A question remains for me most of the suggestions so far depend on the fair play of the client. But we need a way on the server side to reduce update frequency of clients when there are tons of participants. I'm not too sure if TCP congestion control is the best thing to rely on this. One strategy would be to push updates to idle clients at a reduced rate.
What is the strategy for wave on DOS Attacks from maliciouse clients? Is it just the trust model meaning that since participants have to be invited to the wave we assume there are no maliciouse participants or is there some form of active prevention system in place? Stefan 2009/10/6 Michael K <[email protected]> > > On Oct 5, 11:59 pm, > "[email protected]<david.c.hubbard%[email protected]> > " > <[email protected]> wrote: > > Another way to optimize things for the client would be detect an idle > > client - that is, if the client is left open, just watching, but the > > user has not had activity in a while, the client could reduce the > > update rate and even signal the server to send fewer (larger) updates. > > If this happened gradually, the user would not notice. > > More importantly, this is something that can be taken advantage of on > the clients of the other participants. If a client is idle (i.e. no > user activity at all) for more than a few minutes, you can assume that > the user is not near the computer. That's almost the same as if the > user is offline, and thus not viewing the wave. So if you have e.g. 5 > participants online, but 4 of them have been idle for a while, the 5th > can reduce his update frequency as if he's the only one who's viewing > the wave. > > > --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Wave Protocol" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/wave-protocol?hl=en -~----------~----~----~----~------~----~------~--~---
