Is there a way to tell BitTorrent to serve existing files? I'm still willing to serve it up for awhile.
Put the downloaded files in the same format as the.ISO in the download directory (current directory on unixen, not sure about windows) and run the download on the torrent. It'll run a checksum, then start the upload.
There is one, and only one reason that there is a market for Windows Media Formats... They are based on MPEG-4, but Microsoft charges about half the licensing fees for it's use.
It would break backward compatibility with numerous applications, not to mention nvidia's binary drivers (IANAKD). It'll probably be fixed when time_t's go 64-bit.
That seems to me to be a joke project. It may even be against sourceforge's TOS. Also, could someone tell me what the lzip jpeg image is? I'm afraid to look:P
Good theory, but flawed. All ASCII text can be compressed to 7/8ths of its original size of smaller (remove the high-order bit). Note that the engineers might use unicode;P
Maybe you should have pointed out that Windows is booting from a fast harddrive while Knoppix gets the CDROM drive, which is usually slower. Or maybe Windows is just faster (*gasp* Never!).
You contradict yourself. First you say windows gets a speed boost from the hdd, then you say that windows is inherently faster. Since this is based on the premise of equal wallclock time, this creates a contradiction.
Also, AC posting is denied, which is annoying as I use the login form there - why is this?
What do you mean, try that on Motorola iron? The code should work on any platform, provided you remain within the proper ranges for values (0..2^32-1 for unsigned long).
Slashdot Mirror
You try subscribing to the lkml sometime - an email comes in several times a minute. Be glad there are people to sort out individual threads.
Sure, you can use a fusion reactor in your twisted hunk of permenantly magnetized metal^W^W^W^W^W^Wcar.
Why do people say to use D+T reactions when D+D gives more energy and dosen't produce a high-energy neutron?
Where's the bittorrent link from the last article? I'm seeding it, but that won't help if noone has the .torrent file.
PPC and linux, of course. Or, to reduce the legacy even more, PPC and hurd :)
What about hurd?
Does it support the RFC Evil bit?
Splenda isn't perfect, either.
It would break backward compatibility with numerous applications, not to mention nvidia's binary drivers (IANAKD). It'll probably be fixed when time_t's go 64-bit.
Started Apr-01-03 22:27:23 PST
I've been using gentoo for a while. It was great - until the ebuilds stopped working. I'm stuck in the middle of an ABI change :(
dd if=/dev/zero of=bigfile bs=1048576 seek=61440
Instant 60gb.
That seems to me to be a joke project. It may even be against sourceforge's TOS. Also, could someone tell me what the lzip jpeg image is? I'm afraid to look :P
Good theory, but flawed. All ASCII text can be compressed to 7/8ths of its original size of smaller (remove the high-order bit). Note that the engineers might use unicode ;P
Wouldn't it be better to hack mp3 or ogg? These filter out noise (to an extent) and are designed for detectnig similarities *in audio files*.
The Linux uptime counter rolls over after just over a year - those uptime charts will be inaccurate. For true uptime measuring, touch a file at boot.
Also, AC posting is denied, which is annoying as I use the login form there - why is this?
Yes - and the only one you need is at the top.
I don't think this is possible in freenet... does anyone know if this is the case?
What do you mean, try that on Motorola iron? The code should work on any platform, provided you remain within the proper ranges for values (0..2^32-1 for unsigned long).
Possibly, but it's trivial to implement your own:
struct nlong { unsigned char data[4] };
struct nlong htonl(unsigned long foo){
struct nlong x;
int i;
for(i = 0; i < 4; i++){
x.data[i] = foo & 0xffu;
foo >>= 8;
}
return x;
}
unsigned long ntohl(struct nlong x){
unsigned long foo = 0;
int i;
for(i = 0; i < 4; i++){
foo |= x.data[i];
foo <<= 8;
}
return foo;
}
That's only used during the connection stage, and moreover is the original Gnutella. In both protocols, it switches to a binary approach later.
Oh, like this?