Bloglines.com is a great RSS aggregation service, which handles as many RSS feed as you want, lets you sort and search them, and makes it really easy (compared to others) to add any feed you find on the Net. It's free and of course it's available on any OS. It also means that you never miss an update when you're away from your home PC.
And as a cherry on top, they have apps for all 3 major OS's that work with the website to notify you of updates when you're not using your browser. I don't personally use these helper apps though, so I can't vouch for them.
In summation: you should check it out, it's great!
That is so great! You know I've been waiting for SLI for just years now, and thanks to Nvidia, I can finally...
...I can...
...well what the fuck is SLI anyway?
Lately I've been thinking about an idea for a great website, which I'd like to call ">/." (greater than slashdot) or something. For this site I would just rip all of/.'s articles and edit them so that visitors to my site could tell wtf the article is actually about. I would also check for dupes and other things that editors generally do when they haven't been replaced by stuffed penguins sitting on bags of money.
It's easy to surpass EZ-D with almost anything, because nobody bought them. I live in one of the cities they test-marketed these in, and the stores almost couldn't give them away, let alone sell them in some fantastic multi-tiered price gouging orgy.
People just don't want to buy something that becomes worthless as a matter of course, and they probably never will. As far as I'm concerned, these products are just interesting exercises in chemical engineering, and nothing more.
The traffic waves site you linked seems to be in some pain, so here's the Wayback Machine's mirror of the page from last June. (Complete with cute traffic wave animation!)
I had seen this site before (maybe when it was first posted to/.) and I've also tried nullifying traffic waves. Guess what, it works! And it's kinda fun in a serious slowdown, to see the people behind you start out really irritated, but eventually realize how nice it is not to be using the brakes every fifteen seconds.
Freenet doesn't really do any straight pushing, because the files are mirrored during the process of downloading them. So there's no real overhead on that feature of Freenet's architecture.
As for redundant data, only time will tell, but I think it will work nicely. If for no other reason, redundancy will be minimized with the use of identifying metadata (ie "Artist", "Title"), and when searching is implemented in Freenet, I can see it working as well as other systems in those aspects, and better in all the others. (See? I'm optimistic.)
I don't quite understand all of the Freenet protocol myself, even though I've been following its development for a few months. But apparently between DH exchanges, and various hash-based file validation methods, Freenet is supposed to be resistant to all Man In The Middle attacks, as of this new version.
But like I said, I don't understand the protocol. Probably cuz I'm a not a programmer.:-(
Actually, one of the main selling points of Freenet is that, in fact, you can't beat it with a handful of warrents. Getting a warrant would give you access to the computer running a node, but you'd never learn anything because all of the file data is encrypted. Even if you could somehow decrypt the files, the owner of the node has complete deniability because of Freenet's dynamic mirroring. (i.e. "I never downloaded that file. Someone must have requested it through my node.") You would be able to learn the IPs of other nodes that this node has been connecting to, but that sort of information is only relavant if the FBI has some legitimate reason to shut down all of Freenet that it can get it's hands on. Which is unlikely, at least for now.
The Freenet guys are ultimately hoping to have a "subversion-mode" dist of Freenet which will be completely hardened to identification as Freenet traffic, and that sounds fuckin cool:-)
...unless these guys come up with an entirely new and better protocol.
Ever notice how Gnutella just loves to suck all your bandwidth up, as twenty billion people search for "britney spears cumshot"? Well just imagine the joy of having the entire Internet glutted with that kind of crap.
From their site, it looks as though they're still asking for proposals. So let's hope someone has better ideas than "Gnutella to the Stars!!!" if they hope anyone will run it on their servers.
Freenet isn't just "working on" avoiding the problems of Gnutella, it's entire architecture is based on intelligent routing, including:
never broadcasting search queries (thank god).
Intelligently mirroring any and all data to a subset of the nodes that route the file to you
finding a file in the network is done in a chain, with the first node that knows where the data is directing the file request straight to that node
What this means is that, sure, in theory you could have to go through a hundred hops before you find your file. But next time you want that file, it'll be right next door (e.g. one hop away).
That, in my opinion, is leaps and bounds and orgasmically better than what Gnutella can do. Don't you think?
Slashdot Mirror
And as a cherry on top, they have apps for all 3 major OS's that work with the website to notify you of updates when you're not using your browser. I don't personally use these helper apps though, so I can't vouch for them.
In summation: you should check it out, it's great!
...I can...
...well what the fuck is SLI anyway?
Lately I've been thinking about an idea for a great website, which I'd like to call ">/." (greater than slashdot) or something. For this site I would just rip all of /.'s articles and edit them so that visitors to my site could tell wtf the article is actually about. I would also check for dupes and other things that editors generally do when they haven't been replaced by stuffed penguins sitting on bags of money.
So, WHO'S WITH ME?! :P
People just don't want to buy something that becomes worthless as a matter of course, and they probably never will. As far as I'm concerned, these products are just interesting exercises in chemical engineering, and nothing more.
Not to be a googlewhore, but who knew that Back to the Future had any real math in it before they saw this?
I had seen this site before (maybe when it was first posted to /.) and I've also tried nullifying traffic waves. Guess what, it works! And it's kinda fun in a serious slowdown, to see the people behind you start out really irritated, but eventually realize how nice it is not to be using the brakes every fifteen seconds.
here is a direct link to the full size vid:1 a1a1aaa2198c627970773d80669d84574a8d80d3cb12453c02 589f25382f668c9329e0375e81787e85abb28970c7aee1d8de e67ca3297fa65/paycheck_m480.mov
http://a772.g.akamai.net/5/772/51/30378f1c0dfafa/
CHK@fWoX96ph98S3fw6xnuA8PMcKgQwPAwI,iOLxdUpeRNRr0m uzkg17zA/maxpaynehero.zip (you need freenet for this link to work)
If there's one good thing about downloading from freenet, it's that it's more fun than waiting in line at FilePlanet after a /.ing. :P
Also check out linux-on-laptops.com for what looks like a very thorough listing.
Now I can finally get those jewel cases for my record collection!
The paper that was presented, entitled Search and Replication in Unstructured Peer-to-Peer Networks, can be found here (top of the page)
I can't wait.
-=20
As for redundant data, only time will tell, but I think it will work nicely. If for no other reason, redundancy will be minimized with the use of identifying metadata (ie "Artist", "Title"), and when searching is implemented in Freenet, I can see it working as well as other systems in those aspects, and better in all the others. (See? I'm optimistic.)
-=20
But like I said, I don't understand the protocol. Probably cuz I'm a not a programmer. :-(
-=20
The Freenet guys are ultimately hoping to have a "subversion-mode" dist of Freenet which will be completely hardened to identification as Freenet traffic, and that sounds fuckin cool :-)
-=20
Ever notice how Gnutella just loves to suck all your bandwidth up, as twenty billion people search for "britney spears cumshot"? Well just imagine the joy of having the entire Internet glutted with that kind of crap.
From their site, it looks as though they're still asking for proposals. So let's hope someone has better ideas than "Gnutella to the Stars!!!" if they hope anyone will run it on their servers.
-=20
never broadcasting search queries (thank god).
Intelligently mirroring any and all data to a subset of the nodes that route the file to you
finding a file in the network is done in a chain, with the first node that knows where the data is directing the file request straight to that node
What this means is that, sure, in theory you could have to go through a hundred hops before you find your file. But next time you want that file, it'll be right next door (e.g. one hop away).
That, in my opinion, is leaps and bounds and orgasmically better than what Gnutella can do. Don't you think?
-=20