From deb2cf2bdec1eecbeed0231de088f44936d28e57 Mon Sep 17 00:00:00 2001 From: Bryan Newbold Date: Wed, 14 Mar 2018 22:45:58 -0700 Subject: confirmed all-zero hashing key --- proposals/0000-hyperdb.md | 6 ------ 1 file changed, 6 deletions(-) diff --git a/proposals/0000-hyperdb.md b/proposals/0000-hyperdb.md index 8c054f9..19f1d5d 100644 --- a/proposals/0000-hyperdb.md +++ b/proposals/0000-hyperdb.md @@ -564,12 +564,6 @@ mafintosh mentioned this might be in the works. Does this DEP need to "leave room" for those changes, or should we call out the potential for future change? (Probably not, should only describe existing solutions) -Should all-zeros really be used for path hashing, or should we use the -hypercore feed public key? It certainly makes implementation and documentation -simpler to use all-zeros, and potentially makes it easier to copy or migrate -HyperDB content between hypercore feeds. Referencing the feed public key breaks -abstraction layers and the separation of concerns. - There are implied "reasonable" limits on the size (in bytes) of both keys and values, but they are not formally specified. Protobuf messages have a hard specified limit of 2 GByte (due to 32-bit signed arthimetic), and most -- cgit v1.2.3