I rarely take sides in what end up to be political arguments about programming languages. I like them all.
That said, as a pure statement of fact, both Java and COBOL are object-oriented. There's Object Oriented COBOL, and it has been available in IBM's COBOL compilers for more than a decade now. So are C++ and several other programming languages, for that matter. Here's my simplified viewpoint: if path length (i.e. CPU execution time per task; highly correlated with the number of instructions) is your sole consideration, if you are considering COBOL and Java, if you have equally talented programmers, if you have precisely equal functionalities that those programmers are being asked to create de novo (not trying to port from one or the other, which is usually disastrous), if both programming languages can reasonably be used to create those functionalities, if you are applying the best optimizations available to each (most current technologies, best use of compiler optimizations, best run-time tuning, etc.), and if you have reasonably equivalent run-time environments (both monitored the similar ways, for example), if the run-time environments are extremely reductive so as to be as similar as possible (e.g. batch running directly on the operating system, to take out any run-time "noise"), if you give the programmers total freedom on manner of implementation (as long as the function is delivered correctly), and if the programmers know in advance that they are going to be evaluated entirely based on path length ... if all of that is true, then, *on average*, *today*, the COBOL code will almost certainly have a shorter path length than the Java code. (That's the long version of Kirk's placing his bets on COBOL in that specific contest.) But is path length your sole consideration? If so, I suppose it's possible that neither COBOL nor Java is the correct choice. (There are still a few exceptionally talented programmers who can occasionally beat today's optimizing compilers and JITCs, although it's increasingly difficult.) Note that path length does not equal price, for a variety of reasons. The most you can say is that path length and price are sometimes correlated. Path length is almost never the sole consideration in the real world. I haven't heard any CEOs talk about how their companies are going to seek shorter path lengths in the coming fiscal year. And are the above "what ifs?" realistic in particular situations? Probably not, or at least not all of them. So, if you're looking for a path length answer, I guess you've got one ... but why is either the question or the answer *particularly* interesting? Slightly interesting (to some), I'll grant. :-) - - - - - Timothy Sipples Resident Enterprise Architect STG Value Creation & Complex Deals Team IBM Growth Markets (Based in Singapore) E-Mail: [email protected] ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
