I'm not convinced by Ben's arguments. Thanks for pushing back Matt.
https://fountain.fm/episode/zbXnveqNy4RSnIUSwcso
nostr:nevent1qvzqqqpxquqzqa8arj3nz69sudnwhmgfw87plqjvy3dua2c923383qyxxv9nsa9j5nce05
I'm not convinced by Ben's arguments. Thanks for pushing back Matt.
https://fountain.fm/episode/zbXnveqNy4RSnIUSwcso
nostr:nevent1qvzqqqpxquqzqa8arj3nz69sudnwhmgfw87plqjvy3dua2c923383qyxxv9nsa9j5nce05
Quote: "it's a waste of bandwidth if a node keeps giving you transactions that you already have".
Question: would something like neg entropy be useful for the propagation of transactions that are in the mempool?
nostr:npub1gzuushllat7pet0ccv9yuhygvc8ldeyhrgxuwg744dn5khnpk3gs3ea5ds
About 23 minutes in..
nevent1qvzqqqqqqypzqquxdpn0xlh4zqw9k3patfqml9nnndqkyd9e642sfxzlycj5279pqy2hwumn8ghj7un9d3shjtnyv9kh2uewd9hj7qghwaehxw309aex2mrp0yh8qunfd4skctnwv46z7qpq2h0sstzrc7acfjk47xgpvx3v2yuhqz0w55ruqv2fxs0jevksnr8q69urpp
I was also not persuaded by the arguments in favor of the change and agree that nostr:npub1qny3tkh0acurzla8x3zy4nhrjz5zd8l9sy9jys09umwng00manysew95gx had the best position on the issue: let the node operators decide how they run their node.
The OP_RETURN change doesn’t matter to me in and of itself. Removing configurability for node operators and the arguments put forward to justify that removal seem more problematic to me.
The node operators already do decide.
Note that delivering software that has 100 configurable options implies creating 100! different test cases *at minimum* (though of course that can't actually be done, implying some scenarios are left untested).
Yup, found O'Dell's explanation on RHR, direct and to the main points. When ppl make it sound super complicated (just like shitcoiners and their tokens) you know there is some BS in play.
Configurable nodes would be ideal..
Two nuanced points were made that I wasn't aware of:
1. The limits on OP_RETURN were not added by consensus.
2. Bitcoin Knots is maintained by a single person.
Indeed. I appreciate Odell offering OpenSats support for Bitcoin Knots.
Not convinced either. Core devs should fork bitcoin core and follow through with this PR. Then people who agree with this change can run that software. Argument for this is, bitcoin core is clearly a "software monopoly" and this should be humbly respected by core devs as a flaw of the node-landscape today.