Results 2501 ... 2750 for 'f:billymg'...

← | Results 2501 ... 2750 found in asciilifeform for 'f:billymg' | →

billymg: asciilifeform: my understanding was that prb solved the slow sync by allowing out-of-order syncing. i.e. blocks don't need to come in sequentially from 0 - 690k (i admit my understanding on this topic is pretty shallow)

billymg: thestringpuller: just start with asciilifeform's torrent of the block db

billymg: asciilifeform: as of late afternoon yesterday the trb nodes started picking up a bunch of spam peers again http://bitdash.io/nodes/205.134.172.27-8333

billymg: not one that results in bad data but that can result in too many probes in close succession for newly discovered nodes, for a brief period after they are first discovered

billymg: http://logs.nosuchlabs.com/log/asciilifeform/2021-07-10#1044769 << nice, looking at this node helped me identify a bug in my crawler

billymg: asciilifeform: cool, thanks for the link

billymg: asciilifeform: i hadn't thought of adjusting the memory for postgres, i'll look into that

billymg: almost in one third the time

billymg: asciilifeform: the pollers also burn through the queue much faster sitting at your DC pipe vs. my jungle fiber (with same number of max sockets)

billymg: true, the search is where the lag really shows

billymg: asciilifeform: the www runs fast enough on the rk too, i'm guessing has something to do with the 13k rows vs 2M in the logotron db

billymg: asciilifeform: not a feature but it's on my todo list

billymg: asciilifeform: ty, glad you like

billymg: but clicking into each ones shows the spam wave in the last 24hrs

billymg: if you look at the homepage now and scroll to the bottom can see the TRB nodes have returned to normal

billymg: asciilifeform: ty, still very rough around the edges and incomplete but wanted to get something up

billymg: asciilifeform: in most recent poll yours returned 25 http://bitdash.io/nodes/205.134.172.27-8333

billymg: and i only started seeing that happen as of yesterday when i first reported it, before TRB nodes were always connected to a reasonable number of peers and a high percentage were "real" (i.e. responded with valid version msg when probed)

billymg: looking at whaack's node for example, the most recent getpeers returned a reasonable 66, but in the last 24hr it has been jumping around to the 3-4 digit range

billymg: asciilifeform: for when you return, i actually *just* made live a WIP version of the www. you can see from the historical poll results on .27 that the spam peers seem to come and go in waves http://bitdash.io/nodes/205.134.172.27-8333

billymg will check again later

billymg: !w peers 205.134.172.27

billymg: asciilifeform: that's strange, 29 connections seems inline with what i was seeing before

billymg: asciilifeform: the peer lists have been captured (the crawler is now stores probe history up to N probes, as set in conf). i'll dump the results somewhere permanent before the cap is reached

billymg: asciilifeform: so possibly is passing phase and node count/quality will return to normal after some time?

billymg: asciilifeform: it's consistent across all trb nodes that my crawler has picked up, and all in the last 1-2 hours (first time i've observed it since running this thing)

billymg: actually all of them seem to be returning significantly more spam peers in the last 1-2 hours

billymg: another trb node, 205.134.172.26, also jumped in the same timeframe (from an average of around ~175 connected peers to now 1885 in the most recent poll - most of which appear to be spam)

billymg: asciilifeform: is 205.134.172.27 your node? my crawler is showing that sometime in the last hour or so it jumped from around ~40 connected peers (mostly good) to ~1200 peers (mostly fake/spam)

billymg: http://logs.nosuchlabs.com/log/asciilifeform/2021-06-30#1041901 << ty

billymg: so you request an API key from them, probably start paying once you hit a certain threshold, etc.

billymg: asciilifeform: yeah, i haven't looked too hard yet but all the "geolookup" libraries seem to depend on one of the few large API providers, e.g. https://apilayer.com/

billymg: for doing the IP lookup at least, and then his scripts spit out JSON that apparently 'OpenStreetMap' can load in to draw points on one of their maps

billymg: asciilifeform: ah, the map? that is a cool feature, he also appears to be using the main 3rd party API provider for this service

billymg: asciilifeform: thanks, i'll look at that too and see if there's anything useful to extract

billymg: i'm also looking at https://plotly.com/python/ as a potential library for rendering charts/graphs on the crawler www, in case anyone has experience with either, or has other recommendations

billymg: signpost: how did you make the network graphs on wot.deedbot.org? it looks a lot like output from d3.js

billymg: yes, and is the default config out of the box when pressed

billymg: or anything controversial even

billymg: asciilifeform: yes, fully. the theme thing, chan list refactor, log root knob added to the config, and some readme cleanup. there shouldn't be anything offensive in there

billymg: asciilifeform: ty, if you find anything that causes you to barf let me know and i'll update

billymg: http://logs.nosuchlabs.com/log/asciilifeform/2021-06-30#1041812 << if new patches are going to be written, have you had a chance to review the two i published yet? i'd be happy to make updates/regrind if either are offensive enough to prevent patching on top of them

billymg: asciilifeform: ah

billymg: asciilifeform: even if he thought you were rich it still doesn't make sense. 50/50 scenario where one guy puts up the dough, the other the sweat, is extremely common form of partnership. shouldn't even matter if both have dough to begin with, unless one is paranoid and thinks the other is a grifter

billymg: the whole idea would've been so that asciilifeform (or other productive members) could quit day job, not lead cushy life

billymg: asciilifeform: re: fought for motherland: even soldiers receive room and board and small level of pay

billymg: my point was mostly just that salary != disenfranchised bare minimum effort worker in all cases

billymg: asciilifeform: yes, by equity i of course meant non-scam equity from person who plans to make good on the offer

billymg: and only given to those who show they are committed for more than the money

billymg: asciilifeform: it can be an arrangement where very low salary, equity stake, etc.

billymg: it would still not*

billymg: thimbronion, signpost: add me to that list as well. this was one of my biggest questions. man could've paid salaries for a dozen people to work full time and still not have more of an effect on his worth than a rounding error

billymg went straight for index.html and opened that up before looking at anything else

billymg: signpost: ty!

billymg: asciilifeform: or anyone else, do you know if there's a way to force the browser to recognize the files as html? most of them don't have the .html extension so chromium just displays the source rather than the rendered page?

billymg: a few errors when extracting but it looks like it's all there

billymg: asciilifeform: got it and i'm seeding now

billymg bbl, dinner

billymg: ah, just as in, added the torrent file

billymg: asciilifeform: works, ty

billymg: asciilifeform: don't have permission to download

billymg: asciilifeform: cool, sounds good, will let you know when i'm ready to make the switch

billymg: asciilifeform: i'm wondering 1) could you clone the rk drive to a new machine if i lease another from you and 2) could i, at that time, cancel the remainder of the rk lease and have it credited to the new box?

billymg: asciilifeform: i'm also getting close to making the crawler site live, at least a basic version so that others can take a look and provide feedback. at that point i think i'll need to upgrade from my rk to a bigger rig, especially since i also want to run whaack's block explorer on there

billymg: http://logs.nosuchlabs.com/log/asciilifeform/2021-06-28#1041358 << i killed my download of this since it showed 1-2 days remaining, will wait for the torrent and then seed

billymg was kinda hoping for an answer to "who is it that should be hung"

billymg: asciilifeform: sad state of affairs

billymg: asciilifeform: if the clowns being tolerated are a symptom of the complexity collapse, (because "real work" is superfluous anyway), what then triggers the complexity collapse

billymg: asciilifeform: ftr my use of "trannyism" was meant as wide brush label for all of the various "social causes" over actually-doing-things people

billymg: i guess i had the opportunity to see this transformation play out in real time. when started at the company was about 150 lean, when left, 1500

billymg: right

billymg: although i guess what you're saying is doesn't matter, there is no more "real work" to get done

billymg: done*

billymg: the problem is those "games" they're given to play interfere with those trying to get real work don

billymg: asciilifeform: i.e. the "back to no indoor plumbing" path we're on?

billymg: asciilifeform: how is the actual software development "newly superfluous"?

billymg: once i found myself self-policing my language in meetings i knew there was no future in it for me

billymg: i should expand, i'm not even fundamentally against "public facing" roles, and in fact did quite well in them, what i am allergic to is the trannyism that infected all of it in the last 5 or so years

billymg: and realizing that any degree of self-sufficiency in the future will require ability to code and configure servers

billymg: it was also after discovering bitcoin and the logs

billymg: asciilifeform: that is actually what made my shift from the frontend to the backend more active. seeing that the frontend will always be public facing

billymg: i have a small client now that is happy that i just do what's needed and do it well

billymg: plentiful but always of the "let's get excited" variety

billymg: asciilifeform: interface designer by trade, which often included implementing the interfaces, which led to more and more JS as React became "the standard", which eventually led to "full stack" programming

billymg: asciilifeform: ah, this also makes sense

billymg: but i suspect it's not something you "find" but that you cultivate over decade

billymg: asciilifeform: this is what i would love to find

billymg: asciilifeform: isn't the "stack" thing only +ev if "salaried" position? if working by the hour then doesn't matter if 10x3 or 30x1

billymg: asciilifeform: this is the only kind i'm doing these days and it's definitely an improvement

billymg: asciilifeform: i can see it but it seems like a bit of a stretch to compare drilling:harley fixing :: programming for client:programming for self

billymg: asciilifeform: heh, interesting

billymg: probably not more drilling

billymg: asciilifeform: but what does a dentist do then on the weekend?

billymg: asciilifeform: it resonated with me, managers openly graded individuals on "passion"

billymg: asciilifeform: the whore analogy is accurate. saw posted somewhere how working in silicon valley is basically whoring because not just expected to get work done but expected to show enthusiasm/excitement at Nth rewrite of dickpic app

billymg: asciilifeform: yeah i guess that's the issue, once the problem is solved the "repeating/maintaining it for customers" part isn't any fun

billymg: if i suspect you thoroughly enjoy assembling FGs that doesn't mean i still wouldn't want to pay what it's worth to me

billymg: asciilifeform: how much you like what you do doesn't factor into it with potential customers though, does it?

billymg: the "in my non-commercial hours" seems like a bug

billymg: asciilifeform: something that you work on out of "internal motivation" but that also pays the bills / allows you to "own your own time"

billymg: asciilifeform: i'm forever interested in coming up with something that somehow satisfies both of these: http://logs.nosuchlabs.com/log/asciilifeform/2021-06-18#1040055 <> http://logs.nosuchlabs.com/log/asciilifeform/2021-06-19#1040103

billymg: http://logs.nosuchlabs.com/log/asciilifeform/2021-06-19#1040099 << ftr didn't mean 'musing' in the sense of "funny you should mention this thing that was already discussed before" but rather "funny how it took so long"

billymg: asciilifeform: right right. i'm just musing on how it seems much closer to what was talked about before as the desired model -- castles (chans, then) among a republic --and how it's coming about now

billymg: this seems much more the previously discussed "castles among a republic" model than meganet irc

billymg: asciilifeform: that makes sense

billymg: asciilifeform: growth in members is still desirable though, no? if for no other reason than the pile of work is far greater than the number of hands

billymg: http://logs.nosuchlabs.com/log/asciilifeform/2021-06-18#1039864 << this line made the significance of your new proposal click for me btw

billymg: asciilifeform: right, i guess i should've said, "stop in and say 'hi'"

billymg: neat, yeah that sounds pretty ideal

billymg: asciilifeform: so in dulapnet, instead of running znc on my rk, i'd run a relay. and if i find someone on twitter who wants to see what we're up to, they could connect to my rk, rather than having to run their own relay right off the bat

billymg: asciilifeform: yeah, disregard the laptop question, was answered by "shared telephone line"

billymg: asciilifeform: ah, gotcha. so just that at least one relay operator on the net needs to allow "regular users" to connect to their machine

billymg: and home ip changes, or shuts lid

billymg: ...which was my follow up, if someone runs relay on their laptop

billymg: well if person == machine

billymg: asciilifeform: the big question i have in my head right now is, will it be at all possible for someone without a relay of their own to participate in any capacity?

billymg: asciilifeform: interesting, that clarifies it a bit more for me

billymg: a policy like that still ensures it doesn't bloat into another fleanode

billymg: but even so i'm not sure i see the harm in the 'one relay grants you one chan' policy either

billymg: perhaps it still works though, and the new net now has 5-6 relays consisting of the users wanting to join in the other topic. so not the robustness of 30 relays, but more than a single node at least

billymg: not even necessarily 'tired of' but only 'tired of in this one room where we all congregate'

billymg: reason being, let's say there is a healthy group of 40 active users on dulapnet, 30 of which are running their own relays, under 'no chans, just nets' anyone wanting to continue a discussion that users in main channel are tired of will lose out on the robustness of the 30 relays

billymg: i liked earlier proposal of want chan? maintain relay

billymg: http://logs.nosuchlabs.com/log/asciilifeform/2021-06-17#1039741 << i agree that channels seem kind of silly with only a dozen or so users, however i'm not sure how i feel about "no 'channels', just nets"

billymg: http://logs.nosuchlabs.com/log/asciilifeform/2021-06-18#1039826 << ftr i fully support this. one of the things that drew me to the old republic was its denouncement of the 'star-topology'

billymg: also possible to have multiple znc users, i haven't tried this route yet but may do that next

billymg: supposedly possible to have multiple networks in a single znc instance https://wiki.znc.in/Connecting_to_ZNC#Connecting_to_Multiple_Networks

billymg: if anyone else managed to get znc working please let me know the proper hoops to jump through

billymg: asciilifeform: ah ok

billymg: two networks* in znc

billymg: for now i just connected direct as a sanity check to make sure i had the pw right

billymg: i still haven't figured out how to maintain two connections in my znc

billymg: asciilifeform: ty for the explanation, will hop over to your net soon

billymg: i'm not too familiar with how IRC works but why would fleanode migrate servers and not bring the old data with them, e.g. the nick and chan registrations?

billymg: asciilifeform: this guy looks quite interesting, have you tried inviting him into #asciilifeform?

billymg: or znc issue

billymg: asciilifeform: ty, somehow my nick unauthed with freenode (i think?)

billymg: asciilifeform: yeah, i figured it was just taken from the previous logger without modification, and i wouldn't expect much attention to be given to the css considering it was a wartime effort

billymg: asciilifeform: awesome, sounds good, ty

billymg: asciilifeform: i published those logotron patches if you want to take a look

billymg: if that's the case then i misunderstood earlier

billymg: asciilifeform: ahhh, so you were talking more about saving state somewhere in the logger, to prevent unnecessary queries, rather than simply doing a more efficient query?

billymg: i can rewrite the N queries into a single query with N unions, and takes about the same amount of time, maybe coupla ms less. but if db is on a separate box this would be a decent gain considering the network latency cost of making N separate requests

billymg: asciilifeform: i'm just trying to get to root of the sharp edge you mentioned earlier

billymg: sure

billymg: a dozen give/take

billymg: asciilifeform: i just don't have a frame of reference to know if 37ms is a bad time for that query or not

billymg: asciilifeform: here is the query plan for one chan, and the time it takes to execute: http://paste.deedbot.org/?id=5ZXr

billymg: asciilifeform: yes, that makes sense

billymg: http://logs.nosuchlabs.com/log/asciilifeform/2021-05-22#1037281 << by "what it oughta do", do you mean what postgres should be doing in its query plan, but isn't? or that query should be rewritten to coax postgres into using a more efficient plan?

billymg: asciilifeform: will do

billymg: very minor savings (when db/proggy on the same box), but probably worth refactoring just for neatness and for users who may want to set up their db on a separate box

billymg: asciilifeform: actually, perhaps i spoke too soon. more complete results here: http://paste.deedbot.org/?id=mT4_

billymg: my results on my rockchip: http://paste.deedbot.org/?id=v6kO

billymg: and performance is worse using union vs just doing individual queries

billymg: if so, it's a tricky query. only way i could see to get it into one SQL query is with a 'union' statement between, essentially, num_chans queries

billymg: http://logs.nosuchlabs.com/log/asciilifeform/2021-05-18#1036533 << is this the line you're talking about? inside the loop in gen_chanlist?

billymg: asciilifeform: gotcha, do you mind if i separate the version history from the readme? so that the readme always reflects the current state

billymg: perhaps the list is stale, my copy came with a ZNC converter and pagination of search results seems to be there as well

billymg: asciilifeform: another question: the readme lists some yet to be implemented features: http://paste.deedbot.org/?id=ziN8

billymg: kk, will do

billymg: want me to update the readme then?

billymg: ahh, ok

billymg: it was a part of the setup that confused me initially. i have all the components running on the same box so in my setup i didn't have to fuss with any of that, i just added the mod_wsgi vhost config (pointing to my reader.py) to apache and it handled the rest

billymg: asciilifeform: in the logotron readme where you talk about a proxying www server, and running reader.py separately then pointing the www server to its port, are these instructions assuming that the box running the bot and reader is different from the box running apache?

billymg: http://logs.nosuchlabs.com/log/asciilifeform/2021-05-20#1037129 << ty! gonna see if i can publish the patch today

billymg: http://logs.nosuchlabs.com/log/asciilifeform/2021-05-19#1036800 << turns out i did have that knob enabled. all 114 lines of #billymg history are available there if you want to add the channel to your logger

billymg: since most people are in this chan i'll mention here that i mirrored all the logs from http://logs.nosuchlabs.com/log on http://logs.bitdash.io and joined the respective chans with my bot. if anyone doesn't want my bot hanging around their channel just let me know and i'll have it leave

billymg: trinque: lol fair enough

billymg: i know i'm not gonna be the guy who writes a new linux, or new tcp stack, etc. but hopefully i can at least find one or two people who can help with that

billymg: trinque: fwiw my working on the btc network crawler and new www to display obvious centralization of network is to attract more hands

billymg also doesn't write/grok at sufficiently low enough levels to undertake these tasks

billymg: trinque: to which i replied, where are the hands? why not first fossilize and polish something that works before rewriting everything

billymg: trinque: ahhh. i wasn't the one who brought it up though, asciilifeform did

billymg: trinque: not sure what you mean

billymg: asciilifeform: this i understand, and movement towards new silicon for running spherical lisp machine as well, but i don't see army of hands with spare time and resources for doing all of this

billymg: asciilifeform: pip isn't needed though

billymg: asciilifeform: that strategy seems to have worked well enough for dulap gentoo, why not just save old version of flask and psycopg2

billymg: asciilifeform: why does it matter what builds on heathen systems?

billymg: or at least the pieces needed

billymg: so rewriting all of those as well

billymg: right

billymg: so just no psycopg2

billymg: asciilifeform: does 0 deps mean can't use a database, or db must also be written from scratch?

billymg: for the bot or the www portion?

billymg: asciilifeform: what do you want the seriously improved logotron to be?

billymg: asciilifeform: yeah, that part is no problem

billymg: not including lynx and other terminal browsers, obviously

billymg: via css media query, which i'm pretty sure works on any browser anyone here is using (bold assumption, i know)

billymg: however, detecting the w < h also doable

billymg: my suggestion of adding a URL param would not be "auto", but selected the way "show inactive chans" works now (roughly)

billymg: asciilifeform: sorry, i just reread http://logs.nosuchlabs.com/log/asciilifeform/2021-05-19#1036833

billymg: presence of which loads an extra coupla lines of css

billymg: would work with ?navtop=true in url or something

billymg: asciilifeform: answer is yes, and even wouldn't use "media queries" (css definitions targeting different viewport widths)

billymg: asciilifeform: lol in that case i don't see why the fuss over the ~200px sidebar

billymg: asciilifeform: meaning your browser windows are never less than 2160px wide?

billymg: asciilifeform: not sure how wide the support is but iirc been around decade+

billymg: asciilifeform: if you resize your browser window to less than 768px wide you will see it switch the layout to something closer to classic (chan list on top, arranged horizontally)

billymg: asciilifeform: is possible to auto toggle with css

billymg: asciilifeform: i could make a separate patch for the theme

billymg: asciilifeform: ah, i see. imo i like the vertical sidebar chanlist in the desktop view (it mimics many other interfaces i use) -- but will note to others, it's a simple flip of a css switch, no markup changes between what i showed earlier as "classic" and "bitdash" themes

billymg will publish a patch of changes (minus personal theme) when it's cleaned up, most likely on my blog by tomorrow

billymg: additionally, the stacking in lynx as a side effect of converting the chan list table to a ul will be reverted

billymg: asciilifeform: also, minor clarification re: your comment of "eat full vertical chunk of the screen" -- when i said mine stacks on mobile i meant the layout (e.g. chans will be at top of screen, but arranged horizontally)

billymg: nice

billymg: ahh

billymg: asciilifeform: i haven't thoroughly looked through the bot code yet, are there existing knobs so that two bots don't both echo pasted loglines?

billymg: asciilifeform: will look and if not will get that for you

billymg: would this be first instance of two nsabot loggers in the same chan?

billymg: asciilifeform: yes, would be happy to have it hosted on yours as well

billymg: and for "mobile" (< 768px wide) it stacks

billymg: asciilifeform: i haven't tried it on an ipad yet, but the css tightens it up on narrow displays (you can try by resizing your browser window)

billymg: so now user can set to /foo/ or whatever they like

billymg: you'll see mine is living at /

billymg: and i made the /log a knob in the conf file

billymg: lastly, i did fix one bug i noticed with the search bar, where it would cause a server error with queries less than 3 characters

billymg: asciilifeform: in my dark theme, what do you think about the chan list in the right sidebar? does this work well on your vertical monitors? (i tried keeping it very narrow)

billymg: asciilifeform: ok, for the purposes of lynx rendering, i will revert the chan list to a table so that the channels are displayed horizontally

billymg: asciilifeform: same here, i've got different themes going in different windows and machines

billymg could spend a coupla hours making a borland.css for the logotron that mimics this (not suggesting that's what asciilifeform was suggesting, but could)

billymg: asciilifeform: neato

billymg: asciilifeform: blue?

billymg: http://logs.nosuchlabs.com/log/asciilifeform/2021-05-19#1036774 << this is in reference to lynx rendering or the css-ized version?

billymg: asciilifeform: do you use light theme for all terminals and text/code editors as well?

billymg: wanted to see if you prefer it that way or if you'd rather see it in the table like before

billymg: asciilifeform: regarding lynx, would also like to get your feedback on the one change there. i changed the chan list from a table to a list, so lynx now renders it differently

billymg: asciilifeform: also note that this css conf change has zero effect on lynx -- will always render the same there

billymg: but point is, it was a one-line change in the conf. anyone can write their own css file, no need to edit any other code

billymg: asciilifeform: now try reloading and you will see my new theme (may not be to everyone's taste)

billymg: now, the fun part, i will switch one line in the conf to point to a different css file...

billymg: ok, now if you refresh you'll see a few more chans

billymg: one sec, lemme enable a few fake chan links for demo purposes

billymg: asciilifeform: good! that was the point for the first pass on the "classic" theme (the changes are in the html/css)

billymg: as you can see i'm only tracking one channel now, but i have some dummies hardcoded in to show what that looks like

billymg: asciilifeform: ok, here it is with "classic.css" http://logs.bitdash.io/

billymg: asciilifeform: ready to see a quick demo of logotron html/css updates?

billymg: maybe ltc

billymg: thimbronion: i was a little surprised they brought back doge this year, i wonder what it will be next time

billymg: thimbronion: LOL

billymg: asciilifeform: will test there as well

billymg: asciilifeform: ah, got it. and that's what i was using before so never noticed it not centered

billymg: for me it's firefox 78 that's rendering it to the left

billymg wonders if anyone else is using a browser that renders it on the left side

billymg: asciilifeform: do you mind if i ask which browser/version?

billymg: i've always seen it centered, but just opened in a different browser and noticed it's left aligned

billymg: asciilifeform: speaking of inconsistencies, do you see your search bar as centered or left aligned on your currently live logger?

billymg: asciilifeform: when i originally started as an intern one still had to support IE6, after that the new stuff is a walk in the park for me, hardly even have to think about it while doing it

← | →