On Sat, 17 May 2003, Ian Clarke wrote:

> I don't think an ugly platform specific kludge is justified just because 
> a few crazy people try running freenet on really slow machines.

It sounds like you mean 'crazy' to be negative.  It would be preferable, 
surely, to be able to put Freenet on as many machines as possible.  So 
performance for that node suffers?  So what?  If someone with a slow 
machine is able to contribute that little bit to Freenet, don't we want to 
be able to use it?

When you consider, as Greg does, CPU-usage based performance as an issue 
separate from CPU overloading, it starts to make more sense.  (To me, 
anyway.)

You're always going to have widely varying specs on fred's environment.  
My computer is different from your computer is different from anybody's 
computer.  It may be (but is not proven yet) that there won't be a 
"solution" to CPU overloading.  It may be that it makes more sense to 
give fred a limit based on CPU-load.  That would provide for the 
ability to run in a wider variety of environments, including low-resource 
machines, and work on performance can continue as well.

This makes sense.

-todd
_______________________________________________
devl mailing list
devl at freenetproject.org
http://hawk.freenetproject.org:8080/cgi-bin/mailman/listinfo/devl

Reply via email to