What penalty, lowered daily quota? You have no idea how easy it is to make the server think I'm a new host.
Once, LHC was giving 2 tasks per request, and requested a delay of 10 minutes between requests. If you forced an update before the 10 minutes passed, it replied "last RPC too recent" and didn't give you work. By making it think I was a new host, the server thought it was my first request ever and gave me 2 more tasks. And it's not like adding a request/offer/accept facility to the protocol implies there will be a wizard in the GUI to automatically reject workunits with X characteristics. They'd have to edit the client code to (ab)use the protocol that way. And if people can do that, they might as well implement an auto-task-aborter *now*. El Vie 08 May 2009 14:31:54 john.mcl...@sybase.com escribió: > But currently, there is a penalty. > > jm7 > > > > Nicolás Alvarez > <nicolas.alvarez@ > gmail.com> To > Sent by: boinc_dev@ssl.berkeley.edu > boinc_dev-bounces cc > @ssl.berkeley.edu > Subject > Re: [boinc_dev] Work Scheduling (pt > 05/08/2009 01:06 2, Cooperative Scheduling?) > PM > > El Vie 08 May 2009 12:59:59 Rom Walton escribió: > > I mean if I'm a researcher for rose...@home and half my client base > > decides not to run a certain type of task because they don't think it > > doesn't do X, there by doubling the length of time for a CASP run to > > complete, I'm going to be annoyed. > > > > Or I'm working on my PHD thesis and I manage to only attract power > > crunchers who don't like a certain batch of tasks, how am I going to > > complete my research? > > If someone doesn't like a certain small batch of tasks, they already have > the > ability to abort the tasks. > > If someone doesn't like a certain big batch of tasks, they already have the > > ability to detach and contribute to another project. > _______________________________________________ > boinc_dev mailing list > boinc_dev@ssl.berkeley.edu > http://lists.ssl.berkeley.edu/mailman/listinfo/boinc_dev > To unsubscribe, visit the above URL and > (near bottom of page) enter your email address. _______________________________________________ boinc_dev mailing list boinc_dev@ssl.berkeley.edu http://lists.ssl.berkeley.edu/mailman/listinfo/boinc_dev To unsubscribe, visit the above URL and (near bottom of page) enter your email address.
