<oembed><type>rich</type><version>1.0</version><author_name>npub16dt55fpq3a8r6zpphd9xngxr46zzqs75gna9cj5vf8pknyv2d7equx4wrd</author_name><author_url>https://nostr.ae/npub16dt55fpq3a8r6zpphd9xngxr46zzqs75gna9cj5vf8pknyv2d7equx4wrd</author_url><provider_name>njump</provider_name><provider_url>https://nostr.ae</provider_url><html>📅 Original date posted:2018-06-02&#xA;📝 Original message:On Fri, Jun 01, 2018 at 07:02:38PM -0700, Jim Posen via bitcoin-dev wrote:&#xA;&gt; Without the ability to verify filter validity, a client would have to stop&#xA;&gt; syncing altogether in the presence of just one malicious peer, which is&#xA;&gt; unacceptable.&#xA;&#xA;I&#39;m confused about why this would be the case.  If Alice&#39;s node&#xA;generates filters accurately and Mallory&#39;s node generates filters&#xA;inaccurately, and they both send their filters to Bob, won&#39;t Bob be able&#xA;to download any blocks either filter indicates are relevant to his&#xA;wallet?&#xA;&#xA;If Bob downloads a block that contains one of his transactions based on&#xA;Alice&#39;s filter indicating a possible match at a time when Mallory&#39;s&#xA;filter said there was no match, then this false negative is perfect&#xA;evidence of deceit on Mallory&#39;s part[1] and Bob can ban her.&#xA;&#xA;If Bob downloads a block that doesn&#39;t contain any of his transactions&#xA;based on Mallory&#39;s filter indicating a match at a time when Alice&#39;s&#xA;filter said there was no match, then this false positive can be recorded&#xA;and Bob can eventually ban Mallory should the false positive rate&#xA;exceeds some threshold.&#xA;&#xA;Until Mallory is eventually banned, it seems to me that the worst she&#xA;can do is waste Bob&#39;s bandwidth and that of any nodes serving him&#xA;accurate information, such as Alice&#39;s filters and the blocks Bob&#xA;is misled into downloading to check for matches.  The amount of&#xA;attacker:defender asymetry in the bandwidth wasted increases if&#xA;Mallory&#39;s filters become less accurate, but this also increases her&#xA;false positive rate and reduces the number of filters that need to be&#xA;seen before Bob bans her, so it seems to me (possibly naively) that this&#xA;is not a significant DoS vector.&#xA;&#xA;-Dave&#xA;&#xA;[1] Per BIP158 saying, &#34;a Golomb-coded set (GCS), which matches all&#xA;items in the set with probability 1&#34;&#xA;-------------- next part --------------&#xA;A non-text attachment was scrubbed...&#xA;Name: signature.asc&#xA;Type: application/pgp-signature&#xA;Size: 833 bytes&#xA;Desc: not available&#xA;URL: &lt;http://lists.linuxfoundation.org/pipermail/bitcoin-dev/attachments/20180602/9556f60f/attachment.sig&gt;</html></oembed>