If your employer really wants you, they'll pay for that. It costs them nothing (in comparison to the other costs of having an employee, like salary, benefits, office space, etc.).
> Allowing programmers to name serious 'flagship' Linux applications is right in line with letting marketing write them.
Who cares? If you don't like the name, fork the project and rename it.
Programmers don't care about touchy-feelie things like names. They like good code that works for them. If you have different goals, then maybe you should write "flagship" applications and name them what you like.
Free software isn't about taking over the world with cute names. The people that have the coding ability to good work just don't care. Sorry. Use Windows if that's important to you. Freedom wins over superficiality.
Fuck Apple. They're the monopoly in this case, so they deserve the most punishment. I bought a number of songs from iTMS, and that was fine when I had a Mac. As the quality of OS X declined with each release (10.4.5 killed pubcookie on two production servers...), I decided to give up on Apple and bought a Dell running Debian (from which I type this post). Unfortunately, I can't play the music that I legally bought. I have to download music from BitTorrent, harming the artists and the record companies. If I had a choice, I'd pay for music, but I don't have that option anymore. CDs aren't an option, since I only want one song.
Sorry, Apple, but you need to be open. Selling music online legally is great. Discriminating against who you sell music to isn't. (For the record, iHave an iPod, so Apple wouldn't be "losing" anything by selling to me. Only gaining.)
Whatever, I don't like American music anyway. When will I be able to buy JPop and good digital classical recordings online? When will I be able to buy uncompressed 5 channel surround recordings? That sort of stuff would really excite me, and really open my wallet;)
I did slashdot vs. DMCA and slashdot won by a lot. Incidentally, I noticed that the locale in the URL was "en_GB", so I changed it to "en_US" (since the DMCA is a US thing), but all the results were in french. I guess locale != en_GB means locale = fr_FR. Great programmers... (what can you expect from a flash animation that shows punching, though...)
For licensing reasons, we were concerned about distributing code to play MP3 files.
In a future release, however, we hope to provide an interface for you to select your own MP3-playing software.
Fuck MP3, then. Use Vorbis, which is Free of royalties, patents, etc.
Q: Why are movie files so big?
Due to licensing issues with movie codecs, we can't include a motion-compressing codec for making movies. As a result, we can only produce movies that are uncompressed.
If you can obtain a license to use a better codec on your Linux system, we recommend that you use that licensed software to compress the resulting movie files; they should drastically drop in size with any such codec.
Fuck video codecs that require licensing, then. Use Theora, which is Free of royalties, patents, etc.:)
All in all, Google didn't do their homework here, I think. There are plenty of ways to overcome the mentioned restrictions. If it were open source, I'd write the patch myself.
Sadly, it's all closed up, and is useless to everybody. Too bad.
Google should definitely fork F-Spot and call it G-Spot.
Also, someone at Google needs to be told about Ogg/Vorbis and Ogg/Theora. They claim "for licensing reasons" that their software can't compress movies or play music under Linux. Bullshit -- Vorbis and Theora are completely Free, even for proprietary applications like Picasa.
I was also going to post a recommendation for f-spot. Interesingly, f-spot has a Windows flavor as well -- it's written in.NET (erm, Mono) and is "f-spot.exe"!
(I know, it's from Novell and Mono is Miguel de Icazza's little pet project. It's a fine app and runs plenty fast for me to not care one way or the other. Mono is Free, f-spot is Free, and the OS is runs on is Free, so that's all I care about.)
> I still don't see any checking of the contents of $bar or $baz, or if $baz contains "; DELETE *", don't you still suffer the usual problems?
No, because there is no "quoting" going on. Instead of making a string that you'd type into the SQL shell, you're directly telling the database what the query is. If $bar is "'\\\'"\''\""/\//'""''\DROP database foo'\""''\\'', then the database will be told to search for that exact string in the database, not to do whatever that long thing means when you type it into the shell.
The reason many programmers "prefer" the first way is because they do their test queries at the database console, and assume that the database is programmed in the same manner. It's not.
The first rule of secure programming is "Don't quote." and if you do DBI the right way, you will never have to.
Hmm, slashdot won't accept this post due to to many "junk characters". So I guess that gives me an opportunity to explain the second rule of secure programming -- "Don't parse."
The problem with parsing is that you create ambigious situations. If you have a "key: valuekey2: value2" format, what happens when you want to have a value that's "value not a key: not another value"? Difficult situation, and solving it violates the first rule!
Solution? Let something outside your config file separate the keys and values. For example, you can make a configuration directory and make each key a file. Then the filesystem separates your keys from your values. (This is what DJB does in qmail.)
Hope this helps:)
Gah! Still too many junk characters. No wonder there's never intelligent commentary on slashdot.
Why Fedora? Fedora is designed to be a desktop system, not an embedded system. They really should start with something small (might I suggest Debian), and then add what they need.
Plus, if the kids had Debian they could later buy a bigger disk and apt-get install whatever they need!
GCJ runs Eclipse fine, and Eclipse is a pretty big app. I do all my Java development with GCJ, and have never had any problems. As a long-time C programmer, using GDB as the debugger is nice, too.
All in all, GCJ is excellent and will only get better. It won't support Sun's API du-jour, but who cares about Sun?
> didn't realize that dates of the form 2006-05-01 needed quotes
There's a difference between fundamentally flawed thinking and a bug. Someone found this bug, it was fixed in 30 seconds, and suddenly everyone using DBI had a more robust solution. Software's never going to be "perfect", but it's likely that a group of people trying to solve a problem in general is going to have a better product than something you came up with after lunch:)
It especially bugs me because it's easier to Do Things Right. The DBI manpage for perl doesn't even mention the sloppy way that nearly everyone uses... but they do it anyway! In nearly every database application / script I look at, people do things like $dbh->execute("SELECT * FROM foo WHERE bar=$bar AND baz=$baz") after "escaping" $bar and $baz. No, no, no!
It's much easier to prepare a query handle and then execute it as needed: $sth = $dbh->prepare("SELECT a,b,c FROM foo WHERE bar=? and baz=?") $sth->execute($bar, $baz);
Not only is it more efficient (if you're going to use the same query twice), it's secure by default. Let the database programmers handle the Hard Stuff (parsing) so that you can concentrate on your application.
Speaking of which, is there a way to do this in PHP? I've never seen a PHP script that did anything like this (which is probably why bugtraq is 99% php SQL injection holes).
> There is no right to purchase tickets for a concert or event. I seriously don't see why there should be any controversy over this - if Ticketmaster (or anybody) can get $1,500 for a ticket then they should be allowed to get $1,500 for a ticket.
Interestingly, many venues for which Ticketmaster sells tickets are public property, bought for some rich bigwigs by the taxpayers. Case in point in is "US Cellular Field" in Chicago. Paid for by the taxpayers, but no taxpayer could afford World Series tickets last year.
Skype claims to use 256-bit AES to encrypt your call. I say "claims" since there's no proof that they do. The code is closed, so you can't check for yourself. For all you know, it's a direct line to TIA or whatever.
Simple solution. Fork Firefox and port the fixes. Then release a stable version. Problem solved.
I'm sure if the code the Firefox people had functioned perfectly, they'd release it in the next point release. (And if they don't, fuck Firefox, and write your own web browser.)
Re:The diplomatic (accurate) response
on
The CVS Cop-Out
·
· Score: 1
Umm, just get the version from HEAD and you'll probably get some other fixes too.
Rendezvous has nothing to do with physical proximity. What they could do, though, is associate two machines via bluetooth and see how the link quality changes as the machine move nearer to each other. You could do this with 802.11 also, but the proprietary drivers won't give you enough information to do anything useful. This could be solved by using the Linux drivers for the Intel Wireless, which are open source, and give you good information about the link quality. (Do the MacBooks use Intel Wireless, or did Apple choose some other vendor?)
> Why should an M-14 target rifle, which has the detachable box magazine, semi-automatic fire, a portruding pistol grip, and a flash suppressor/muzzle brake be banned?
Ummmmm... dunno. Maybe because it has a detachable box magazine, semi-automatic fire, a portruding [sic] pistol grip, and a flash suppressor/muzzle brake!!?
My Core Duo Dell runs nice and cool, even when I'm compiling something and not running cpufreqd. It's possible to keep laptops cool, but then they don't look as cool:)
Slashdot Mirror
> And how does one save up for that?
If your employer really wants you, they'll pay for that. It costs them nothing (in comparison to the other costs of having an employee, like salary, benefits, office space, etc.).
> Allowing programmers to name serious 'flagship' Linux applications is right in line with letting marketing write them.
Who cares? If you don't like the name, fork the project and rename it.
Programmers don't care about touchy-feelie things like names. They like good code that works for them. If you have different goals, then maybe you should write "flagship" applications and name them what you like.
Free software isn't about taking over the world with cute names. The people that have the coding ability to good work just don't care. Sorry. Use Windows if that's important to you. Freedom wins over superficiality.
Fuck Apple. They're the monopoly in this case, so they deserve the most punishment. I bought a number of songs from iTMS, and that was fine when I had a Mac. As the quality of OS X declined with each release (10.4.5 killed pubcookie on two production servers...), I decided to give up on Apple and bought a Dell running Debian (from which I type this post). Unfortunately, I can't play the music that I legally bought. I have to download music from BitTorrent, harming the artists and the record companies. If I had a choice, I'd pay for music, but I don't have that option anymore. CDs aren't an option, since I only want one song.
;)
Sorry, Apple, but you need to be open. Selling music online legally is great. Discriminating against who you sell music to isn't. (For the record, iHave an iPod, so Apple wouldn't be "losing" anything by selling to me. Only gaining.)
Whatever, I don't like American music anyway. When will I be able to buy JPop and good digital classical recordings online? When will I be able to buy uncompressed 5 channel surround recordings? That sort of stuff would really excite me, and really open my wallet
I did slashdot vs. DMCA and slashdot won by a lot. Incidentally, I noticed that the locale in the URL was "en_GB", so I changed it to "en_US" (since the DMCA is a US thing), but all the results were in french. I guess locale != en_GB means locale = fr_FR. Great programmers... (what can you expect from a flash animation that shows punching, though...)
Fuck MP3, then. Use Vorbis, which is Free of royalties, patents, etc.
Fuck video codecs that require licensing, then. Use Theora, which is Free of royalties, patents, etc.
All in all, Google didn't do their homework here, I think. There are plenty of ways to overcome the mentioned restrictions. If it were open source, I'd write the patch myself.
Sadly, it's all closed up, and is useless to everybody. Too bad.
Google should definitely fork F-Spot and call it G-Spot.
Also, someone at Google needs to be told about Ogg/Vorbis and Ogg/Theora. They claim "for licensing reasons" that their software can't compress movies or play music under Linux. Bullshit -- Vorbis and Theora are completely Free, even for proprietary applications like Picasa.
I was also going to post a recommendation for f-spot. Interesingly, f-spot has a Windows flavor as well -- it's written in .NET (erm, Mono) and is "f-spot.exe"!
(I know, it's from Novell and Mono is Miguel de Icazza's little pet project. It's a fine app and runs plenty fast for me to not care one way or the other. Mono is Free, f-spot is Free, and the OS is runs on is Free, so that's all I care about.)
> I still don't see any checking of the contents of $bar or $baz, or if $baz contains "; DELETE *", don't you still suffer the usual problems?
:)
No, because there is no "quoting" going on. Instead of making a string that you'd type into the SQL shell, you're directly telling the database what the query is. If $bar is "'\\\'"\''\""/\//'""''\DROP database foo'\""''\\'', then the database will be told to search for that exact string in the database, not to do whatever that long thing means when you type it into the shell.
The reason many programmers "prefer" the first way is because they do their test queries at the database console, and assume that the database is programmed in the same manner. It's not.
The first rule of secure programming is "Don't quote." and if you do DBI the right way, you will never have to.
Hmm, slashdot won't accept this post due to to many "junk characters". So I guess that gives me an opportunity to explain the second rule of secure programming -- "Don't parse."
The problem with parsing is that you create ambigious situations. If you have a "key: valuekey2: value2" format, what happens when you want to have a value that's "value not a key: not another value"? Difficult situation, and solving it violates the first rule!
Solution? Let something outside your config file separate the keys and values. For example, you can make a configuration directory and make each key a file. Then the filesystem separates your keys from your values. (This is what DJB does in qmail.)
Hope this helps
Gah! Still too many junk characters. No wonder there's never intelligent commentary on slashdot.
Why Fedora? Fedora is designed to be a desktop system, not an embedded system. They really should start with something small (might I suggest Debian), and then add what they need.
Plus, if the kids had Debian they could later buy a bigger disk and apt-get install whatever they need!
> The entire Java industry relies on those APIs du jour.
And whose fault is that?
> and you're not using them as the beginning of a sentence
Umm... yes, he is.
GCJ runs Eclipse fine, and Eclipse is a pretty big app. I do all my Java development with GCJ, and have never had any problems. As a long-time C programmer, using GDB as the debugger is nice, too.
All in all, GCJ is excellent and will only get better. It won't support Sun's API du-jour, but who cares about Sun?
> NEWSFLASH: Sony and the RIAA are not law enforcement agencies.
What country do you live in?
> didn't realize that dates of the form 2006-05-01 needed quotes
:)
There's a difference between fundamentally flawed thinking and a bug. Someone found this bug, it was fixed in 30 seconds, and suddenly everyone using DBI had a more robust solution. Software's never going to be "perfect", but it's likely that a group of people trying to solve a problem in general is going to have a better product than something you came up with after lunch
It especially bugs me because it's easier to Do Things Right. The DBI manpage for perl doesn't even mention the sloppy way that nearly everyone uses... but they do it anyway! In nearly every database application / script I look at, people do things like $dbh->execute("SELECT * FROM foo WHERE bar=$bar AND baz=$baz") after "escaping" $bar and $baz. No, no, no!
It's much easier to prepare a query handle and then execute it as needed:
$sth = $dbh->prepare("SELECT a,b,c FROM foo WHERE bar=? and baz=?")
$sth->execute($bar, $baz);
Not only is it more efficient (if you're going to use the same query twice), it's secure by default. Let the database programmers handle the Hard Stuff (parsing) so that you can concentrate on your application.
Speaking of which, is there a way to do this in PHP? I've never seen a PHP script that did anything like this (which is probably why bugtraq is 99% php SQL injection holes).
> There is no right to purchase tickets for a concert or event. I seriously don't see why there should be any controversy over this - if Ticketmaster (or anybody) can get $1,500 for a ticket then they should be allowed to get $1,500 for a ticket.
Interestingly, many venues for which Ticketmaster sells tickets are public property, bought for some rich bigwigs by the taxpayers. Case in point in is "US Cellular Field" in Chicago. Paid for by the taxpayers, but no taxpayer could afford World Series tickets last year.
Your government at work for you, as always.
Skype claims to use 256-bit AES to encrypt your call. I say "claims" since there's no proof that they do. The code is closed, so you can't check for yourself. For all you know, it's a direct line to TIA or whatever.
Oh, you mean things like this:
. 6.x86/OpenSSH-57/
http://www.opensource.apple.com/darwinsource/10.4
Looks like stock OpenSSH to me.
So Apple calling their version of FreeBSD OS X is plagarism then?
Simple solution. Fork Firefox and port the fixes. Then release a stable version. Problem solved.
I'm sure if the code the Firefox people had functioned perfectly, they'd release it in the next point release. (And if they don't, fuck Firefox, and write your own web browser.)
Umm, just get the version from HEAD and you'll probably get some other fixes too.
Rendezvous has nothing to do with physical proximity. What they could do, though, is associate two machines via bluetooth and see how the link quality changes as the machine move nearer to each other. You could do this with 802.11 also, but the proprietary drivers won't give you enough information to do anything useful. This could be solved by using the Linux drivers for the Intel Wireless, which are open source, and give you good information about the link quality. (Do the MacBooks use Intel Wireless, or did Apple choose some other vendor?)
Fortunately B::Deparse deobfuscates the code instantly, resulting in:
/\./; /[wW]in/ ? 'cls' : 'clear';
:)
@A = (25, 0);
@B = (0, 24);
@C = (49, 24);
@X = ($") x 49;
@_ = (@X, $/) x 25;
$_[$A[1] * 50 + $A[0]] = '.';
$_[$B[1] * 1 * 50 + $B[0]] = '.';
$_[$C[1] * 50 + $C[0]] = '.';
@X = ($C[0], $C[1]);
'???';
while (394 > join($", @_) =~ tr/.//) {
do {
$R = 3 * rand;
@X = (int((${('A', 'B', 'C')[$R];}[0] + $X[0]) / 2 + 0.5), int((${('A', 'B', 'C')[$R];}[1] + $X[1]) / 2 + 0.5 + 0))
} while $_[$Z = $X[1] * 50 + $X[0] + 0] =~
$_[$Z] = '.';
system $^O =~
print @_;
}
which is actually pretty easy to understand
> Why should an M-14 target rifle, which has the detachable box magazine, semi-automatic fire, a portruding pistol grip, and a flash suppressor/muzzle brake be banned?
Ummmmm... dunno. Maybe because it has a detachable box magazine, semi-automatic fire, a portruding [sic] pistol grip, and a flash suppressor/muzzle brake!!?
My Core Duo Dell runs nice and cool, even when I'm compiling something and not running cpufreqd. It's possible to keep laptops cool, but then they don't look as cool :)