Hi. I made a github account just to post an issue (issue #10054), but the issue is not showing up publically. It only appears to me when I am logged in. I guess the account is shadowbanned. If I chuck the issue in a pastebin and put it here, please could someone here post it on my behalf.

note: The GitHub issue I posted is titled: ChainLocks Inspired Idea To (sometimes) Prevent Selfish Mining

<o​frnxmr:xmr.mx> Github doesnt like accounts created over tor

<o​frnxmr:xmr.mx> Send the link

<o​frnxmr:xmr.mx> Seems to be hidden for me

This is the github link: monero-project/monero #10054 I am just searching for a decent pastebin to use to send the text.

<o​frnxmr:xmr.mx> Yeah, gh nuked it

<o​frnxmr:xmr.mx> paste.debian.net

perfect: paste.debian.net/hidden/38699502

<o​frnxmr:xmr.mx> "if the miners of 60% (that number can be tweaked) of the last 100 blocks (that number can be tweaked) send signed messages to the Monero network that the latest block produced at the chain-tip is final, then that block -- at its height -- can be considered authoritatively "final" and cannot be orphaned as per consensus. Any block at the same height which is broadcast will be rejected by nodes. "

<o​frnxmr:xmr.mx> Problem: selfish mining allows you obtain a higher % of blocks than your hashrate indicates

<o​frnxmr:xmr.mx> Its been demonstrated that ~35% hashrate can obtain 50+/100 on chain blocks

<o​frnxmr:xmr.mx> This seems like it turns a 51% attack into only requiring 40%

There's a later paragraph where I made a (maybe correct) comment on why the idea I gave might still work

My expectation is that the written idea by itself is insufficient. However, my hope is that the core idea might be workable (with tweaking and ammending).

*If* it can be made to work, it looks like it could be more promising than the other ideas I've read about so far. The idea is simple and doesn't require a large changes or additions to the fundamental of the monero system.

<o​frnxmr:xmr.mx> Its similar to rolling checkpoints

v0.18.4.2 is tagged

<o​frnxmr:monero.social> (build) party time?

yes

<n​obfg9000:matrix.org> Hey do you mind if I ask a question about github.com/monero-project/supercop/tree/monero ?

<e​veroddandeven:monero.social> Hi guys, I hope I'm not asking this in the wrong room, but why if I specify default values for parameters `prune=false` and `split=false` for `gettransactions` rpc method, I get empty response? It's a bug?

<o​frnxmr:xmr.mx> what?

<o​frnxmr:xmr.mx> Oh nvm. Rpc method

<o​frnxmr:xmr.mx> Docs could be wrong

<e​veroddandeven:monero.social> matrix.monero.social/_matrix/media/…ero.social/idvMEHhOXYMYoikzpPuIMQwc

<e​veroddandeven:monero.social> Aah okok

<n​obfg9000:matrix.org> From what I can tell, this supercop library is just to benchmark different implementations. basicially, I was wondering if you would accept a PR to add more (architecture-independent) implementations like donna, etc. It would make supporting monero on gentoo on different arches easier.

You might be running a pruned node and the node doesn't have that whole tx ?

Or by "empty" you mean not an error ?

More impls mean more likelihood of a fork. We'd switch to a well reviewed impl if it were faster though.

Though support of more archs is also a good argument :)

IIRC the wallet switched to x86 asm for speed, but the node did not, due to the concern about forks.

<n​obfg9000:matrix.org> Do they not have faith in the asm implementations?

<n​obfg9000:matrix.org> that is a bit surprising

<n​obfg9000:matrix.org> I think it's just because the monero cli and monerod packages are the same in gentoo, because they're from the same codebase

<o​frnxmr:xmr.mx> Moneromooo, this? monero-project/monero #9856

Long ago.

I assume this is the same asm code being used in this path ?

n​obfg9000: we use the supercop lib for amd64-64-24k / amd64-51-30k because it causes a large speedup over the existing implementation

oh that was already answered

<j​effro256:monero.social> everoddandeven @everoddandeven:monero.social: the field name should be "txs_hashes", not " tx_hashes"

<g​ooseworld:matrix.org> There is an application that is very similar to matrix but doesn't use servers because it is P2P. It actually works very well, better than Matrix in some cases like large rooms and file transfers and obviously you don't need to run a server. keet.io is the website and it is available on most platforms. There is a link in the app on the profile page to the community room to get you<clipped message>

<g​ooseworld:matrix.org> started in the app.

<o​frnxmr:xmr.mx> Wrong room sirmaam