Hi again. Is it correctly assumed that a query command is handled by sending the actual query in a drawinfo to the text screen, and then just the prompt and query type in the query command? If that really is so (it sure seems like it in the Protocol draft) I sure wonder why. Why not send the the query text in the query packet? This way the client can display a nice dialogue instead of just outputting some text in the window and expect an answer. By doing it this way the risk is high that the user dont notice the query and just wonders whats up when nothing happens. A dialogue that pops up would be much nicer in my opinion, but that would require that the actual query is in the query packet and not already written to the output area. Comments? // Isak Styf -------------- next part -------------- An HTML attachment was scrubbed... URL: http://shadowknight.real-time.com/pipermail/crossfire/attachments/20010522/3434e51e/attachment.html