Re: intent of package seti@home
> That's not completly fair. Distributed.net has released the core code for
> their client, they just haven't released their networking protocols.
Ok, perhaps I did word that a little harshly, at least in the case of
distributed.net..
> It's really an interesting problem, and perhaps one of the places where open
> source breaks down - how can you be sure if you release the code that
> someone doesn't fake results? Until someone coms up with a method of
> preventing this type of abuse, I don't expect to see them change their ways.
but this is still security through obscurity though.. it may stop your
average script kid from messing with their keyspace, but it is hardly
protection from someone reverse engineering the protocol and doing the
same..
sending a false positive would be pointless.. it would almost instantly
be exposed.. but there is a fairly large time window to perfect the art
of sending false negatives.. (still a statistically fruitless exercise
for one person given the size of the keyspace, but then people *do* buy
lotto tickets ;-)
I _can_ understand distributed.net's caution in this though.. they have
committed several years of processing time to what (barring successful
sabotage, or a bug in their code ;-) must definitely be a positive result..
In the case of seti@home though, I can't see how the same applies.. It's
not like the flat earth society is going to purchase a Cray to send
billions of "nobody's home" messages back to them.. and it's not like it's
going to be particularly hard for them to confirm that, while the message
"Live long and prosper" may have originated from a greater intelligence,
it certainly didn't come from thousands of light years away ;-)
(yes I'll bet someone pulls that one within the first 12 months whether
they release the code or not 8)
You are right, it would be a very interesting problem - except that it
is not, because we have been kept on the other side of the fence..
If J. Random. Hacker found a 1% optimisation in the distributed.net
code then it could slash months off their worst case completion date.
But in their case the risk of total failure may not be worth it to
them.
seti@home on the other hand have only their funding to lose if the
well remains dry for too long. While their peers still grope for
gravitons and ignore evidence of ftl objects, I'd have thought they
would do well to enlist all the help they can get!
..but that's another issue entirely and I've rambled enough already,
best,
Ron.
Reply to: