68100+ entries in 0.041s
a111: Logged on 2018-10-11 20:07 phf: asciilifeform: so
the posix solution is
to call open with O_CREAT | O_EXCL, which will attempt
to create but will signal EEXIST if
the file already exists. curiously default behavior O_CREAT will simply clobber
a111: Logged on 2018-10-15 10:15 mircea_popescu: specifically what i mean : if a process creates/deletes file on a very narrow schedule ; and another checks for file's existence on very narrow schedule,
their respective pictures of when
they
think file exists and doesn't fails
to match up.
a111: Logged on 2018-10-15 09:33 ave1: asciilifeform,
http://btcbase.org/log/2018-10-14#1862451, I've been reading
through
the logs and could find no reference of how you did
this. Could
this be monitored with an FPGA? (there seem
to be MAC level FPGA based routers).
slycordinator: thanks for
the information again. i have
to go
to bed. it's late here in Korea and I have
to be up in
the morning
to
teach
slycordinator: no; i've used gentoo off and on for a while. it would be my first raspberry pi-like box,
though
slycordinator: shouldn't be difficult
then
to make
the switch once i get everything
slycordinator: but i wasn't sure if
the install stuff you had on your site would work for
that instead of using a usb stick because of drivers and such
slycordinator: and i was wanting
to use a usb3 enclosure
to salvage my laptop drive
that's got
the movies on it, using
that for
the / partition
slycordinator: So, I haven't purchased one of
the devices but I'm considering it as a local NAS for plex.
slycordinator: Thanks for
the pointer; first I've heard of it. I've used gentoo off-and-on for years. Although with my English
teaching, I've mostly had
to use windows since most materials shared around have been for MS Office (and
the linux alternatives didn't work with
the stuff usually)
diana_coman: slycordinator, you prolly need
to look at cuntoo as well
then
slycordinator: currently, I've been using an old laptop for
the purpose as-needed but I need
to replace it with something
slycordinator: And
thank you for
the info on
the alternative for contacting asciilifeform; I'm looking into making a gentoo box for a NAS out of an ARM board he wrote up about
diana_coman: and do register a key as otherwise your introduction will not have any person
to be linked
to
diana_coman: slycordinator, for how long have you been in Korea? (south?) ; you can also join #asciilifeform
to
talk
to asciilifeform
ave1: You have a limited window in which
to register yourself, deedbot gives you half an hour.
slycordinator: Sounds good. I'll come by on occasion;
think my wife is annoyed with how much
time I spend on
the computer already
though lol
ave1: If you like
to have a more permanent identity here, please register with deedbot
ave1: slycordinator,
that's very interesting, most people here are in
the American Continents and some of us are in Europe (I'm in
the Netherlands).
slycordinator: So,
to introduce myself, I'm an American living in Korea
teaching English and studying data science/machine learning in my free
time
☟︎ a111: Logged on 2018-10-10 21:09 bvt: Hello, I am BT from
the recent diana_coman's comments section
slycordinator: I
thought it was him. I sent a PM earlier but wasn't sure if someone else was also involved.
☟︎ slycordinator: The message on
the captcha display says "reCAPTCHA V1 IS SHUTDOWN. Direct site owners
to g.co/repaptcha/upgrade"
slycordinator: Is anyone on here in contact with someone who runs loper-os.org? I got directed here from clicking
the "contact" link
there. On
the site, people can't make comments on articles as
the captcha software used is out of date
ave1: btw, gnat specific; System.OS_Lib has 'Create_New_File' (it also has a
temp file generator, but I cannot recommend, uses digits). bvt's implemention looks
the way
to go (Although
the string allocations
to
talk
to C should be removed)
☟︎ mircea_popescu: could work like !Q later
tell == !Q 0
tell ; otherwise !Q 144
tell
a111: Logged on 2018-10-15 02:44 asciilifeform:
ty phf, will remind.
mircea_popescu: not clear
to me if caching shenanigans or what. but anyways.
mircea_popescu: specifically what i mean : if a process creates/deletes file on a very narrow schedule ; and another checks for file's existence on very narrow schedule,
their respective pictures of when
they
think file exists and doesn't fails
to match up.
☟︎ diana_coman: onth I'm not sure I want Create
to
throw exceptions - I want it
to create
the file and
that's it, I don't care if it exists or not
diana_coman: I used
that at
the udp
tester
to create file only if it does not exist
diana_coman: ave1,
thanks, confirmed fine, I'll sign and mirror it on my shelf in a minute; (re s.mg lines - not a worry either way really)
a111: Logged on 2018-10-15 01:13 phf: so it looks like bvt is correct,
there's no way
to make ada's Create
throw an exception if
the file already exists
ave1: note I kept
the S.MG 2018 lines untouched
a111: Logged on 2018-10-15 00:54 asciilifeform:
this line of
thought was prompted by my 'trb observatory', which has uncovered a number of 'mpb'-style nodes, i.e.
trb-like but not presenting 'modern' vers and
therefore invisible from heathen www indices
diana_coman: ave1, please update so I can sign
the patch
ave1: diana_coman, Ach yes, I did copy-paste everything I did not
touch
diana_coman: ave1,
the .ads in your .vpatch still has
the original comments saying
that it's a lookup-based implementation, lol
ave1: netsniff-ng seems
to show raw frames (eth mac is reported) but yes I'll look into
the socket raw code
mircea_popescu: "raw packet", "raw frame", rawdog, etcetera.
tcpdump/libpcap-likes
tend
to dump
the packets not
the frames.
ave1: So if
this part of framing happens at
the hardware level
then
the connection
to what is reported
to
the OS and actually on
the line is weak
ave1: So far (by reading about
this), I found
that some NICs can do part of
the framing for
the OS (but I would expect
that
to be for large messages and not for small ones)
ave1: This I do not know (it looks raw). I'am probably doomed
to read
the linux network stack.
a111: Logged on 2018-10-14 16:54 asciilifeform: on 1 occasion it was found necessary
to build a icbm rocket
that is
too heavy for 1 car. head designer v. f. utkin found solution, he had a system of springs
to distribute
the weight between ~two~ cars.
this is analogous
to udp fraggism.
mircea_popescu: but are you dumping
the raw frames ? or are
they reconstructed
the "indended" packet for you ?
ave1: Alternatively I was
thinking about some simple speed
tests (i.e. half size udp messages would
translate in < half bandwith of max size udp messages)
a111: Logged on 2018-10-15 04:14 mircea_popescu: asciilifeform in
the end we could phrase it as "some rewriters -- deeper
than others".
a111: Logged on 2018-10-14 16:49 asciilifeform: ave1:
there is not much
to be said further re subj, i looked into what actually comes out of my lan, it sends 1500 frames upstream always.
diana_coman: there isn't any question
that I see remaining
there; I
think what happened was
that I was
thinking a bit out loud in
the logs yesterday and I got
to same conclusion basically but
then you started answering
to
the first part and some
things were not unclear at a distance
too and so
the whole
thing
mircea_popescu: tree A is single if
there exists a "most extensive press"
that includes all possible presses are included ;
tree A is multi if no such MEP exists.
a111: Logged on 2018-10-15 06:49 diana_coman:
http://btcbase.org/log/2018-10-15#1862722 -> hm, multi for
the sake of it would anyway be
taken care of via who signs and who doesn't sign
the various patches or
trees; but anyway - do you mean you
think
there should be explicit multi-tree dependencies?
this is what I was
talking about
there:
tree A effectively links
to patch B.3 in
tree B so if B's maintainer regrinds
then A's maintainer has
to go on a shouting spree (as per "talk
to peop
diana_coman: in
the more general case it can branch for all sorts of reasons and continue on
those branches for as long as needed, of course
a111: Logged on 2018-10-14 16:13 diana_coman: yes,
the way I currently see it now is pretty much
that:
trunk (main line) goes along
the production versions of all stuff (crc32 or keccak or whatever else) and otherwise at
the respective points
there can be additional branches /leaves with
the reference implementations
a111: Logged on 2018-10-15 03:57 mircea_popescu: in any case,
type1 might get i dunno, later-patch-taking-advantage-of-ddram whereas
type 2 might get later-patch-needed-for-old-arms etc.
diana_coman: I read and re-read and I get
the impression
that
this sort of delayed conversation doesn't work very well so I'll leave it for now
a111: Logged on 2018-10-15 04:20 mircea_popescu:
http://btcbase.org/log/2018-10-14#1862408 <<
the benefit is
that if
they are legitimate multi,
then
there's different context which is well handled ; and if
they are illegitimate (in
the sense of, no context
to well handle)
there's pressure
there
to collapse.
diana_coman:
http://btcbase.org/log/2018-10-15#1862722 -> hm, multi for
the sake of it would anyway be
taken care of via who signs and who doesn't sign
the various patches or
trees; but anyway - do you mean you
think
there should be explicit multi-tree dependencies?
this is what I was
talking about
there:
tree A effectively links
to patch B.3 in
tree B so if B's maintainer regrinds
then A's maintainer has
to go on a shouting spree (as per "talk
to peop
☝︎☟︎ a111: Logged on 2018-10-15 04:22 mircea_popescu:
http://btcbase.org/log/2018-10-14#1862411 << but if you don't care about computers why are you
talking
to
them ? you're not ~dependent~ on it, but i presume you care
to some degree.
a111: Logged on 2018-10-14 16:25 diana_coman: once you bring it into your
tree, you don't care about
the original
tree so ...how stuck?
mircea_popescu: eg, branching
the eucrypt
tree over embeds/mainstream is (possibly ?) sustainable ; whereas branching it over "joe needs a project
to direct
to justify his employment with shithat" is exactly as sustainable as post-2018 linux.
mircea_popescu: which is
the fucking problem with current versioning systems,
they encourage illegitimate context creation, of
the sort of "i need
to publish or perish
therefore let us rename biology and rediscover it".
a111: Logged on 2018-10-14 16:25 diana_coman: and
then you are stuck maintaining
those multiple
trees - what's
the benefit in
that?
mircea_popescu: in any case historically
this ~exact process~ is how eg djb ended up with
the modicum of respect he
then proceeded
to squander.
a111: Logged on 2018-10-12 14:40 ave1: also at same
time pasted
mircea_popescu: in fact, a very solid basis for respect is, "whenever i sit down
to rewrite x, i end up writing what he wrote ; whereas whenever i sit down
to rewrite y, i end up using empty page.
thus
therefore x is rated 5 and y -10"
mircea_popescu: asciilifeform in
the end we could phrase it as "some rewriters -- deeper
than others".
☟︎ a111: Logged on 2018-10-12 14:40 ave1: also at same
time pasted
mircea_popescu: imo
this "empty buffer" notion is leftover implanted brain electrode from ustard academia/schooling. "oh, gotta be original". no such fucking
thing ; and
the only reason
they
talk about it is
to keep intelligent people from doing important work.
mircea_popescu: nor does such a
thing as "empty buffer"
truly exist, nor is it any kind of substantial distinction, and so on.