I'm looking for opinions: gist.github.com/tevador/639d083c994c1ef9401832c08e2b7832#appendix-d

The choice is between CSIDH-512 and CSIDH-1024. The table lists UX-related parameters. Would the cost of CSIDH-1024 be acceptable (regardless of security)?

<jpk68:matrix.org> Encoded address lengths could always decreased by ~14% using Base58 instead of Base32

<jpk68:matrix.org> It could also be argued that the large increase in address length (even using Base58) is alone enough to distinguish between Jamtis and legacy addresses. Plus, starting with a 4 or 8 is pretty uniquely Monero, IMO

<jpk68:matrix.org> However there could very well be some other disadvantages with Base58 that I am not aware of

<jpk68:matrix.org> Actually, since there are no 'secondary' Jamtis addresses, that wouldn't make sense. Maybe they could start with a 12 or 16 (add 4, or next power of 2). This is probably a dumb idea but oh well

<rbrunner7> @jpk68:matrix.org: You will find some background info about the design choices of these addresses, like using Base32, in my post about Jamtis addresses as they were thought to be used together with Seraphis: old.reddit.com/r/Monero/comments/zl…jamtis_addresses_will_be_so_awfully

jpk68: That's not the problem I'm trying to solve. I'm deciding between CSIDH-512 and CSIDH-1024, with the latter one having ~50% longer addresses in all encodings.

<jpk68:matrix.org> Apologies :)

FYI, I contemplated using base62 for the address data while keeping the prefix in base32, but decided against that due to QR code sizes.

rbrunner: In the meantime, Jamtis addresses grew from ~200 to ~400 or ~600 characters, lol.

<-----0:zano.org> Hello, Question about the spynode situation; is this notna result of not tying mixing to proof of work ?

<-----0:zano.org> I mean: decoys are added without some POW being needed. If it were the case then at least they would have contributed to the network.

<ofrnxmr> #monero-research-lounge:monero.social