| 11:06 |
asciilifeform |
!!ticker btc usd |
| 11:06 |
deedbot |
$26,779 |
| |
~ 27 minutes ~ |
| 11:34 |
awt |
dump incoming? |
| |
~ 19 minutes ~ |
| 11:53 |
awt |
https://twitter.com/BitcoinMagazine/status/1665736512451883008 << SEC sues Binance |
| 11:57 |
awt |
http://logs.bitdash.io/pest/2023-06-04#1026654 << https://twitter.com/aisolopreneur/status/1665695277997498370 |
| 11:57 |
bitbot |
Logged on 2023-06-04 14:13:10 signpost: next hilarious step in the jocko relationship is LLMs trained to be your own personal jocko. |
| |
~ 4 hours 44 minutes ~ |
| 16:42 |
awt |
I can now simulate a pestnet of arbitrary size: http://paste.deedbot.org/?id=IlcK |
| |
~ 1 hours 6 minutes ~ |
| 17:48 |
phf |
i was thinking about pest as a kind of universal messaging protocol, this is perhaps a niche application (lol, niche), but one of the things i've been keeping in the back of my mind is torturing the implementation into routing telegram messages for me |
| 17:52 |
phf |
so there's a counterparty station, that sends "fake" hearsay messages, which would be easy. the feature that's not supported is dm's to hearsay peers. obviously it would be insecure under normal circumstances, since relay counterparty decodes messages, but maybe in cases where you |
| 17:52 |
phf |
control the relay counterparty? or cases where relay counterparty is a proxy for some heathen protocol |
| 17:58 |
phf |
but the other idea that i had is an explicit threaded view for the messages, so you could use pest as a proxy for various forums. each message (or multi-part message) is treated as single post, and the replies are netchained to their corresponding parent messages. so instead of bei |
| 17:58 |
phf |
ng a linear progression of messages, it's instead an outline. |
| 18:03 |
phf |
when all you have is a graph™ |
| 18:09 |
awt |
phf: I sort of bumped into the threading idea when contemplating a telegram/discord bridge bot. Gotta translate the threads somehow. |
| 18:11 |
phf |
we're doing some haram here, asciilifeform will not be pleased :> |
| |
↖ |
| 18:13 |
phf |
you have two different modalities though. one is chaining, that is the graph is used to establish a sequence, and the other one is outline, where graph is used to establish a tree. pest uses the first one, where's telegram/discord will necessarily have to use the second one, which doesn't work very well with pest. |
| 18:13 |
awt |
probably someone more interesting walkins from telgram, while discord might bring in a younger crowd. |
| 18:14 |
phf |
like PROD breaks |
| 18:15 |
awt |
yes - tree is a challenge since no way to say - this is the next message vs. this is a reply. |
| 18:16 |
phf |
also you can use netchain for responding to individual hearsay messages from fake peers presented by the proxy station, but you can't initiate a conversation that way. the first message will have nothing to hang off |
| |
↖ ↖ |
| 18:26 |
phf |
test |
| 18:28 |
phf |
this is weird. how come i'm the only one relaying messages from awt to nosuchlabs? i had a bug in my code, debugged it live because lisp, did a restart and that's when this message http://logs.nosuchlabs.com/log/pest/2023-06-05#1027111 materialized |
| 18:29 |
phf |
but it should've arrived to nosuchlabs way earlier by virtue of being relayed by someone else? |
| 18:30 |
awt |
I think not the only one - you can see my messages coming through asciilifeform above. |
| 18:30 |
dulapbot |
(asciilifeform) 2023-02-27 deedbot: Peering info: http://paste.deedbot.org/?id=TyXN |
| 18:30 |
dulapbot |
Logged on 2023-06-05 18:24:30 awt[phf]: yes - tree is a challenge since no way to say - this is the next message vs. this is a reply. |
| 18:31 |
awt |
I just got a replay of that message from you. |
| 18:33 |
awt |
Which is odd because it should already be in my db. |
| 18:34 |
awt |
Not sure who else is connected to dulapbot. |
| 18:34 |
awt |
I'm not. |
| 18:34 |
awt |
obv. |
| 18:36 |
awt |
Wait, replay was of the log reference. |
| 18:39 |
awt |
So for this log reference, I didn't get it from anyone else, had to getdata your station for it. |
| |
~ 41 minutes ~ |
| 19:20 |
awt |
might be nice in the gui to have some sort of indication of "connection" quality |
| 19:20 |
phf |
live peers: asciilifeform, awt, cgra, crtdaydreams, deedbot, dulapbot, jonsykkel, lobbes, signpost |
| 19:21 |
phf |
lol that's actually a lie… |
| 19:22 |
phf |
live peers: awt, deedbot, dulapbot, jonsykkel, lobbes, signpost |
| 19:23 |
awt |
I suspect my normal route to dulapbot is currently flakey |
| 19:24 |
phf |
broadcast-live-peers, http://okturing.com/src/16033/body |
| 19:29 |
awt |
some kind of a darker and darker warning color for a decreasing moving average of packets/min from a peer might be nice |
| 19:29 |
awt |
rather than on/off |
| 19:30 |
awt |
won't help much with l2 peers. |
| 19:30 |
awt |
actually might |
| 19:35 |
phf |
none of the packet stream is mandatory. if you're behind firewall you want to continue sending IGNORE to keep the socket alive, but otherwise… |
| 19:38 |
awt |
In that case depending on the moving average window if the peer sent a broadcast regularly once a month, should be normal color until a couple months of no broadcasts. |
| 19:39 |
phf |
that's fair, but i'm still reading my messages by looking at the raw console output, so i'm definitely not there yet |
| 19:39 |
awt |
But yes, if completely silent, would go dark. |
| |
~ 3 hours 19 minutes ~ |
| 22:59 |
asciilifeform |
http://logs.bitdash.io/pest/2023-06-05#1026695 << bridges b/w pestnet & aols, www doors, etc. imho are worth baking -- iirc we had coupla thrds re subj |
| 22:59 |
bitbot |
Logged on 2023-06-05 18:11:08 phf[4]: we're doing some haram here, asciilifeform will not be pleased :> |
| 22:59 |
asciilifeform |
( for that matter, with classical irc , ditto ) |
| 23:00 |
asciilifeform |
'gate drug' or how did it go |