hahaha all I was saying was just because everyone says it will fail that does not mean it will truly fail. It might fail yes, I admit the probability it will fail is significantly higher than it will not fail, but I want to do it because I want to do it and maybe I can learn a bit more about server hosting while I do it not because "I know it will be successful" Even the IT company said they are not going to do it if it is going to deal with the internet in general like google imaeg search engine.

The most well-known p2p at the moment, gnutella, works in the architecture you specified. Every client also works as a server and that's why it is very difficult, if not impossible, to shut it down. ISPs and authorites might be able to track down who is using gnutella client, by port scanning, but the gnutella network itself can't be shut down.

but Napster was based on a central server. That was why it was so easy to shut Napster down when Metallica and other people in the music industry first started yapping

I read... the dude who built Napster... he stayed up 72 hours to write codes for Napster.

I definitely want to meet him one day in person.

I wonder why people think....

1. I am going to cache powerpoint files, ms word files, pdf files, and so on, like google does. The fact is I am not even going to cache the contents of html files.
2. the database will be complicated.
the fact is my database table will be simple!! it will only have a few columns, namely image address, keywords, thumbnail's name, resolution, size.
that will be it!! thumbnails and the database table.

well, I might alternatively consider coming up with my own interface for google image search engine as well. that's a definite possibility.

I wish you luck even though I still don't know what you are making. Are you in Austin? (you don't have to answer that either)

I am in Canada... quite far away from Austin Texas :P