whaack: lobbes: erm following the #t thread... at least in mp's example that is a string i.e. alphabetical comparison. you can't use < or > to sort dates unless you have YYYY-MM-DD format or are converting the date string into some sort of date type.
whaack: s/to sort dates/to sort strings representing dates/
lobbes: whaack: technically I was converting a unix timestamp (e.g.1459468727) via e.g. line_time = time.gmtime(float(1459468727))
lobbes: then I was running that through time.strftime(%d %b %Y, line_time)
lobbes: which, you're right, converts it to string huh
whaack: lobbes: okay but if you do the comparison on the string returned from time.strftime you are going to have the same problem.
whaack: mhm
lobbes: whaack ty, this makes sense to me now at least haha. Was indeed an implementation issue on my end
whaack: lobbes: np.
jfw: http://logs.ossasepia.com/log/ossasepia/2019-12-04#1011900 - thanks diana_coman. re "how secure", sure - expected number of clock cycles required to break (collision, preimage etc.)
ossabot: Logged on 2019-12-04 15:27:38 diana_coman: jfw: hah, certainly makes for less tense reading! can confirm too that it builds and matches expected hashes; further than that, the code will require some pen-and-paper reading though; re "as secure as SHA3", do you have some sort of "how secure" measurement at all?
jfw: granted there's no proof on lower bounds there, but upper bounds can be found.
jfw: In thinking about the sponge construction, I had convinced myself that 2^capacity was such an upper bound. I'm struggling to precisely recall why, perhaps I should try to have it out as a proof
whaack: diana_coman: EOD Report: I did roughly 9hrs of saltmines and 30 mins of journaling. I practiced the guitar for 1hr as well. An expected-unexpected task was I had to exchange the gas cylinder for my stove, which took a little bit over 45mins.
jfw: but if so, smaller capacity is less secure in that sense. However I have no idea whether larger capacity might make it less secure in some other way (more frequent iteration of the permutation 'leaking bits' or something)
jfw: http://logs.ossasepia.com/log/ossasepia/2019-12-04#1011901 - interesting, it was deliberate (e.g I ended up with same for yrc genesis); I'd been thinking of each V 'project' as its own namespace for patches just as with the files in its tree
ossabot: Logged on 2019-12-04 15:27:54 diana_coman: jfw: that "genesis.vpatch" is a horribly-chosen name.
jfw: or are main.c or Makefile also horrible names?
diana_coman: jfw: eh, re genesis vs main/makefile - don't mix stuff there really; the idea with V overall is of building yggdrasil not just having tree-stumps all over the place (and all of them starting at genesis because ofc!)
ossabot: (trilema) 2018-10-15 mircea_popescu: kinda the fucking point, building yggdrasil over here.
diana_coman: how do you reason anyway re similarity genesis-main/makefile? because "genesis" is whatever vpatch is the root of a tree but not really substantially different in any way from another vpatch otherwise.
diana_coman: whaack: sounds like a normal saltmines working day,huh.
diana_coman: http://logs.ossasepia.com/log/ossasepia/2019-12-05#1012033 - this is pretty much the core trouble - there is no clear proof of "this is indeed more secure than that"; you can pick and choose various criteria, sure, but what they mean exactly is not that clearcut.
ossabot: Logged on 2019-12-05 00:53:58 jfw: but if so, smaller capacity is less secure in that sense. However I have no idea whether larger capacity might make it less secure in some other way (more frequent iteration of the permutation 'leaking bits' or something)
whaack: diana_coman: it was normal indeed. no ocean waves or ladies over for coffee to interrupt the grind at least.
dorion_road awoke to a thin, sticky blanket of wet snow to shovel. good for snowballs, not sledding.
dorion_road: I'd appreciate an !!up in #t when someone's available.
whaack misses skiing.
diana_coman: whaack: ha, can't have it all?
diana_coman: dorion_road: !!up done.
whaack: diana_coman: aha well not all at once. I would love to go skiing on a trip to Europe
diana_coman: quoth Eulora: can't have it all - where would you put it?
diana_coman: whaack: so ...plan that trip? you do have quite a lot of stuff to sort out first though, iirc.
whaack: diana_coman: Aha that reminds me also of some George Carlin skit MP linked to once about how life can be seen as a game of figuring out where to put your stuff. The eulora servers are back up, correct? I need to get in there if not just to read the MOTD's
diana_coman: whaack: the game server is up ; the webpage is not up though; do you have the client?
diana_coman: eulora is more of "figure out how to hold on long enough & make the most of it"
whaack: diana_coman: I do have one installed on this computer...but I believe it is out of date / I didn't wind up getting it working. The last working client I had is on a desktop that didn't make it to Costa Rica.
diana_coman: iirc there is no incompatibility problem as such ie at most you need to update the server IP.
whaack: diana_coman: Towards the end of January I will plan a trip to Europe. I have a more immediate need to plan a border run by ~January 18th. Which prompts the question: jfw and dorion_road: are either of you available to meet for dinner/coffee/etc somewhere in Panama City between now and Jan 18th? My mother is visiting Costa Rica between Jan 5th - Jan 12th I believe, so those dates will not work.
whaack: diana_coman: Okay, ls -l tells me that I set up Eulora in 2015...I remember having trouble getting the client to run and ./euclient gives me 'cannot execute binary file'. If I find some extra time I will try to get the game installed on this machine, otherwise I'll wait until I have my new comp built.
diana_coman: whaack: that sounds like some mismatch there ie you have it compiled with something other than you have running on the machine currently.
diana_coman: whaack: do you have a eulora account already?
whaack: diana_coman: Yes
diana_coman: so then possibly a recompile + change of server IP is all it takes.
diana_coman: anyways, not a priority atm.
diana_coman: jfw: this para is not very clear at all if one doesn't import already quite a lot; just re-phrase it really.
jfw: hmm ok, as an update?
diana_coman: jfw: just change it in there to make it clearer, not an issue.
jfw: http://logs.ossasepia.com/log/ossasepia/2019-12-05#1012039 - I recall discussion of using the hash of a genesis patch to identify the trees it creates, suggesting to me that the genesis should contain some unique context
ossabot: Logged on 2019-12-05 03:37:56 diana_coman: how do you reason anyway re similarity genesis-main/makefile? because "genesis" is whatever vpatch is the root of a tree but not really substantially different in any way from another vpatch otherwise.
diana_coman: I can't figure out which discussion you are referencing there, hm; but even as you state it there, that would be something done from outside *with* a patch considered "genesis"; basically any vpatch has a hash anyway, could equally well use that to identify the (sub-)trees that start from it, no?
jfw: also probable I don't quite grasp the overall vision. ATM what I see does look more like a forest of separate projects than one big tree
jfw: let me see if I can find it.
diana_coman: yes, it is; partly because parts are still missing (most glaringly moving files)
ossabot: (trilema) 2019-11-06 diana_coman: mircea_popescu: not as far as I know; it was always *not done* but always popping up in people's memory as "working"; except every time I wanted to *move* a file rather than have it del/add, it turned out that ...no.
jfw: the discussion I recalled is hereabouts
ossabot: (trilema) 2016-06-13 mircea_popescu: mod6 the part where each press should end up collected in the /genesis-hash/ dir is settled, i thought ?
diana_coman: jfw: the way I read that is that a v-press from a leaf that has several possible roots should end up in as many dirs as you have roots - ie as many presses effectively - with each dir named based on the hash of the corresponding root (admiteddly I didn't go back fully to see maybe there's more I don't recall on that older convo)
diana_coman: ie not literally "genesis-hash" but "root1Name-its-hash" of sorts.
jfw: I'm still finding the linked convo pretty confusing fwiw, a couple different aspects being discussed possibly.
jfw: I think the "several possible roots" scenario is ruled out by the manifest
diana_coman: yes, it is not very clear and moreover it's also a bit old by now so some things have changed, most notably - yes - the manifest part.
diana_coman: jfw: anyways, now you basically give people a lot of work to do to catch up with your code that's finally published; I suppose that balances at least the writing-pain of sorts but I still feel like poking you in the eye for keeping silent for so long.
jfw: More directly to the main.c/Makefile question, if patches are all existing in some unified tree (though I'm as yet unsure quite how that would work), top-level names like that could come in conflict and perhaps each 'project' needs its own subdir, as seen for example in trb and mp-wp
jfw winces, rubs eye
diana_coman: I think that and similars will get clarified only as the thing finally gets moving towards that yggdrasil; so yes, not clear yet and it's been stuck for ages too.
jfw: alright, small steps then. Think I should regrind that 'keksum' genesis to include name? Could just rename the file too but the name is also in the manifest.
diana_coman: please do, yes; it's anyway less pain now before anyone else signed it.
jfw: aite.
diana_coman: better indeed.
diana_coman: jfw: do you have your dates for the Dec/Jan holidays?
jfw: sadly not certain yet
jfw: (whaack: ^)
diana_coman: kind of weird the sort of uncertainties you choose to live with :P
jfw: I think I need to figure it out today really.
jfw: get it booked, rather.
diana_coman: sounds sensible indeed.
jfw: I hear the 'weird' too, heh.
diana_coman: :P
whaack: jfw: alright, I will likely go to Nicaragua and save my trip to Panama for late March / early April. If I do a Panama trip for this visa's border run, the most realistic dates are cerca January 15th.
diana_coman vaguely hopes whaack does border runs on some proper motorbike at least.
whaack: diana_coman: that sounds like my mode of travel i'd be using if i wanted to avoid the men in uniform with stamps
jfw: whaack: pretty sure both dorion_road and I will be back by the 15th fwiw.
diana_coman: whaack: otherwise why... run, yes.
BingoBoingo: whaack: If you fly anywhere, since you'll probably hub through panama that'll probably be an opportunity to take a couple days there too.
diana_coman: whaack: I suppose you can equally visit BingoBoingo while he is still in Uruguay.
BingoBoingo: For however long I'm still here...
jfw: hey, a bitcoin foundation server is still for sale iirc...
BingoBoingo: It is. Tengo un Dell PowerEdge 610 con 2x 2.67 Ghz Intel E5640 CPUs, 48 GB RAM, 2 fuentes de alimentación y 2 fuentes de alimentación adicionales.
whaack: BingoBoingo: well, I like the excuse to quickly plan a Panama+Uruguay trip. I'll consider this seriously the next few days.
BingoBoingo: Well, jfw will be able to offer some thoughts on Uruguay in the very near future.
dorion_road: http://logs.ossasepia.com/log/ossasepia/2019-12-05#1012094 << the week of the 16th is still good for visiting vt.
ossabot: Logged on 2019-12-05 15:02:59 jfw: I think I need to figure it out today really.
whaack: huzzah two days after arriving and my UPS is paying dividends. I am connected via backup internet running on power from my ups. Looks like Costa Rica is having an extended rainy season. Now to figure out how to make it stop beeping...
dorion_road: whaack nice, good for you. good luck repressing the beeps.
whaack: dorion_road: All I had to do was click any button on my APC. I read online though that some apc's don't come with an off button for the beeps and I got pretty worried.
asciilifeform: whaack: i used to have a series of those. desolder wick cures.
whaack: asciilifeform: I may have to do that if I can't figure out how to turn it off completely. I don't need to be woken up at night to be told my power is out.
BingoBoingo: whaack: Since it runs lead-acid batteries, there are situations where your health might benefit from some beeping.
asciilifeform: BingoBoingo: ~those~ ( overheat, short, etc ) tend to silent..
BingoBoingo: I was thinking more along the lines of pre-mature failing battery gas, but probably silent too. Should however smell.
whaack: BingoBoingo: Hm okay, I will take action if the battery starts to smell. Are there other best practices / safety tips I should know about for this brick?
asciilifeform: whaack: (unless yer a miser) -- 1x/2y battery swap.
BingoBoingo: whaack: Just if you smell sewer gas or the battery starts bulging remove either it or yourself from the residence as fast as you can. Like out the window fast. But yes, usually yearly battery changes will take the risk of this to almost zero.
asciilifeform: if miser -- 1x/3y. beyond that, box becomes 'decorative'
whaack: hm okay. I would not put a bomb in my house to save ~$150 yearly
BingoBoingo: 1 year is the sorta interval that should give you a wide margin of safety.