone's opinion on C++, Java, MS, Borland, etc. I appreciate your help on this... -Michael Long Live The Code Project!
I used to love Borland (Inprise). However since they have turned all of their C++ development efforts to C++ Builder I have stopped using them. Borland's C++ 5.x compiler was a great compiler. It supported templates and some of the other new C++ features years before Microsoft did. I believe that it was ANSI compliant as well, and if it wasnt, I know I really enjoyed working in their IDE as well. But they stopped expanding upon it. So when I started developing with ATL and COM, I decided that Visual C++ was the way to go. Borland C++ Builder 4.0 was to flakey for my tastes. I built a project that took about 9 months to develop it some of difficulties that I faced at the end were really challenging. I would not recommend using this table for large scale applications. I would use it for an IT department where the programmers prefer C++ over VB. As far as JAVA, I like it as a language, I think that it is fun to program in. But I still prefer the weak type-casting that C++ offers, and the ability for data to just be raw data rather than every data item requiring an object.
Build a man a fire, and he will be warm for a day
Light a man on fire, and he will be warm for the rest of his life! -
Again, you're right. Perhaps I should revise my question: I need to write a technical paper on anything concerning the C++ language. So far I've ruled out the history of C++ as well as 'showing' that C++ is superior to Java even if I find myself leaning in that direction due to the material provided (minus the issue of using the right tool for the right job). Does anyone have any suggestions? Is there any particular facet of C++ that seems completely fascinating to you? Maybe I should start a new thread.
-Michael Anderson-
完全な円Generic programming using Templates. There are so many things that you can accomplish and so many ways to save time by using templates. You can write paramterized data abstraction containers like in STL. You can simplify complex math equations during compile time rather than runtime. I have implemented a client / server architecture by using a set of templates to do all of the core dirty work, then using a set of specialized classes that were plugged into the middle to of the architecture to make the entire thing work.
Build a man a fire, and he will be warm for a day
Light a man on fire, and he will be warm for the rest of his life! -
Brit wrote: Have you ever written code as en educational tool? None that was hamstrung. Ever. If you are going to show off a technology, in this case J2EE, even for educational purposes, you are not going to create shit. Oracle and Sun have staked a lot in J2EE and were upset about this test. They've had ample opportunity to optimize and modify it, yet these optimizations didn't hold up to a third party examination. I reference again: http://www.gotdotnet.com/team/compare/veritest.aspx[^] Given that the stock for both companies is tanking, perhaps I am expecting too much of them.:) As for the claim of apples and oranges, I completely disagree. Sun's J2EE white papers staked a claim that they can provide a total solution superior to Microsoft's--this included the hardware, operating system, database etc.--it is how real purchase evaluations are done in the real world. (Several years ago, Apple claimed that they made supercomputers. To prove this, Adobe tweaked the crap out of Photoshop for them. The results were impressive and I think totally fair because what matters is the solution and the users benefited from what Adobe had done--now had Adobe refused to release this version, then it would have been disingenuous, but still fine with me.) Both Oracle and Sun make a big deal about making claims of the superiority of their products and produce a plethora of white papers arguing the theoretical merits of their solutions against not just Microsoft, but IBM, SAP, etc. But when a third party, for example, published data showing DB2 to be faster than Oracle 9, Oracle throws a fit and starts changing the argument. They've been playing this shell game for years and are still at it. If J2EE is so great, prove it. Oracle and Sun haven't yet. Their results cannot be verified by an independent third party. And they have yet to answer some of the serious issues raised not just by the white paper in question, but other papers and articles as well. (Remember when Ellison claimed Oracle 9 was unbreakable. And it was broken. Nuf said.)
these optimizations didn't hold up to a third party examination Macromedia is a third party. None that was hamstrung. Ever. If you are going to show off a technology, in this case J2EE, even for educational purposes, you are not going to create shit. According to Macromedia's tests, .NET really has nothing on Java Pet Store. Further, if we assume Macromedia's test are wrong, and use VeriTest's (which tried to match the setup used by Oracle, but didn't actually test their setup), then according to the numbers at godotnet, Java Pet Store is generally 2-3 times slower than .NET. I'd hardly call that "hamstrung" or "shit". If Java Pet Store is an educational tool, that is actually good performance. As an example, I took my 3D engine and was able to get 5x better performance out of it by making code optimizations. 5 times! That's not at all uncommon, either. I've seen people get 100x speed increases out of small bits of code. ------------------------------------------ "Isn't it funny how people say they'll never grow up to be their parents, then one day they look in the mirror and they're moving aircraft carriers into the Gulf region?" - The Onion
-
Generic programming using Templates. There are so many things that you can accomplish and so many ways to save time by using templates. You can write paramterized data abstraction containers like in STL. You can simplify complex math equations during compile time rather than runtime. I have implemented a client / server architecture by using a set of templates to do all of the core dirty work, then using a set of specialized classes that were plugged into the middle to of the architecture to make the entire thing work.
Build a man a fire, and he will be warm for a day
Light a man on fire, and he will be warm for the rest of his life!
