That's not true, Ancestry.com lets people have more than one entry in the tree - perhaps your experience with it is quite old. It's also relatively straightforward to add anything you want as a source, though you get extra bells & whistles if it comes from a database they've got all mapped out.
Neural networks have been investigated for years as a text classification tool, and no, they don't tend to work very well. In addition, it usually takes so long to train them that most categorization researchers don't bother too much with them anymore.
They're saying that the AltiVec Linux apps are [occasionally] 1000% faster than the non-AltiVec Linux apps. Linux is the common factor, not the difference. AltiVec is the difference, so Apple (and IBM & Motorola, who make the chips) has plenty to brag about. Their hardware is mighty fine.
I've been thinking about this too. The main ways for OSS developers to make money from their products (not that they always want to) are training & support. It seems like the parallel in the music industry is that artists like to play live, and distributing your music for free is a good way to keep people interested in you and get them to show up for the concerts.
It's really not so different from the hugely popular tape-trading networks that bands like the Grateful Dead have been using for years and years, except in that case people _do_ pay for the music (they have to buy tapes, stamps for mailing, etc.), but the band never saw a dime directly. Most people agree that tape-trading is something that sustains the life of the bands, though - Dead shows sold out every time.
In the Internet age, as music is distributed quickly over the world, it can be difficult for people to make it to concerts, if only for reasons of proximity. So there's a little incongruity between the distribution & "support" methods (if you will), because most people can't take advantage of the support. That's why I think we'll see a lot of effort put into the creation of "virtual concerts", which people will pay to see if the show is good enough.
There are lots of models one can set up for distributing independent music, and obviously the models are really the issue here, not the file format. I run an mp3 site (www.mirrormusic.com) that has a totally different model, at least for now: we're not _selling_ any mp3s, we simply distribute them. Artists upload for free, listeners download for free. The only people paying anything are us, who pay to have the site hosted. We're not selling ad space either (yet, anyway).
We plan to offer artists a way to charge for their music if they want to, but so far nobody's really asked for that kind of service. People just seem to want distribution.
One interesting thing about the site (we do this just because it's interesting) is that we impose no genres on the songs. Songs are associated with each other by comparing various people's ratings of various songs to come up with correlations between songs.
Also, the technology we're using is pretty fun - Apache, mod_perl, HTML::Mason, CVS - the stuff I like.
What a ridiculous comment. Of course every computable function can be written in Perl. Pretty much any C code can be translated directly into Perl, and this is the way a lot of horrible Perl gets written.
Here's why a Perl algorithm book is worthwhile:
Perl's builtin data structures have features that can be used to great effect in algorithms, particularly hashes. For instance, here's a classic example of an algorithm to remove duplicates from a list, or return the union of two lists:
sub unique { my %temp = map {$_,1} @_; return keys %temp; }
Or, more succinctly:
sub unique { return keys %{{ map {$_,1} @_ }}; }
This is way different from the algorithm a transplanted C programmer would use. Things that are appropriate in one language aren't necessarily appropriate in another.
Slashdot Mirror
That's not true, Ancestry.com lets people have more than one entry in the tree - perhaps your experience with it is quite old. It's also relatively straightforward to add anything you want as a source, though you get extra bells & whistles if it comes from a database they've got all mapped out.
Why do you keep saying "muslim types"?
Neural networks have been investigated for years as a text classification tool, and no, they don't tend to work very well. In addition, it usually takes so long to train them that most categorization researchers don't bother too much with them anymore.
-KenYou've got it exactly backwards, so calm down.
They're saying that the AltiVec Linux apps are [occasionally] 1000% faster than the non-AltiVec Linux apps. Linux is the common factor, not the difference. AltiVec is the difference, so Apple (and IBM & Motorola, who make the chips) has plenty to brag about. Their hardware is mighty fine.
I've been thinking about this too. The main ways for OSS developers to make money from their products (not that they always want to) are training & support. It seems like the parallel in the music industry is that artists like to play live, and distributing your music for free is a good way to keep people interested in you and get them to show up for the concerts.
It's really not so different from the hugely popular tape-trading networks that bands like the Grateful Dead have been using for years and years, except in that case people _do_ pay for the music (they have to buy tapes, stamps for mailing, etc.), but the band never saw a dime directly. Most people agree that tape-trading is something that sustains the life of the bands, though - Dead shows sold out every time.
In the Internet age, as music is distributed quickly over the world, it can be difficult for people to make it to concerts, if only for reasons of proximity. So there's a little incongruity between the distribution & "support" methods (if you will), because most people can't take advantage of the support. That's why I think we'll see a lot of effort put into the creation of "virtual concerts", which people will pay to see if the show is good enough.
There are lots of models one can set up for distributing independent music, and obviously the models are really the issue here, not the file format. I run an mp3 site (www.mirrormusic.com) that has a totally different model, at least for now: we're not _selling_ any mp3s, we simply distribute them. Artists upload for free, listeners download for free. The only people paying anything are us, who pay to have the site hosted. We're not selling ad space either (yet, anyway).
We plan to offer artists a way to charge for their music if they want to, but so far nobody's really asked for that kind of service. People just seem to want distribution.
One interesting thing about the site (we do this just because it's interesting) is that we impose no genres on the songs. Songs are associated with each other by comparing various people's ratings of various songs to come up with correlations between songs.
Also, the technology we're using is pretty fun - Apache, mod_perl, HTML::Mason, CVS - the stuff I like.
What a ridiculous comment. Of course every computable function can be written in Perl. Pretty much any C code can be translated directly into Perl, and this is the way a lot of horrible Perl gets written.
Here's why a Perl algorithm book is worthwhile:
Perl's builtin data structures have features that can be used to great effect in algorithms, particularly hashes. For instance, here's a classic example of an algorithm to remove duplicates from a list, or return the union of two lists:
sub unique {
my %temp = map {$_,1} @_;
return keys %temp;
}
Or, more succinctly:
sub unique {
return keys %{{ map {$_,1} @_ }};
}
This is way different from the algorithm a transplanted C programmer would use. Things that are appropriate in one language aren't necessarily appropriate in another.