tmsr - Search: i

44600+ entries in 0.317s

asciilifeform: sometimes i see obvious spamola on trilema. but always assume that mircea_popescu let it through for lulz/collectiblevalue

mircea_popescu: that's the past 5, covering a coupla days. so i guess it's more than 1/day

asciilifeform: i'm satisfied that hanbot showed mp-wp to be usable on nfs

hanbot: asciilifeform kk, i'ma give it a shot

mircea_popescu: these are all distinct ips as at some point (maybe 2015 ish ?) i wiped dupes. but i get maaaybe 1 spam item in review queue per day

asciilifeform: hanbot: if you post one, i'd like to try it.

asciilifeform: hanbot: i'd luvv a working-with-nfs patchset on mp's wp. unfortunately my own wwwism proved insufficient to make it go.

mircea_popescu: i have Spam (17,852)

hanbot: 10k marked spam comments means spammy ips are stored and attempted comments from same don't show up in admin, subjectively over time it's seemed i need to spend less time mopping shitcomments. and i didn't post proper patches, but will.

mircea_popescu: asciilifeform i dun expect it is possible for it to not work if wp otherwise does work.

mircea_popescu: one thing it does is <input type="text" name="author18c6e55" with that hash changing daily (or how often you please) ; another thing it does is a "refuse from ips in spam list" ; there's more i might nor recall right off.

hanbot: asciilifeform i'm using mp-wp on nfs and spam works pretty much as expected ---though having ~10k marked spam comments helps.

asciilifeform: i'd luvv to find that i botched it with idjit mistake and in fact it worx

mircea_popescu: i think so, yes.

mircea_popescu: i have the feeling we discuss this periodically. what's dynamic ip to do with it ?

asciilifeform: and i dun have a dedicated slave to delete'em.

asciilifeform: i get a spamturd PER MINUTE

mircea_popescu: noob joins trilema, someone asks sto etcetera, noob answers.... "here's a thing i made", and it goes over well.

mircea_popescu: i wanna see jesus' carpenting school exam work, not that he fucking impressed a buncha journahos with his "profundity".

asciilifeform: i suspect that the 'benefits from reading x' circle is never all ~that~ much larger than the corresponding 'could have written x' circle.

a111: Logged on 2016-05-30 15:37 asciilifeform: but i find it hard to see how it is not obvious that 'every sewer rat can publish' is a considerable damper on culture

mircea_popescu: ygrade", i mean there WILL NOT, period, be anything he could have fucken found.

mircea_popescu: and i don't even mean ~subjective~ functions here! it's not a case that "to the exceptionally gifted as identified by ivy league, worth as he is maybe $1000 with clothes included, the subjectively perceived marginal value of information is slightly negative. no, no, no, OBJECTIVE, you can tell him whatever the fuck you will and he will not have any use for it -- i don't mean "he won't be able to find any use, above his $1k pa

a111: Logged on 2018-01-20 22:55 diana_coman: http://btcbase.org/log/2018-01-20#1773347 <-as I realised I never said this anywhere: this trouble with no-pingbacks on page is part of the reason why I have post+page for this sort of thing (the code reference shelf is one, the bac data another) - so that pingbacks can go at least to the post introducing the page; ugly workaround of sorts

mircea_popescu: http://btcbase.org/log/2018-01-20#1773406 << yeah i think ima have the page style changed. ☝︎

asciilifeform: i thought it was pretty clear .

phf: i don't grok the first homework: do you mean produce random computations, and programmatically verify ffa against them?

trinque: http://btcbase.org/log/2018-01-19#1773147 << I've got several cleanup patches coming before the walletsnip. anybody have a serious objection to me introducing a changelog in the first? ☝︎

asciilifeform: i.e. Product is only written to as result of the FZ_Mod call , http://btcbase.org/patches/ffa_ch7_turbo_egyptians#L148 , which in turn , inside FZ_Mod, is a clean 1-shot, http://btcbase.org/patches/ffa_ch7_turbo_egyptians#L102 . QED.

asciilifeform: http://btcbase.org/log/2018-01-17#1771712 << at the risk of thread necromancy ( item was candidate for ch8, so i gotta ) this is false. FZ_Mod_Mul as seen in ch7 IS bufferized, because ch7's FZ_Mod is. ☝︎

a111: Logged on 2018-01-20 17:12 mircea_popescu: i have a coupla problems in re http://btcbase.org/log/2018-01-20#1773273 that design input is welcome on. 1. the intuitively correct format for this would be to make it a page (like http://trilema.com/eulora-shop/ ) rather than an article. but mp-wp pages don't take comments/pingbacks, which im starting to think is the height of idiocy.

diana_coman: http://btcbase.org/log/2018-01-20#1773347 <-as I realised I never said this anywhere: this trouble with no-pingbacks on page is part of the reason why I have post+page for this sort of thing (the code reference shelf is one, the bac data another) - so that pingbacks can go at least to the post introducing the page; ugly workaround of sorts ☝︎☟︎

asciilifeform: i fucking hate printer-generated pagebreaks.

phf: also the whole exercise only becomes worthwhile if i can also produce reasonable html out of the cweb source, since there's general allergy here to pdf/ps

phf: it's considered, i'm just thinking from which side to start unraveling this spitoon

phf: works out to about 170 pages of very nicely typeset source code. this is actually first time i'm working with printed source code proper (unlike ascii i prefer to work on-screen) and it's quite a joy

phf: dvi>ps i had to use a full tex install, because i haven't found a non-modern dvips implementation yet.

phf: in unrelated i managed to produce a literate programming dvi out of diff's source code from first principles. the actual content is not really a literate programing, since it was mechanically produced, and it doesn't yet compile since cat of headers and source produce a bunch of duplicate definitions, but first step.

diana_coman: hence the end of the line there: "I didn't really find a reasonable way to keep comments in"

diana_coman: ftr I'm still on wp purely because there has always been something else topping the pile; I just about switched to statically generated pages too but I didn't really find a reasonable way to keep comments in

diana_coman: http://btcbase.org/log/2018-01-20#1773347 <- I think pages should take pingbacks! and tbh I never really understood why are "pages" so special; the way I see it, "pages" are just posts outside of the time-flow, "sticky" posts basically ☝︎

phf: http://btcbase.org/log/2018-01-20#1773313 << i wrote a little checker for eater, and the other one binary is asciilifeform_aggressive_pushgetblocks.vpatch.asciilifeform ☝︎

mircea_popescu: i... notice

asciilifeform: i wouldn't even have conceived of doing image munging with a php proggy. masochism.

mircea_popescu: wtf is the remote box even for, if not to do the shit i don't need supervised.

mircea_popescu: 2. if i upload items, i'd very much like to use the already extant uploader ; but that thing puts everything in year/month directories. tho i think this is actually a good thing, tell people "do not link item directly, link the code shelf itself".

mircea_popescu: i have a coupla problems in re http://btcbase.org/log/2018-01-20#1773273 that design input is welcome on. 1. the intuitively correct format for this would be to make it a page (like http://trilema.com/eulora-shop/ ) rather than an article. but mp-wp pages don't take comments/pingbacks, which im starting to think is the height of idiocy. ☝︎☟︎☟︎

spyked: http://btcbase.org/log/2018-01-20#1773311 <-- pingback -- yes. XML-RPC -- dunno; I am not running php, so having a file called xmlrpc.php in my tree seems a bit odd; although I guess this is particular to wordpress, not to XML-RPC ☝︎

spyked: calls should go only to WP links though, right? I don't remember how Wordpress did it (did it look for /xmlrpc.php ?). though I suppose it should work either way (non-WP links would 404 and that's that)

a111: Logged on 2018-01-20 11:42 spyked: I am working on adding comments (one of the yet-untested ideas is reading only gpg-signed comments), but that's a low-priority item on todo list atm

mircea_popescu: http://btcbase.org/log/2018-01-20#1773258 << i dun think this is such hot stuff ; i personally wouldn't bother to sign just to leave a comment. but if you care, here's an alternative : make a bot, have comments be sent to it via command. iirc danielpbarron had something somewhat in this vein, dun recall how automated tho ☝︎

spyked: (what I'm trying to do is separate the browser/linking text snippets part from the need to embed an entire js interpreter in there.)

spyked: yeah. ideally, I would like to be able to look at the link and figure out to which text snippet it refers manually. or at least have a way to compute without parsing html. hm.

mircea_popescu: no, i know, it is.

spyked: aha. in particular, trilema anchors are generated based on the occurence of js dom-tree elements. wait, I'm gonna find an example

spyked: but yeah, ^ w3m code is weird. and besides, I don't know how text highlighting is supposed to work in a text-based browser. highlighting seems to be a terminal thing, which is weird, because both w3m and lynx/links can be compiled with mouse support! which is essential for that js snippet in trilema, because iirc it relies on onMouse* events.

spyked: mircea_popescu, I wanna first get a working item (js highlighting for trilema anchors) then make everything link into a static binary, then post them. as they are right now, I expect them to be unusable on many systems (e.g. musl-gentoo).

mircea_popescu: at least with you i know it's not intentional.

a111: Logged on 2018-01-08 17:18 btcvixen: but i am female

a111: Logged on 2018-01-19 01:30 asciilifeform: in other folx who apparently had gone fullbore usgtard since i last looked : lubos motl. e.g. https://motls.blogspot.com/2018/01/korean-backlash-to-planned-bitcoin.html .

mircea_popescu: as to the chat, i'd much hear your malformed oppinions than http://btcbase.org/log/2018-01-19#1772852 ; http://btcbase.org/log/2018-01-19#1772915 ; http://btcbase.org/log-search?q=%22+orlo%22 ; http://btcbase.org/log-search?q=yarvin ; http://btcbase.org/log-search?q=Yudkowsky and the rest of the http://btcbase.org/log/2018-01-08#1767149 ☝︎☝︎☝︎

spyked: which reminds me that there are a lot of things that I've left hanging since december, e.g. the w3m-js shinohai found that I rebased, and the ns-js library. left hanging because I haven't yet figured how to make trilema js anchors highlight in w3m-js (or any other text browser)

spyked: hm. "This exists because sourceforge went down for a week during July 2015, making it necessary to maintain local copies."; yeah, publishing these makes a lot of sense. I don't know if I'd find all the libraries I used for the blog if I were to look for them again.

spyked: so I'd probably sign them separately?

spyked: hi mircea_popescu. sure, I remember russell's thing. but is that really a problem? (as long as there's a republic to squash the cocksure idiots)

spyked: shinohai, I use a self-rolled static page generator based on Common Lisp (CL-WHO) (and Pandoc for parsing Markdown, but pandoc and markdown are both optional, of course). I have the source code published on shithub, but it's mostly hacked together, so in this sense not very different from any static site generator. https://github.com/spyked/thetarpit.org <-- coad and text

shinohai: What do you use to generate your pages spyked, if I may ask? I have wp-mp of course, but my host charges more to set up mysql and crap so haven't ever done that yet.

shinohai: I was pretty impressed with your blog for sure, I am a terrible blogger myself. I might do better if I actually had a server to set up a blog on, but future project.

spyked: thanks shinohai. :D the plan is to (at the very least) periodically share items of republican interest that I'm working on (e.g. ada lisp). progress has been very slow so far, but it's picking up

spyked: I am working on adding comments (one of the yet-untested ideas is reading only gpg-signed comments), but that's a low-priority item on todo list atm ☟︎

spyked: http://btcbase.org/log/2018-01-19#1773164 <-- it doesn't send them, but now that you mention it, I might be able to add a (manually-operated) thing that does this. otherwise, thetarpit is just a set of statically-generated pages via handcrafted lisp-based contraption ☝︎

a111: Logged on 2018-01-19 19:11 phf: his 4th sig is non plaintext by the way, i threw it in since the eater ate, but this probably should be an anti-policy

spyked: http://btcbase.org/log/2018-01-19#1773176 <-- yikes! sorry phf, I fixed it (and double-checked, as I should have in the first place) --> http://lucian.mogosanu.ro/v/seals/ffa_ch4_ffacalc.vpatch.spyked.sig ☝︎

a111: Logged on 2018-01-19 16:21 asciilifeform: hey spyked ! i like your summary of the calc ops.

phf: i think it's adorable

asciilifeform: i'm still waiting for anybody to suggest a pill.

a111: Logged on 2017-11-10 10:14 diana_coman: for the curious there are in fact 2 prolific-stamped documents describing pl2303 and pl2303x; pl2303x seems to be a sort of upgrade to 64bits but why couldn't it be properly identified as distinct I don't know

shinohai: I knew of straight serial port, will have to buy solder

asciilifeform: i suppose those are scarce, rationed, these days. like car bomb.

mircea_popescu: what, am i to surmise we're not as cool as snowden and strauss-kahn ? or rather that the empire sunk to such a depth it is not even capable of finding mendacious prostitutes/chambermaids anymore ?

asciilifeform: i dun think i have enuff bits in my registers, to count the reddit scumola

mircea_popescu: oh, you mean other than the "i have my spurious "bitcoin is scamming" verbiage all over reddit " guy ?

asciilifeform: and i have this notion that this is the n-th, not 1st or even 2nd case of the pathology. but for some reason can't immediately find in the logs the others.

lobbes: http://btcbase.org/log/2018-01-19#1773183 << I dunno if 'interesting' was der Unterschied here; I'd wager it was more: "$realname1 appears to live outside of normal pantsuit spirit summoning range.. I'll try $realname2". Note this was -after- failing at summoning pantsuit spirit to #trilema itself ☝︎

asciilifeform: this prompted me to try the 'googleate own name' exercise. hilarious snoar, as always. finds 1) my www 2) buncha folx who are not related to me in any way i know of 3) lame softwarepatent from a defunkt slaveship i used to row on 4) bitcoin-otc

phf: his 4th sig is non plaintext by the way, i threw it in since the eater ate, but this probably should be an anti-policy ☟︎

asciilifeform: that's a rented 'ford' 150, i recognize the cardboard on the key

a111: Logged on 2018-01-19 15:50 asciilifeform: http://btcbase.org/log/2018-01-19#1773051 << where did i say 'bw is free' or that i compare with spam vps ? trinque posted a heathendom figure of 700 ( trinque can answer when he wakes up , for what bw, i presume 100/100 ) for comparison.

asciilifeform: i'ma link to this item.

asciilifeform: hey spyked ! i like your summary of the calc ops. ☟︎

a111: 2017-11-22 <spyked> http://p.bvulpes.com/pastes/yVbXl/?raw=true <-- most of it is config.sub and config.guess. two lines at the end may fix shinohai's troubles. anyway, I'ma post the whole thing (w3m+gc+js+whatever else) once I manage to do a static build.

BingoBoingo: Will do, and I will check to see if these numbers are 22% VAT inclusive or if I missed that again.

mircea_popescu: BingoBoingo dun worry about it ; sign the contract, explain about the cash, all is well. what gear did you have physically present i forget ?

BingoBoingo: I have seen the hall, the rack, the UPS, etc. Will meet with them and make sure nothing has been overlooked.

BingoBoingo: I will let the datacenter know we are moving forward with the whole rack 20/200 asymmetric.

asciilifeform: i thought BingoBoingo had no bank, can't eat wires..?

mircea_popescu: aite, anyone wanna do the above 3.5k wire and pocket the bitcoin implied by current lulzexchange ? before i give it to rando latino via bb's meatwot.

asciilifeform: as i understood, merely the pipe into rack. rack itself costs also.

mircea_popescu: this is a whole rack being discussed here, i dunno if you realise.