[tahoe-dev] Tahoe-LAFS is widely misunderstood

[Greg Troxel]

Chris Palmer <chris at noncombatant.org> writes:

> Isn't my scheme simply the most reliable and most expensive form of erasure
> coding?

No, it's *N expensive, for some factor N.

I think what Shawn is comparing is

    1-of-N
    k-of-k*N

An example is 1-of-4, and 10-of-40.

For 1-of-4, and 50% reliability, you have a 1/16 odds of losing.
For 10-of-40, you win if any 10 are up; my head is too foggy to do the
math but it seems lossage is much less likely than 1/16.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 194 bytes
Desc: not available
URL: <http://tahoe-lafs.org/pipermail/tahoe-dev/attachments/20110202/5d67613a/attachment.pgp>