>Who do you think the people evangelising RoR are? Do you think they are actually >people who have only learned Ruby, so they don't know any better in trying to get >other to try it? For some reason, I doubt it... While I don't necessarily have any >hard evidence on user profiles...
Well, one of the biggest RoR evangelists is Dave Thomas, author of The Pragmatic Programmer...
I'd assume all passenger/cargo planes work like this now... My dad flies a Dash 8 and has pretty much the same functionality. Pre-flight he enters everything into the computer, the autopilot does everything while the plane is above 2000 feet.
That's a great practice for most projects, but is it realistic for a kernel? I'd guess that a lot of bugs result from multithreading, obscure hardware, etc, it makes me think that unit testing wouldn't be realistic. Or maybe everything gets tested except stuff in the arch directory...
I'd love to know how one does automated testing when you're working with bare metal, if any slashdotter knows!
> Unit tests aren't to make sure that your code is working correctly right now
Well, if you're doing test-first development, you're writing unit tests to make sure your code running correctly right now. And you get the long-term benefits for free:)
When I compile a program, I put everything in it's own directory in/opt (use the --prefix option when you run the configure script) It makes for some long PATH and MANPATH variables, but it works for me Since there are only 20-ish extra programs that I need, I find it quite managable
But how can they screw open source software? I don't see how they can destroy OSS through collaboration unless they manage to convince some GPLers to go BSD, dual license, or sing over copyrights.
Off the top of my head, these are the albums I own that utilize the moog: Hugo Montenegro - Moog Power Hugo Montenegro - Neil's Diamonds Return To Forever - Romantic Warrior Edan - Beauty And The Beat
There is a distinct sound with the moog. I'm not a musician, so maybe it's an effect I'm making up, but I notice it.
> Put a function called hello() in your CherryPy application, and the return value of > that function is displayed in your web browser when you visit http://address/hello
Rails and Struts already have this feature.
(Of course, for Struts you need a some XML for every class you want to act this way.)
Where in the summary or the article does it say that the retailers are actually changing the customer's reviews? Seems like it's just about accepting/rejecting reviews.
You get the same benefits doing test-first development... But on top of that, you *must* update the tests when you update the code (unless the code change is trivial).
I fudged my stoichiometry results during a couple of my labs in Grade 11. Operating a pipet is hard when you're trying to check out the hot ladies in front of you.
Ask a profiler. Write the clear code, and profile it. If there's a performance problem, then optimize. If you don't think your optimizations will show up when profiling your app, then your optimizations are worthless.
Slashdot Mirror
>Who do you think the people evangelising RoR are? Do you think they are actually
>people who have only learned Ruby, so they don't know any better in trying to get
>other to try it? For some reason, I doubt it... While I don't necessarily have any
>hard evidence on user profiles...
Well, one of the biggest RoR evangelists is Dave Thomas, author of The Pragmatic Programmer...
I'd assume all passenger/cargo planes work like this now... My dad flies a Dash 8 and has pretty much the same functionality. Pre-flight he enters everything into the computer, the autopilot does everything while the plane is above 2000 feet.
I'm from the south, you insensitive clod!
That's a great practice for most projects, but is it realistic for a kernel? I'd guess that a lot of bugs result from multithreading, obscure hardware, etc, it makes me think that unit testing wouldn't be realistic. Or maybe everything gets tested except stuff in the arch directory...
I'd love to know how one does automated testing when you're working with bare metal, if any slashdotter knows!
> Unit tests aren't to make sure that your code is working correctly right now
:)
Well, if you're doing test-first development, you're writing unit tests to make sure your code running correctly right now. And you get the long-term benefits for free
When I compile a program, I put everything in it's own directory in /opt (use the --prefix option when you run the configure script)
It makes for some long PATH and MANPATH variables, but it works for me
Since there are only 20-ish extra programs that I need, I find it quite managable
Yeah, his flash clone of an old video game is something to get all high and mighty about
My eyes are rolling at an amazing speed
But how can they screw open source software? I don't see how they can destroy OSS through collaboration unless they manage to convince some GPLers to go BSD, dual license, or sing over copyrights.
A collection of electropop records. Most of them feature the Moog as the main instrument.
Off the top of my head, these are the albums I own that utilize the moog:
Hugo Montenegro - Moog Power
Hugo Montenegro - Neil's Diamonds
Return To Forever - Romantic Warrior
Edan - Beauty And The Beat
There is a distinct sound with the moog. I'm not a musician, so maybe it's an effect I'm making up, but I notice it.
Still looking for a copy of Moog Indigo on vinyl.
You can this with Rails too. There's a lot you can do with routes.rb since really, it's just code, not a configuration file.
> Put a function called hello() in your CherryPy application, and the return value of
> that function is displayed in your web browser when you visit http://address/hello
Rails and Struts already have this feature.
(Of course, for Struts you need a some XML for every class you want to act this way.)
Basewars was much better.
http://www.loudthinking.com/arc/000479.html
In particular... "The point is that the cost per request is plummeting, but the cost of programming is not."
And the cost of deployment is not either. Or any other of the people-related tasks in development.
Where in the summary or the article does it say that the retailers are actually changing the customer's reviews? Seems like it's just about accepting/rejecting reviews.
You get the same benefits doing test-first development... But on top of that, you *must* update the tests when you update the code (unless the code change is trivial).
You honestly think performance will be an issue on a 3 ghz machine?
LOL, in fact LMAO
It interrupts me from my work every hour :-)
I fudged my stoichiometry results during a couple of my labs in Grade 11. Operating a pipet is hard when you're trying to check out the hot ladies in front of you.
I never knew of this... Thanks!
I always thought it should have been .cum.
> (like the fool I worked with who thought he'd save disk space on a Win 3.x machine by setting up the swap space on the server)
THAT IS AWESOME
I would suggest learning J instead. It's essentially APL, but replaces all of APL's special characters with ASCII characters.
rofl
Ask a profiler. Write the clear code, and profile it. If there's a performance problem, then optimize. If you don't think your optimizations will show up when profiling your app, then your optimizations are worthless.