what exactly is innovative about describing a UI declaratively? surely you don't mean the fact that the surface syntax of the declaration language has angle brackets in it?
> FM transmitting- I love it, but it's not really powerful enough here in the Washington, DC area with all the background noise to be picked up more than 3 or so feet away.
install a wired FM modulator in your car. they are available for around 40$ at places like crutchfield.com, and are not too hard to install.
I'm afraid the "heavyweight" thing in the interview doesn't mean what you think it does. it means "the server needs to compute stuff in order for the client to do anything with the repository". and this approach is not a good idea for distributed development, on account of it not scaling too well. think how much more bearable SourceForge would be if it only had to host files for "official" mainlines.
GTK's object orientation actually survives past compilation time, which is very nice. this is not the case in the language you probably mistake for being object-oriented (I'm taking a somewhat-educated wild guess, yes).
(and anyway, why does the implementation language matter for you? there probably is a binding for the language you like.)
[ I'm not very good at math, but I'll try to be helpful anyway ]
if Matlab-like functionality is appropriate, then try Octave (look for it on the GNU site)
you said you don't like Emacs. well, if your dislike is strong then I guess Jacal and Mockmma are not your cup of tea. they are written in Scheme and Common Lisp respectively, so presumably they are most convenient to run with the prompt in Emacs.
So, by this logic, Debian and RedHat would be different OS's, no? (keeping in mind that various linux vendors, such as RedHat do put extra patches into the stock linux kernel)
yup, they *are* different OS'es. but: they use the same binary format, the same libc, and (with luck) will use the same standardized filesystem layout and sysinit scriptology (if not the same, then *compatible*), if something ever comes out of the LSB.
note: I'm not trying to argue in favor of or against BSD or Linux, just clarifying.
That's not a question of being a die-hard or not. The real question is why the f*&% people still invent new syntax. The beauty of Scheme (or Common Lisp, for that matter) is that you learn one basic syntax, and you can construct most any domain-specific syntax in terms of it. Yes you can have infix math in Lisp. Do your homework or something.
All this glitzy networking stuff could be a nice Scheme or Lisp extension, along with the syntax. But noooo, we have to invent a new damn language to have the networking. We'll have Perl for text processing, Lisp for heavy AI, Rebol for networking and Python for, uh, "rapid prototyping" (slow prototyping, anyone?).
Disclaimers: 1. I needed to vent. 2. I think the computing field needs more people that know what they are doing, and "easy" languages only help impostors and give people unfounded hopes (like, hey, I'm a bad-ass algorithm designer. I know Basic). 3. I don't take parentophobes seriously. If you have such problems distinguishing syntax from semantics, than you probably don't have a good grasp of the distinction between content and representation.
Slashdot Mirror
what exactly is innovative about describing a UI declaratively?
surely you don't mean the fact that the surface syntax of the declaration language has angle brackets in it?
> FM transmitting- I love it, but it's not really powerful enough here in the Washington, DC area with all the background noise to be picked up more than 3 or so feet away.
install a wired FM modulator in your car. they are available for around 40$ at places like crutchfield.com, and are not too hard to install.
I'm afraid the "heavyweight" thing in the interview doesn't mean what you think it does.
it means "the server needs to compute stuff in order for the client to do anything with the repository". and this approach is not a good idea for distributed development, on account of it not scaling too well. think how much more bearable SourceForge would be if it only had to host files for "official" mainlines.
> Would you like Van Gogh to work on your teeth?
he's more of a plastic surgeon, actually.
leave your teeth to Ecsher.
GTK's object orientation actually survives past compilation time, which is very nice. this is not the case in the language you probably mistake for being object-oriented (I'm taking a somewhat-educated wild guess, yes).
(and anyway, why does the implementation language matter for you? there probably is a binding for the language you like.)
that paradigmatic whining is on the wane, yes.
you were saying?
what is insightful about this post? the ellipses?
if Matlab-like functionality is appropriate, then try Octave (look for it on the GNU site)
you said you don't like Emacs. well, if your dislike is strong then I guess Jacal and Mockmma are not your cup of tea. they are written in Scheme and Common Lisp respectively, so presumably they are most convenient to run with the prompt in Emacs.
hth
So, by this logic, Debian and RedHat would be different OS's, no? (keeping in mind that various linux vendors, such as RedHat do put extra patches into the stock linux kernel)
yup, they *are* different OS'es.
but: they use the same binary format, the same libc, and (with luck) will use the same standardized filesystem layout and sysinit scriptology (if not the same, then *compatible*), if something ever comes out of the LSB.
note: I'm not trying to argue in favor of or against BSD or Linux, just clarifying.
Ugh.
That's not a question of being a die-hard or not. The real question is why the f*&% people still invent new syntax. The beauty of Scheme (or Common Lisp, for that matter) is that you learn one basic syntax, and you can construct most any domain-specific syntax in terms of it. Yes you can have infix math in Lisp. Do your homework or something.
All this glitzy networking stuff could be a nice Scheme or Lisp extension, along with the syntax. But noooo, we have to invent a new damn language to have the networking. We'll have Perl for text processing, Lisp for heavy AI, Rebol for networking and Python for, uh, "rapid prototyping" (slow prototyping, anyone?).
Disclaimers:
1. I needed to vent.
2. I think the computing field needs more people that know what they are doing, and "easy" languages only help impostors and give people unfounded hopes (like, hey, I'm a bad-ass algorithm designer. I know Basic).
3. I don't take parentophobes seriously. If you have such problems distinguishing syntax from semantics, than you probably don't have a good grasp of the distinction between content and representation.
There, I feel better now.