Thanks Brandon. Here's the revised plan: * Java code defaults to current directory, but location of config file can be specified (this is probably true now).
* Wrappers specify absolute location - this is how an enduser will do it. * Use platform-specific configuration file naming - code this into wrapper scripts. * Client can read and re-write node configuration file. * Once I can figure out a good way, make it so client can start and stop node. -- I can understand why you want the proxy (/servlet) separate from the node. The only problem is that a lot of people will want to run both, and having two OS processes will consume a LOT of memory. (This is Java we're talking about!) > > > This is a much better idea. But why do you need a -HUP signal that can be > > > sent remotely? Why not just have the configurator kill the node process > > > and restart it? Unless you're talking about running the configurator > > > remotely, which I see no need for. > > > > Only because, if you are writing pure Java, this is very difficult to > > achieve. > > Pah. It's easy. You can do like the distributed.net RC5 cracker does and > have it periodically check for a file whose existence means it needs to > exit immediately. OK. That's a good way to have the thing shut down. Now all I need to do is figure out how to make the client spawn stuff. This is not possible without the aid of *some* platform-specific component. The trick is to make it as clean and pleasant a mechanism as possible. I'm sure I can think of something. > > Yes, my plan was to integrate fproxy, not change it (unless it needs > > changing). Perhaps the HTTP proxy could either be standalone, or > > run as part of the client, at the user's discretion. Perhaps it doesn't > > belong in the node. Perhaps it does. > > fproxy isn't a proxy, it's a servlet. So that's a decision to make, proxy > or servlet (both! yay!). I don't think it should go in the node. Just have > the launcher thingy be able to launch that too. I think it would be EXTREMELY cool if we could find a way to make browsers respond to URLs of the form "freenet:<key>" and direct them to the freenet. If not, we'd need to make the "proxy" re-write all the URLs (they would be stored on freenet in 'freenet:<key>' form) as they come back so that links would work. Yuck. I'm sure Mozilla and Konqueror won't be hard to convince, anyway. > > If you force users to run nodes, then you will get zillions of people > > running a node on a dialup modem for 15 minutes at a time. This would > > kill the network. Surely it's better to let them freeload than bugger up > > the network. (My facts may be wrong here.) > > This won't kill the network. That's what the transient option is for. You > can freeload AND run a node. You should run a node. Fairy nuff. Steve _______________________________________________ Freenet-dev mailing list Freenet-dev at lists.sourceforge.net http://lists.sourceforge.net/mailman/listinfo/freenet-dev
