Hey, I was wondering if anybody else plays trade wars on an ssh connection, rather then telnet or other means. The reason I am asking is that I am trying to get a helper program up and running, but I run into the same hitch with every one that I try, they don't support ssh. If anybody has experience or ideas it would be much appreciated, thanks!

I'm sure new versions of zoc do, but they won't route thru twxproxy if you do that.

TWGS doesn't run ssh, so where are you interjecting ssh?

I'm guessing he's talking about a bbs interface.

*blink* If that's the case...

Why not just get putty, setup an SSH tunnel, and then proxy chain everything together?

Swath | Zoc | Telnet | Whatever -> TWXProxy -> SSH -> BBS's SSH front end -> TW

Lengthy and full of lag, but what the heck, eh?

I have the same problem as the OP, and I am trying to get what you've described to work. I am just not getting it... the explanation in the docs just baffles me. Could you describe specifically how to do this?

Forgive my complete ignorance here, but what's the benefit of running TW with SSH?

eleq did this a long time ago through a portal he wrote. i kinda liked it. it would add an encryption load to the twgs though and arguably provide no real benefit. also, it could break twx and swath. that being said, most online games secure traffic - but mostly to keep players from cheating with proxies. hehe.

Not much need for encription, but SSH can confirm the identity of the player.

Really, this could improve our ability to limit players to one account per game?

From the sysop end of things I don't get it either, but then I'm relatively ignorant about the technical side of [inter]networking. I really just wanted to be able to use a helper and scripts in a game run on a BBS I have to ssh to. I had started writing some EXTREMELY rudimentary REXX scripts (pretty much just glorified macros) and I was wondering if that was really necessary.

You would have to distribute client certificates to the players, probably by email. It still doesn't prove that the player you sent it to isn't a dupe though.

SSH confirms identities by means of asymmetric keys -- public and private key pairs. All it confirms is that the person who is able to receive messages encrypted with the public key must also be in possession of the private key. The protocol itself does not confirm that the person doesn't have more than one key pair, or that the private key wasn't stolen.

That's where certificate authorities come in. They "officially" (so far as they are trusted) associate a public key with some sort of identity.

If you wanted to enforce identity in TW, you wouldn't need to use SSH. You'd just need to exchange some encrypted tokens. You'd also need to create a CA and certify people's public keys. People would have to register with you by proving their identity somehow. And there's the problem. Even if you demanded three forms of ID and charged for the service, it would still leave opportunity to cheat if someone were determined to do so. (Borrow a couple friends' IDs, use a couple different addresses and phone numbers...)

And there is the problem in a nutshell. Even if I voice validate a player over the phone, how do I know I wasn't talking to his next door neighbor, friend, or co-worker?

Ok, that helps.

I've definitely thought about ways to authenticate, and we've discussed it quite a bit here. I know it would never keep players from having dupes if they want, but it does limit the dupe activity. I've never been too concerned about players who play more than one account as long as they're actually PLAYING those accounts. The problem comes in using accounts as free resources, and that's less likely to be a problem if you're using some kind of authentication.