40700+ entries in 0.284s
mod6: got
a bit of
a fade?
BingoBoingo: Guess who finally got
a trendy Latino haircut!
phf: the content of these articles is as much of
a mystery to me as they are to the log reader. maybe it's zavalishin writing obituary for bardin, but i doubt it.
phf: "Reading in the Dark: Does fiction matter in
a post-fact age? harpers.org"
mod6: Tuning the best interval between when we would query
a connected host seems like it might take some doing. Right now, looks like 20 minutes (in ben's vpatch).
mod6: yaeh, i have
a feeling that's kinda what was happening to me. but untested yet.
mod6: ah, no haven't had
a chance yet.
BingoBoingo: I have to say the silence this morning is
a bit extraño
mod6: "If you try to throw yourself upon the wheels of the great machine, you're gonna have
a bad time, Komrade"
a111: Logged on 2016-10-04 15:36 asciilifeform: ACHTUNG, PANZERS! pc engines 'apu2' (the board with the intel nics - vs. 'apu1', with realtek) , turns out, is crippled, hdt probe barfs with it, the cpu is reputed to have
a drm fuse set.
laplinker: hmm... i might actually do that, for the xdp3, they should be popping up more and more the more time passes and most folks use xhci, but thats just
a guess
laplinker: sometimes those xdp3 boxed pop up oon ebay, i found mine for about 50$
a few years ago, but now intel is debugging over usb3
laplinker: you should really get one of those minnow boards and
a lure board, works OOB with intel DAL at VLV2 config
laplinker: yeah, im not so sure i can afford something that costs: "send me
a quote" :)
laplinker: oh, its
a long story about AMD, HDT, and looking for sage probes online :)
trinque: and also, handled incorrectly that'll be
a lot of boxen that are suddenly unreachable
BingoBoingo: getting some from them and then resubmitting my application to LACNIC. Gotta utilize 64 ip addresses before LACNIC will allocate
a /24
BingoBoingo: !~later tell mircea_popescu
a MUSL box is now ready to plug in and serve
a form of Qntra... after
a weekend of head banging
a111: Logged on 2018-01-29 01:14 asciilifeform: !~later tell trinque do you have
a working gnat on ppc (g5 or any other) ?
ave1: Meanwhile, I'll fix the links, also I found the problem will report
a little later
ave1: Hmm, I'm using mp-wp and I cannot find
a setting that is supposed to scrub links or allow them. I am looking into it...
☟︎ ave1: I do not get "TODAY'S LOG' IS
A LINK" ...
ave1: aha, I made the check for the email to
a check for the website (needs to be 6 chars long)
ave1: I did! it is more
a question for other readers
ave1: Which makes the process
a stupid trying / building / running / failing exercise.
a111: Logged on 2017-11-29 01:38 asciilifeform yet again, for 3rd time in 2 yrs, attempted and failed to build
a 'zero foot print runtime' for gnat -- to abolish the 3MB of liquishit it shits into every executable.
ave1: In the end it took only
a small set of changes, but finding these was no fun...
ave1: I'm enjoying the series
a lot! (I've now just read ch2 and starting to ingest it...)
xenmen_: (The catch-up, that is; rereading older material I spread over
a few weeks to
a month)
xenmen_: I come by ~once
a year, an annual tradition, to reread
mircea_popescu: behold, that there was an empty lot with
a lot of hungry people in it that then later became
a restaurant.
phf: asciilifeform: i.e. ??``[(let ((
a #x]#[) (b #x]#[) (c #x]*#[) (sz #x].0.1-#[)) (= (nth-value 0 (truncate (*
a b) (1+ sz))) c))]
phf: asciilifeform: so i spent
a long time on your homework since it's
a kind of trick question, but the naive solution that you've provided will not always produce True for obvious reasons.
mod6: omg it is so hard not to babysit this thing after
a year of babysitting.
trinque: people don't show up at some empty lot hungry and then later there's
a restaurant
phf: shinohai: here's
a service you could provide: bouncers for sluts :) i believe e.g. znc supports multi-tenant
phf: yes, that's cause fg looks like some mil spec shit. you just needed to put inside
a cheap plastic box and stencil "fax modem" on it
mod6: ok 3 blocks behind, now just doing
a bunch of mem pool stuff. we'll see if it falls back behind
BingoBoingo:
A common recurring answer is they don't know because everyone they know goes to Miama with one bag and returns home with 5-10
mod6: but *shrug* maybe it just needed longer to talk to others. it had only been up for
a little while at that point.
mod6: if it does it again, we'll take
a look
mod6: well, probably wasn't "stuck" just takig
a long time or whatever.
mod6: in other news, got within 2 blocks of HEAD lastnight... never quite did make it... then for some reason fell like 80 blocks behind while stuck on
a block.
mod6: maybe the more specific with '
a' and 'b' is the ticket tho
mod6: <+asciilifeform> diff -uNr $1 $2 | awk 'm = /^(---
a|\+\+\+)/ << diff -uNr $1 $2 | awk 'm = /^(---
a|\+\+\+ b)/ << ya?
phf: mircea_popescu: i don't think it makes sense to talk about bit order on any of the architectures that we're discussing, since you can only address by bytes. big endian systems could have little endian bytes and you would never know, likewise little endian systems could have big endian bytes and you would never know. logically though there's no reason you can't view big endian to be
a bit reverse of little endian
BingoBoingo: Well, that everything we dig into turns into shit upon examinition show how great of
a handicap not having
a republic is.
mircea_popescu: but this is
a problem of display not of ontology, on small endian box.
phf: right, i'm more saying that neither have merit, and the only reason there's
a narrative is because you have to have one or the other, and by nature of having both in theory you have both in practice
phf: it's
a completely arbitrary decision, that has
a bunch of cons and pros based on time, hardware, architecture, etc. etc. big iron used to be bigendian, micros were primarily little endian, both for reasons
phf: i find all arguments pro or cons unconvincing, as encoded by elders i nthe very name, which is
a reference to gulliver's travels
phf: mircea_popescu: our current strategy with architecture support has been "it works on my machine, but if you want it to work on your machine submit
a vpatch", and it's been pretty consistently applied (e.g. my openbsd patch that's been floating independently, despite never getting vpatched).
mircea_popescu: asciilifeform do
a "cosmetics" at the end of the story, by then oyu'll have
a proper vdiff