|
|
|
| Authored by: Anonymous on Thursday, May 10 2012 @ 02:11 PM EDT |
I'm sure that the "expert" knows all about hello world apps and knows
its not a game. I think *all* programmers have seen hello world examples - for
new programmers, its probably the first thing they ever did.
I'm guessing its just taken out of context in the tweet.. I wish we had more to
go on.. Last week was much more exciting! Maybe we can get someone to write
mirror_slap a note so he can take the rest of the week off from work? :)[ Reply to This | Parent | # ]
|
| |
| Authored by: tknarr on Thursday, May 10 2012 @ 02:26 PM EDT |
What's funny is that "Hello, World" really isn't about teaching programming
in a language per se. It's more about the mechanics of the environment. It's the
simplest program you can write that does enough that you can tell that it
compiled and ran and did what you expected it to do. It's used to get students
past the point of knowing how to get the development environment started, edit a
source file, write the boilerplate needed to actually make a legal program, run
the compiler and linker, run the resulting program and view the output. It's the
programming equivalent of sitting down in a car and figuring out where all the
controls are in this model, getting the seat adjusted so you're comfortable,
adjusting the mirrors and getting the key turned and the engine running. If you
can get it to run, you've gotten all the overhead dealt with and you're ready to
begin learning to program.
As a speed test, it's like doing a drag-strip
speed test of cars when the finish line is only a car's length from the start.
The distance is so short that you aren't measuring how fast the cars are, you're
measuring how fast they can start to move. "Hello, World" is so simple and short
that it really spends no time at all running, all the time is spent in the
overhead of getting the code loaded, getting the run-time environment set up,
and then getting everything shut down after the program's finished.
And
if Mitchell is any expert, he knows all this. It's standard practice when
testing performance to run a large number of iterations to get a long enough
run-time to get the errors and random variations down to an insignificant
fraction of the total. Think flipping a coin. If you only take 2 flips, it's not
uncommon for them to be both heads or both tails. In theory it should be a 50/50
split, but with only a few trials it's not unusual for it to heavily favor one
or the other. Everybody knows that, so if you want to check whether a coin's
fair or loaded you flip it a lot of times, fifty or a hundred. And even
then you don't expect it to be exactly 50/50, if you flipped it a hundred times
and got 51 heads and 49 tails you don't take that as a sign the coin's loaded in
favor of heads. In testing it's not uncommon for me to set performance tests to
run 10,000 iterations to get me enough run-time to produce a solid average, and
when the code's really fast I've had to run millions of iterations to get
numbers I'm confident of. Running a single iteration of the smallest program
possible? That's like flipping a coin once. [ Reply to This | Parent | # ]
|
| |
| Authored by: Anonymous on Thursday, May 10 2012 @ 03:05 PM EDT |
| http://www.theverge.com/2012/5/10/2991236/oracle-
google-java-
android-trial-damages-phase
Probably true even if Oracle win everything
else with wilful
infringement. It's still got to be less than the case
costs.
[ Reply to This | Parent | # ]
|
| |
| Authored by: Anonymous on Thursday, May 10 2012 @ 05:25 PM EDT |
Oracle declined to comment on the PowerPC-OpenJDK
proposal.
So much for the collaboration between
Oracle and IBM for Java huh? :p[ Reply to This | Parent | # ]
|
| |
| Authored by: Anonymous on Thursday, May 10 2012 @ 05:32 PM EDT |
Article
link.
PJ: That's copyright damages. Patents are pending, if you will
allow me to phrase it that way.
ROFL
I actually read right
passed the pun and got to the "allow me to phrase" before I realized what was
said :)
RAS[ Reply to This | Parent | # ]
|
|
|
|
|
