[bitcoin-dev] BIP 158 Flexibility and Filter Size

[Gregory Maxwell]

On Sat, Jun 2, 2018 at 12:01 AM, Olaoluwa Osuntokun <laolu32 at gmail.com> wrote:
>> A typical network attacker (e.g.  someone on your lan or wifi segmet, or
>> someone who has compromised or operates an upstream router) can be all of
>> your peers.
>
> This is true, but it cannot make us accept any invalid filters unless the
> attacker is also creating invalid blocks w/ valid PoW.

I wish that were the true, but absent commitments that wouldn't be the
case unless you were always downloading all the blocks-- since you
wouldn't have any sign that there was something wrong with the
filter-- and downloading all the blocks would moot using the filters
in the first place. :)

Or have I misunderstood you massively here?

For segwit originally I had proposed adding additional commitments
that would make it possible to efficiently prove invalidity of a
block; but that got stripped because many people were of the view that
the "assume you have at least one honest peer who saw that block and
rejected it to tell you that the block was invalid" security
assumption was of dubious value. Maybe it's more justifiable to make
use of a dubious assumption for a P2P feature than for a consensus
feature?  Perhaps,  I'd rather have both filter types from day one so
that things not implementing the comparison techniques don't get the
efficiency loss or the extra work to change filter types for a
consensus one.

[I think now that we're much closer to a design that would be worth
making a consensus committed version of than we were a few months ago
now, since we are effectively already on a second generation of the
design with the various improvements lately]