...I'd also suggest taking a look at wxWindows. They're open source, the results look very mac native (they have screenshots), and it seems to be very portable....having libraries for windows, most *nux (using GTK, Xt, or Motif), and macs, and maybe a few others. Oh, and it's c++ for those with an objective-c phobia, like me:)
I honestly don't know why more isn't done with
this framework outside of those crazy python people. It looks good, and is completely free (both as in beer and speech), unlike Qt
(not trolling, just stating a fact!)
Not 100% sure (because I'm too lazy to look at the trolltech website) but I think he was refering to the fact that getting the QT developer licence is not free for mac os x, unlike in linux, where they give free license. This has nothing to do with the too often uninstalled extra CD apple ships with their operating system.
I define large as, well, large. I'm a scientist. Everyone I know in my field (computational chemistry) has run code on intel/amd machines, and many do so as their primary source of computer power.
And I have seen a cray. I've run code on the university's cray. Guess what, my university is getting rid of its cray to buy a massive linux cluster. Check any university you please and I'll lay 50-50 odds that they have recently or will soon purchase a 100+ node commodity cluster. Hell, even my puny research group is buying a 50 processor cluster. There's just too much power for too cheap to keep supporting expensive hardware for anything but things that just don't distribute well.
I never said vector processing was dead, I just said it's not safe to assume they're the only game in town anymore for serious science.
2)Unlike C, Fortran 90 (NOT Fortran 77!) has parallel features
built in language, not libraries. Loops can
be parallelized without programmer interaction, for example. It makes creation
of parallel programs easier.
1. There are many c compilers that have autoparallization features. The IBM compiler (xlc) that shipped with my university's IBM regatta has such a feature. The C compiler I use on the Origin 2000's at NCSA also has it. It's dangerous to use, because it mostly sucks.
2. OpenMP is a compiler specification, so writing SMP code in C is supported by the compiler - no libraries needed for basic parallelization. Even FORTRAN has to resort to libraries to get parallization on a distributed system (ie a cluster).
Not to be rude or anything, but I do get it. "Unrolling" inner loops is a risky thing and works sometimes and not others. I've seen autoparallization fail miserably. Far better to find critical areas yourself and use openMP to do it for you, anyway.
As for the FORTRAN fluency thing. I graduated in 1999 with a degree in chemistry. I work now in the computational chemistry field. Almost everything is written in fortran. I was forced to take fortran, and I agree it's a good thing. But I much prefer C for the reasons I gave above.
Because C is also a very parallelizable language with hooks for both MPI and openMP and even PVM if you really want. Because automatic parallelization sucks, even in fortran. Because I can either find numerical libraries in C or I can link with the fortran ones. Because at the end of the day, I can read C easier than I can fortran and with correct use of the compiler, I can get it as (nearly as?) fast as fortran. The only reason fortran is still around in science is legacy code (which is fine by me, I don't want to port it) but new code should be written in what the programmer is more comfortable with debugging. And I disagree strongly that writing even numerical code is easier in fortran... for me.
And FORTRAN is a niche language. FORTRAN is only really useful for number crunching. If you learn C you can do it all.
Slashdot Mirror
Well, it runs right now in x windows. You can get the evolution package from fink.sourceforge.net. I've run it, and it works (mostly) fine.
Actually, I did get it, and greped extensively. No dice. Don't be so damned quick to judge, git-brained coward.
Uhhhhh....everybody has access to the CVS repository. Look here
Four words: "It's still in beta."
...I'd also suggest taking a look at wxWindows. They're open source, the results look very mac native (they have screenshots), and it seems to be very portable....having libraries for windows, most *nux (using GTK, Xt, or Motif), and macs, and maybe a few others. Oh, and it's c++ for those with an objective-c phobia, like me :)
I honestly don't know why more isn't done with this framework outside of those crazy python people. It looks good, and is completely free (both as in beer and speech), unlike Qt (not trolling, just stating a fact!)
Not 100% sure (because I'm too lazy to look at the trolltech website) but I think he was refering to the fact that getting the QT developer licence is not free for mac os x, unlike in linux, where they give free license. This has nothing to do with the too often uninstalled extra CD apple ships with their operating system.
Sheepishly...
Uhhhhhh.... I meant char crap[255];
Ironic, eh?
Buffer overflows are features! What else do you do it you've forgotten the root password?
Not new, you're right. pine (the old classic command line, well ncurses, mail client) has a bounce feature.
I define large as, well, large. I'm a scientist. Everyone I know in my field (computational chemistry) has run code on intel/amd machines, and many do so as their primary source of computer power.
And I have seen a cray. I've run code on the university's cray. Guess what, my university is getting rid of its cray to buy a massive linux cluster. Check any university you please and I'll lay 50-50 odds that they have recently or will soon purchase a 100+ node commodity cluster. Hell, even my puny research group is buying a 50 processor cluster. There's just too much power for too cheap to keep supporting expensive hardware for anything but things that just don't distribute well.
I never said vector processing was dead, I just said it's not safe to assume they're the only game in town anymore for serious science.
2)Unlike C, Fortran 90 (NOT Fortran 77!) has parallel features built in language, not libraries. Loops can be parallelized without programmer interaction, for example. It makes creation of parallel programs easier.
1. There are many c compilers that have autoparallization features. The IBM compiler (xlc) that shipped with my university's IBM regatta has such a feature. The C compiler I use on the Origin 2000's at NCSA also has it. It's dangerous to use, because it mostly sucks.
2. OpenMP is a compiler specification, so writing SMP code in C is supported by the compiler - no libraries needed for basic parallelization. Even FORTRAN has to resort to libraries to get parallization on a distributed system (ie a cluster).
Not to be rude or anything, but I do get it. "Unrolling" inner loops is a risky thing and works sometimes and not others. I've seen autoparallization fail miserably. Far better to find critical areas yourself and use openMP to do it for you, anyway.
As for the FORTRAN fluency thing. I graduated in 1999 with a degree in chemistry. I work now in the computational chemistry field. Almost everything is written in fortran. I was forced to take fortran, and I agree it's a good thing. But I much prefer C for the reasons I gave above.
...which assumes that your processor has vector units, and that your compiler is smart enough to know about them.
A large amount of science is now done on commodity intel pentium chips with gcc (g77). Not a big win, I'd say.
Because C is also a very parallelizable language with hooks for both MPI and openMP and even PVM if you really want. Because automatic parallelization sucks, even in fortran. Because I can either find numerical libraries in C or I can link with the fortran ones. Because at the end of the day, I can read C easier than I can fortran and with correct use of the compiler, I can get it as (nearly as?) fast as fortran. The only reason fortran is still around in science is legacy code (which is fine by me, I don't want to port it) but new code should be written in what the programmer is more comfortable with debugging. And I disagree strongly that writing even numerical code is easier in fortran ... for me.
And FORTRAN is a niche language. FORTRAN is only really useful for number crunching. If you learn C you can do it all.
Macs are now unix, don't you know.
Just joshin'!
Not that I disagree with the jist of what you're saying, but fink does not a unix make.
Cygwin sits on top of windows and provides most of what fink does (minus the snazzy apt-get functionality) but no one would argue windows is unix.
Mac OS X is a unix IMHO, but fink is not the reason.