RE: [Blocks-development] crazy anti-flooding idea:

[Spencer Jr., M. <sp...@ac...>]

Hmm...true, that makes sense.

Oh well.  Back to day 3 of my "Teach yourself C for Linux Programming in 21
Days" book, and chapter 1 of "Linux Socket Programming By Example"  At this
rate, I should be capable of contributing code to Blocks, oh, sometime next
year.  *kidding*

(Actually, I already know C++ rather well...I just need help and experience
tying my C++ experience to an API.  That C for Linux in 21 days book seems
to be focusing on C rather than on Linux.  Oh well.  :) )

--Michael Spencer
bl...@ms...

> -----Original Message-----
> From:	Blat Froop [SMTP:pet...@ho...]
> Sent:	Friday, August 18, 2000 9:01 AM
> To:	blo...@li...
> Subject:	Re: [Blocks-development] crazy anti-flooding idea:
> 
> >From: "Spencer Jr., Michael" <sp...@ac...>
> >To: blo...@li...
> >Subject: [Blocks-development] crazy anti-flooding idea:
> >Date: Fri, 18 Aug 2000 08:37:10 -0500
> >
> >CPU-bind the Eve clients!  ^_^
> >
> >Is there some kind of cryptographic function we can require to be used
> with
> >all broadcast messages passed, which is difficult to create (perhaps ~5
> >seconds of P3-500's processing to encode) but not difficult to decode?
> >
> >Could this be reversed, so someone could create a large number of packets
> >that are more difficult to decode than to encode (even though they'd all
> be
> >bogus packets) so they'd CPU-bind the good client instead?
> >
> >Do we want to require encoding only for people who originate the packets,
> >where packet routers just pass the data along?  Or should all broadcast 
> >data
> >(which should be protected in a special way) be encoded every time it
> gets
> >routed?
> >
> >What do you guys think?
> 
> I think the CPU requirement would kill the Blocknet. If this is
> designed to limit spam, then it would also limit normal throughput,
> and in 6 months time some script kiddie with a hacked server farm, beowulf
> 
> cluster, or with the latest $$$ box would be able to
> out perform the Blocknet. Also, for sending broadcast messages
> we would have to choose a func which was tragetted at taking 5
> mins for 16 connections. A spammer would simply send it only
> to one connection in order to flood faster... you can go on and on
> with this.
> 
> Good but of brainstorming, but I dont think this ones got much
> legs. :-)
> 
> ttfn
> 
> PG.
> 
> ________________________________________________________________________
> Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com
> 
> 
> _______________________________________________
> Blocks-development mailing list
> Blo...@li...
> http://lists.sourceforge.net/mailman/listinfo/blocks-development