The main reason I don't feel like continuing on in Perl is because I once seriously shot myself in the foot due to Perl's lack of easy to use and easy to understand complex data structures. Python has classes
Were you using perl v 4?? (released in about 1992) Modern perl has classes! man perltoot.
use Net::FTP; my $ftp = Net::FTP->new('ftp.server.com', Debug => 0); $ftp->login('anonymous', 'e@mail.address'); $ftp->get('file.txt'); etc.
And error checking as well. The following is probably wrong Perl:
$variable = 1; print $varaible; But it isn't an error.
I take it you never put "#!/usr/bin/perl -w" at the top of your perl scripts? Look at man perl, and search for "-w" to see quite how many times you're told to enable warnings like this. Also put "use strict" at the beginning of your scripts.
BTW, if you do "-w", the error you get is this: Name "main::varaible" used only once: possible typo at -e line 1.
I've written a LOT of perl in the past but I switched to python because of the strange artefacts like $_.
An alternative option is to put "use English;" at the top of your script. Then you can call the $! $? $_ $% $, $; variables by their more readable names: $OS_ERROR $CHILD_ERROR $ARG $FORMAT_PAGE_NUMBER $OUTPUT_FIELD_SEPARATOR $SUBSCRIPT_SEPARATOR etc.. You can tell they're "magic" variables, cos in Perl the convention is that upper-case variables are ones which mean something special to the interpreter.
open HANDLE, "hello.txt" or die "cannot open hello.txt: $OS_ERROR".
Actually, I guess I should've said "0, 2, 4, 6, 8, 10" in those examples. Also, if the example had been this:
for i in range(0,1000000,2):
print i
Then the examples I've given break, but yours doesn't. I don't quite know why people replying to you think that a cryptic "(0,10,2)" is clearer than a percent sign, which is known for being a remainder operator. Maybe the best would be this:
Ahhggg. This brings back horrible memories of huge FORTRAN that would be flattened by one *SPACE* out of wack. It's a pain. A real pain.
The trick is to use a decent editor which has good syntax highlighting. Then, at the point where your error is, everything will turn red (or green or...). This also applies to C++ and Perl and other languages which don't rely on whitespace - it becomes much easier to spot the odd unescaped quote.
Actually, one thing that annoys me about perl is the fact that it's almost impossible for an editor to syntactically highlight it perfectly. Although this is hardly ever a problem unless you're intentionally being obfuscated (e.g. print qq]hello, world\n];, or s/([^ ])([abc])/local $_ = $2;y!abc!ABC!;$1.$_/ge;, or ${rand > 0.5 ? "b" : "a"}++;).
What people are missing in this forum is that Python is probably the most extensible language out there
Is there something available for Python like the Inline:: modules for Perl? These are modules which make it possible to stick C, C++ or Python right in the middle of Perl code. What would be really cool is if there is a way to hook up Python with any of the thousands of Perl modules which exist out there. One of the nicest thing about Perl is CPAN, a huge online repository of Perl modules for doing just about everything. It would be cool if a Python script can access, say, Perl's Net::AIM module (which provides a nice class interface to AOL Instant Messenger protocol).
Python and Perl are both really nice languages. Anyone who's never used them should give them a try. For certain tasks, it's amazing how much faster you can do things in comparison to, say, C++ or Java or VB. (But every language has its place; I don't believe the "one language fits all" claims that people sometimes make).
Python vs Perl is largely down to personal preference, and what code/modules are currently available that you can build on top of.
Kinda shoots down your central assumption ("girls do better than boys when younger, but worse later, in education").
There is evidence that a good proportion of women have a hard time coping with prejudiced environments in universities though - see this interesting Caltech article about some of the problems which they may face.
In other words, now it's time to rest on his laurels? Coast the rest of the way through high school?
It would be really nice to believe that his [high] school might be providing him with work that's stretching him to the limits and really flexing his mathematical brain muscles.
It would also be really nice to believe in Father Christmas.
I remember getting told off at [high] school, because we were supposed to show (i) that something was true for any rectangle, and then (ii) that the same thing was true for any square, and for part (ii) I just said "this is implied by (i) because a square is a rectangle". Apparently I was supposed to blindly work through the argument again.
Teachers sometimes don't realise that laziness is the mother of mathematics, and hence aren't aware that spotting generalities is important.
So no, you're right, now isn't the time for him to rest on his laurels. But if he wants to learn to be good at actual mathematics, then maybe he should buy a good book (e.g. "What is Mathematics" by Courant and Robbins, updated by Ian Stewart) and bunk maths lessons to read that instead.
To *not* allow the school to require its enrollees to abide by these policies would be a *limitation* of the freedom of individuals to contract, and is thus repugnant to the principles of our system (which we got from you, and stem in turn from the tribes the romans were never able to quite control).
Interesting that you say that - because now we have quite a lot of limitations on contracts which I believe you don't. For example, a lot of employment rights which you can sign away in the US, you can't sign away in the UK. (I.e. even if you sign a waiver, you can enforce that right later).
We've had a number of test cases here where 'Private Clubs' (the MCC, various gentlemans clubs etc.) have been taken to court over discriminatory rules (No Women, generally). AFAIK the clubs have all lost in the end.
Not quite sure where "we" is, but in the UK, there's a number of private clubs which don't admit certain people on grounds of creed or gender. E.g. Jewish-only golf clubs, and also Men-only golf clubs.
OTOH if they are employing somebody, then discrimination is illegal. E.g. a Men-only golf club can't refuse to employ women bar staff on grounds of gender.
You realize that in all the recent school shootings the students involved mentioned their plans either online or in person but were ignored because nobody was prompted to do anything because it "was just speech"
How many more dead teenagers do you want to see before people are supposed to be held accountable for their words and actions?
I see what you mean, but I think it's ridiculous to think that the solution is police investigation.
The *reason*, at least partly, that the latest shooting happened, was because the kid who shot people was getting bullied at school. The previous shooting, IIRC, was by kids who were social outcasts at school too. It would be so much simpler, so much cheaper and so much more effective to actually start *dealing* with the bullying problem in schools, than to do almost random geek profiling, website policing and speech suppression on the assumption that anyone who talks about guns is a potential murderer.
Sorry to rant like this, but it really bothers me that people go chasing stupid, ineffective "solutions", yet ignore the real bullying problem that is staring everyone in the face.
Because if the UK company wants to _SELL_ anything in the US, it better comply with all the laws and regulations of the US.
As has been pointed out, that won't apply if someone from the US is browsing a website based in the UK and gets a product shipped out.
It would apply if the UK company actually had a US branch. But by the time you are a multinational company, you probably have the resources to check out patents and stuff more easily.
Packet design is protectible (as it is a concrete implimentation) provided it is part of the whole application. Likewise with APIs.
That sounds really horrible to me, if I've read you correctly, i.e. that you mean APIs should be patentable per se. I think it's a really bad idea to allow patents on a method of communication. Then, your power to lock people into your system doesn't just stem from the programs you write, but from all the programs which anyone else writes to interact with your system. That could easily give you an unassailable lead and allow you to run a fat, bloated, consumer-unfriendly monopoly.
"1.Programs == Data. John von Neumann proved this way back in the 50s"
Huh? Somebody actually tried "proving" this?
You could argue that this is what Church's Thesis says. The usual statement of Church's Thesis is essentially that "any reasonable computer can emulate any other reasonable computer". Here, we assume that every computer has an infinite storage medium, (and we don't mean real-time emulation!).
It's not actually something which is precise enough to prove, because it's a statement about the real world and so the word "reasonable" is fuzzy. But almost all computer scientists today believe Church's Thesis to be true.
The court ruled against Napster (the company) not against OpenNap (people running servers).
This is an important point. There's a significant difference between a Napster server and an OpenNap server - the latter will index any sort of file, not just an MP3. This should give them more chance of passing the "Substantial non-infringing uses" test which Napster failed. This would render them legal.
It would be a strange situation if they decided that a version which ignores the copy control bit is illegal. No MP3 program I know of takes any notice of the copy control bit which exists in all MP3s, and that includes the ones included in Debian.
Also, being technically unable to manipulate a document is different from being legally unable to do so. There are many examples which would count as fair use; for example, a blind person might want to convert a document to ASCII to pipe through a text reader (that's just a random example off the top of my head).
OTOH they might want the patched version to be in non-US to avoid falling foul of the delightful DCMA.
they even took the trouble to write correct manual pages for the GNU compiler, which the FSF seem not to be interested in doing
That's a somewhat misinformative statement. The GCC info pages are perfectly correct. It is a bit disingenuous not to make clear they do actually write such documentation, although they use info rather than man.
Actually, the problem with Hubbard is not that his religion is drivel. It is that his religion is a very nasty cult. The fact that it is drivel is almost incidental in comparison.
Slashdot Mirror
Anyone know what the license is of the enhancements? I couldn't see anything directly on the homepage, and 655 MB over a modem is a bit unrealistic :-(
Could the graphical installer be used in Debian proper, or is it non-free?
Were you using perl v 4?? (released in about 1992) Modern perl has classes! man perltoot.
use Net::FTP;
my $ftp = Net::FTP->new('ftp.server.com', Debug => 0);
$ftp->login('anonymous', 'e@mail.address');
$ftp->get('file.txt');
etc.
I take it you never put "#!/usr/bin/perl -w" at the top of your perl scripts? Look at man perl, and search for "-w" to see quite how many times you're told to enable warnings like this. Also put "use strict" at the beginning of your scripts.
BTW, if you do "-w", the error you get is this: Name "main::varaible" used only once: possible typo at -e line 1.
An alternative option is to put "use English;" at the top of your script. Then you can call the $! $? $_ $% $, $; variables by their more readable names: $OS_ERROR $CHILD_ERROR $ARG $FORMAT_PAGE_NUMBER $OUTPUT_FIELD_SEPARATOR $SUBSCRIPT_SEPARATOR etc.. You can tell they're "magic" variables, cos in Perl the convention is that upper-case variables are ones which mean something special to the interpreter.
open HANDLE, "hello.txt" or die "cannot open hello.txt: $OS_ERROR".
The trick is to use a decent editor which has good syntax highlighting. Then, at the point where your error is, everything will turn red (or green or ...). This also applies to C++ and Perl and other languages which don't rely on whitespace - it becomes much easier to spot the odd unescaped quote.
Actually, one thing that annoys me about perl is the fact that it's almost impossible for an editor to syntactically highlight it perfectly. Although this is hardly ever a problem unless you're intentionally being obfuscated (e.g. print qq]hello, world\n];, or s/([^ ])([abc])/local $_ = $2;y!abc!ABC!;$1.$_/ge;, or ${rand > 0.5 ? "b" : "a"}++;).
Is there something available for Python like the Inline:: modules for Perl? These are modules which make it possible to stick C, C++ or Python right in the middle of Perl code. What would be really cool is if there is a way to hook up Python with any of the thousands of Perl modules which exist out there. One of the nicest thing about Perl is CPAN, a huge online repository of Perl modules for doing just about everything. It would be cool if a Python script can access, say, Perl's Net::AIM module (which provides a nice class interface to AOL Instant Messenger protocol).
Python and Perl are both really nice languages. Anyone who's never used them should give them a try. For certain tasks, it's amazing how much faster you can do things in comparison to, say, C++ or Java or VB. (But every language has its place; I don't believe the "one language fits all" claims that people sometimes make).
Python vs Perl is largely down to personal preference, and what code/modules are currently available that you can build on top of.
Cool - care to elucidate? I gotta admit I don't like the thought of basing the universe on very strong AC :-)
It would be really nice to believe that his [high] school might be providing him with work that's stretching him to the limits and really flexing his mathematical brain muscles.
It would also be really nice to believe in Father Christmas.
I remember getting told off at [high] school, because we were supposed to show (i) that something was true for any rectangle, and then (ii) that the same thing was true for any square, and for part (ii) I just said "this is implied by (i) because a square is a rectangle". Apparently I was supposed to blindly work through the argument again.
Teachers sometimes don't realise that laziness is the mother of mathematics, and hence aren't aware that spotting generalities is important.
So no, you're right, now isn't the time for him to rest on his laurels. But if he wants to learn to be good at actual mathematics, then maybe he should buy a good book (e.g. "What is Mathematics" by Courant and Robbins, updated by Ian Stewart) and bunk maths lessons to read that instead.
Not quite sure where "we" is, but in the UK, there's a number of private clubs which don't admit certain people on grounds of creed or gender. E.g. Jewish-only golf clubs, and also Men-only golf clubs.
OTOH if they are employing somebody, then discrimination is illegal. E.g. a Men-only golf club can't refuse to employ women bar staff on grounds of gender.
I see what you mean, but I think it's ridiculous to think that the solution is police investigation.
The *reason*, at least partly, that the latest shooting happened, was because the kid who shot people was getting bullied at school. The previous shooting, IIRC, was by kids who were social outcasts at school too. It would be so much simpler, so much cheaper and so much more effective to actually start *dealing* with the bullying problem in schools, than to do almost random geek profiling, website policing and speech suppression on the assumption that anyone who talks about guns is a potential murderer.
Sorry to rant like this, but it really bothers me that people go chasing stupid, ineffective "solutions", yet ignore the real bullying problem that is staring everyone in the face.
As has been pointed out, that won't apply if someone from the US is browsing a website based in the UK and gets a product shipped out.
It would apply if the UK company actually had a US branch. But by the time you are a multinational company, you probably have the resources to check out patents and stuff more easily.
You could argue that this is what Church's Thesis says. The usual statement of Church's Thesis is essentially that "any reasonable computer can emulate any other reasonable computer". Here, we assume that every computer has an infinite storage medium, (and we don't mean real-time emulation!).
It's not actually something which is precise enough to prove, because it's a statement about the real world and so the word "reasonable" is fuzzy. But almost all computer scientists today believe Church's Thesis to be true.
It would be a strange situation if they decided that a version which ignores the copy control bit is illegal. No MP3 program I know of takes any notice of the copy control bit which exists in all MP3s, and that includes the ones included in Debian.
Also, being technically unable to manipulate a document is different from being legally unable to do so. There are many examples which would count as fair use; for example, a blind person might want to convert a document to ASCII to pipe through a text reader (that's just a random example off the top of my head).
OTOH they might want the patched version to be in non-US to avoid falling foul of the delightful DCMA.
See what happened to Lisa McPherson. Sure, Christianity might have been this bad several hundred years ago. But then, that was *very* bad.
Actually, the problem with Hubbard is not that his religion is drivel. It is that his religion is a very nasty cult. The fact that it is drivel is almost incidental in comparison.