| diana_coman: | whaack: sounds all right; try jfw's keksum, good idea + you should read that too anyway; bot investigation sounds ok so continue from there and do take notes so you have them at write-up time. | [03:40] |
| whaack: | diana_coman: I can try and use keksum this weekend and then try to digest the source next week. Before reading jfw's implementation I assume I should read the keccak reference, which looks like a sizeable task by itself. | [10:19] |
| diana_coman: | whaack: oh huh, you don't quite know what Keccak is? | [10:56] |
| diana_coman: | but yes, you will have to read and digest Keccak, ofc, otherwise uhm, what sense exactly can you possibly make of any implementation of it, whether it's jfw's or mine or anyone else's really. | [10:58] |
| whaack: | diana_coman: Yes, Keccak is a black box for me. "s/I assume I should/I need to." | [11:03] |
| diana_coman: | whaack: look perhaps through the eucrypt series, as I think I go in some detail over the Keccak mechanics as such too; iirc jfw also links some tutorials from his article introducing the genesis of his keksum. | [11:07] |
| dorion_road: | enjoyed the lobbes / diana_coman conversation, congrats to both! | [13:19] |
| dorion_road: | http://ossasepia.com/2020/04/21/ossasepia-logs-for-12-Dec-2019#1012741 << is this a mandatory on premise saltmine ? or do you have room there to negotiate a contractor basis wrapped in a corporation, say deleware llc, and deliver the excel jockey work and administer the corporation from abroad ? | [13:20] |
| ossabot: | Logged on 2019-12-12 16:50:27 lobbes: I stayed in that line of work for about 6 years until I got my current gig as an "excel jocky" (which I've been doing for about 3 yrs) | [13:20] |
| dorion_road: | http://ossasepia.com/2020/04/21/ossasepia-logs-for-12-Dec-2019#1012755 << how much have you looked into selling the house and renting ? that'd at least provide you more liquidity. | [13:22] |
| ossabot: | Logged on 2019-12-12 17:04:09 lobbes: diana_coman: well I do want to sell my house, and I do want to travel. However, I realized that even if I hop the fence I'm still faced with the "wut do now?" question, and the problem of my being poor (both in capital and skill) | [13:22] |
| dorion_road: | http://ossasepia.com/2020/04/21/ossasepia-logs-for-12-Dec-2019#1012811 << glad to hear it. do you have any questions/comments about | [13:25] |
| ossabot: | Logged on 2019-12-12 17:56:12 lobbes: diana_coman: The tmsr os work interests me because I think it could help me improve my overall computer literacy. Plus I think the dorion/jfw combo has some chance of bringing in money (the republic is indeed lacking profit centers). I would like to help out in that aim in some capacity | [13:25] |
| dorion_road: | the business plan ? | [13:25] |
| jfw: | hi all, my net connection's been down about an hour for reasons unknown. Mobile tethering right now and unsure if that'll be up to the task of finishing this photo-heavy article. For now I think I'll focus on investigating the net problem. | [14:19] |
| BingoBoingo: | jfw: As much as I don't like the additional suspense, it is a joy seeing these come out in pieces. Prolongs the time I get to see where I live through other people's eyes and lenses | [14:22] |
| jfw: | whaack: I'm curious at least from an academic standpoint whether keksum will build on Mac. We'll see if it's true what I heard that they killed static linking even more than Drepper did. Simple Makefile change if so. | [14:23] |
| jfw: | io.c and main.c won't require Keccak knowledge so there's all sorts of parallel approaches here. | [14:24] |
| jfw: | I certainly found diana_coman's articles helpful to whet the appetite for the full reference; I linked the first in the series. | [14:24] |
| jfw: | Digesting Keccak was pretty fun too I thought, rubik's bit cubes and sponges and all. | [14:26] |
| jfw: | BingoBoingo: I doubt I can do justice to all the sights at least without spending another week on it but will try! | [14:31] |
| BingoBoingo: | jfw: Well, kinda how these things work. Eventually you'll tire of the men on horse statues as you did when they were in front of you | [14:32] |
| whaack: | diana_coman: Change of plans. My keyboard is going to arrive in Santa Cruz (the nearby town I always refer to) tomorrow, so it would be a waste of time for me to go today and only be able to pick up the bank card. I am going to move the trip to town to Tuesday. I'm going to use the extra time today to keep working on studying / setting up trinque's bot. | [14:42] |
| whaack: | jfw: I'm running OSX 10.9 and I have some problems in general with installing various items, since my 'brew' (the os's package manager) is broken. So whether or not I can get keksum setup probably says little as to whether your 'average' mac user will be able to. I will report my success/failure regardless. | [14:45] |
| jfw: | 'average mac user' doesn't even have compiler or bash these days; long as you have those (which can be got right from Apple) you should be fine. | [14:47] |
| diana_coman: | http://ossasepia.com/2020/04/21/ossasepia-logs-for-13-Dec-2019#1012881 - the unexpected blows to all plans; did you sort it out meanwhile? | [15:29] |
| ossabot: | Logged on 2019-12-13 14:19:39 jfw: hi all, my net connection's been down about an hour for reasons unknown. Mobile tethering right now and unsure if that'll be up to the task of finishing this photo-heavy article. For now I think I'll focus on investigating the net problem. | [15:29] |
| diana_coman: | http://ossasepia.com/2020/04/21/ossasepia-logs-for-13-Dec-2019#1012885 - glad to hear it! | [15:30] |
| ossabot: | Logged on 2019-12-13 14:24:56 jfw: I certainly found diana_coman's articles helpful to whet the appetite for the full reference; I linked the first in the series. | [15:30] |
| diana_coman: | http://ossasepia.com/2020/04/21/ossasepia-logs-for-13-Dec-2019#1012889 - at least that makes for a better reason to spend those 5-6 hours on town-visit, certainly | [15:30] |
| ossabot: | Logged on 2019-12-13 14:42:03 whaack: diana_coman: Change of plans. My keyboard is going to arrive in Santa Cruz (the nearby town I always refer to) tomorrow, so it would be a waste of time for me to go today and only be able to pick up the bank card. I am going to move the trip to town to Tuesday. I'm going to use the extra time today to keep working on studying / setting up trinque's bot. | [15:30] |
| diana_coman: | whaack: is that town so far away or is it that poorly connected or why does it take anyway the best part of a day ? | [15:31] |
| diana_coman: | dorion_road: where are you with the plan & changes this week? | [15:36] |
| dorion_road: | diana_coman I'm planning to cover it in my review to be published later today, is that okay ? | [15:56] |
| diana_coman: | dorion_road: ok. | [15:57] |
| jfw: | diana_coman: I determined the outage is not just me, cable provider reported ~200 endpoints down in the neighborhood and has people working on it but no ETA. I've been focusing on offline work (wallet dev) for now, no shortage of that heh. | [15:59] |
| diana_coman: | jfw: heh, sounds even good for your wallet work which was highest priority anyway iirc. | [16:00] |
| jfw: | yep. Looks like I'm on track for deadline of online / node part today, minus the sendrawtransaction TRB patch. Had some technical debt to pay - schema and code refactoring, but through that and back to the top-level commands. | [16:04] |
| jfw: | dorion_road: ^ | [16:05] |
| diana_coman: | not bad. | [16:05] |
| jfw: | The change I made based on MP's comments was to add address tags (many-to-many) so you can define arbitrary address sets and query them independently. Thus one node can serve many wallets or have decoy sets | [16:11] |
| whaack: | diana_coman: It is a ~1hr drive each way. The roads are 1 lane though and so problems can add 30mins+ to this. The bank takes about 30min-1hr depending on the line. Then I go grocery shopping which takes about another 40min. I also run around picking up miscellaneous items on every trip which takes another 30mins-1hr. | [16:24] |
| whaack: | jfw: Upon running make I get http://paste.deedbot.org/?id=PZLN . I ran gcc --version and got http://paste.deedbot.org/?id=33tu I learned that some version of Xcode swept gcc away and replaced it with Clang. So do you know if it is possible to compile keksum with Clang? If not I will try to get gcc installed, but it looks like I will run into the same headache I had with gnat. | [17:19] |
| jfw: | whaack: yes they switched to llvm to stick it to RMS or something; lots of effort went into making it as gcc-compatible as possible donw to the CLI options. | [17:21] |
| jfw: | Your problem there looks like the static linking thing. | [17:21] |
| jfw: | ("ld: library not found for -lcrt0.o" for the log.) | [17:22] |
| whaack: | jfw: ok so then do you know a "Simple Makefile change" I should try? Sorry if this should be self evident. | [17:24] |
| jfw: | see where it says -static in the makefile? ;) | [17:24] |
| whaack: | ahaha yeah i changed that to -dynamic | [17:24] |
| jfw: | nah just remove it. | [17:25] |
| whaack: | cool, works | [17:25] |
| jfw: | nice, runs too? | [17:25] |
| whaack: | jfw: yes, wonderful. now I can finally have a modern V on this comp. | [17:26] |
| jfw: | oh and I should mention, I updated my keksum and yrc articles with new genesis for patch renaming and changing the capacity default, last night. | [17:27] |
| jfw: | might want to make sure you have the latest, in case you want to build on it. | [17:28] |
| jfw: | whaack: does your box also have python 2? if so you can run the tests via 'make check' | [17:29] |
| whaack: | jfw: I get various errors. http://paste.deedbot.org/?id=yJdp | [17:30] |
| whaack: | (the first line is missing the first c character, copy paste error) | [17:31] |
| jfw: | gah, one test broken anyway by the default capacity change, but otherwise working for me, let me check on those errors. | [17:32] |
| jfw: | that TEST_STEPS should be defined in testvectors.h which is generated by the python script. Does it look sane? | [17:33] |
| whaack: | ah testvectors.h is completely empty. maybe the python script failed silently | [17:34] |
| jfw: | did you miss copying more 'make check' output before that 'c'? | [17:35] |
| jfw: | I don't see it running at all. | [17:35] |
| whaack: | jfw: No the c was the only missing thing from the output of 'make check' (the full output here: http://paste.deedbot.org/?id=ucSE ) | [17:38] |
| jfw: | try again after 'make clean' please | [17:39] |
| jfw: | that empty testvectors.h is preventing it from re-running so we don't see why it's run wrong | [17:40] |
| whaack: | jfw: I ran "python munge-testvecs.py testvectorszero.txt testvectorsnonzero.txt > testvectors.h" manually and then "make check" and got back http://paste.deedbot.org/?id=pPCn | [17:41] |
| whaack: | jfw: Now I will make clean and try again. | [17:41] |
| jfw: | interesting, you get different hash than me for the 'smoke test'. Wonder if 'yes' and 'dd' work different on mac, lol | [17:44] |
| jfw: | meanwhile you could "./keksum *" and see if hashes match what's in the vpatch. | [17:45] |
| whaack: | jfw: I found my problem. I had to switch "python2" to "python" in the Makefile. Then make clean make check gave me the same error as above ^ http://paste.deedbot.org/?id=Hc40 | [17:45] |
| jfw: | aha, always something dumb like that. python2.7 might work too, I thought major-version-only was supposed to 'always exist' but whatevs. | [17:46] |
| jfw: | plain 'python' will sometimes turn out to be python3 though, /me glares at Gentoo | [17:47] |
| whaack: | jfw: soon to be python4 no doubt | [17:47] |
| jfw: | not in my vtree :) | [17:48] |
| jfw: | any of those hashed sources pass a manual check against the vpatch at least to the first copule digits? | [17:54] |
| whaack: | jfw: yes one source file passed a full check (using grep) | [17:55] |
| whaack: | jfw: i've updated my v to use keccak and now i'm going to try to properly press trinque's bot source without getting dumped a bunch of errors for unmatched hashes | [17:56] |
| whaack: | to use keksum* | [17:56] |
| jfw: | hokay good, it's probably just that hacky pipeline to generate test data that comes out different on mac. Think I'll remove that in favor of one of the sources. | [17:56] |
| jfw: | thanks for the testing! | [17:57] |
| whaack: | jfw: thanks for the tool! | [17:57] |
| jfw: | glad it's of use. | [17:58] |
| jfw: | arguably there's a 'user error' of make on my part here too; if you create files by redirection, make won't know that the command failed because an empty file is created. | [18:01] |
| whaack: | jfw: hm a small detail I noticed about keksum. the output contains 1 space between the hash and the filename, whereas my shasum on my mac and sha512sum on ubuntu contains 2 spaces between them. Afaik there's nothing wrong with your choice, and i don't know why those other hash functions put 2 spaces instead of 1, but I have to tweak my v.pl a little more to make it be compatible with this. | [18:47] |
| jfw: | the reason those utils use 2 spaces is the second is a 'format' char that could be * for systems that distinguish text from binary. I threw that out because such systems = Windows and there's no such distinction worth making imho. Not sure what phf's "ksum" does here, that might be the precedent worth following if any. | [18:51] |
| whaack: | jfw: okay cool. | [18:55] |
| jfw: | "worth making" here I mean strictly in the sense of "a file is a series of octets and the system has no business 'helpfully' mangling newlines or whatever", obviously text vs. binary is important for human habitation. | [18:55] |
| lobbes: | http://ossasepia.com/2020/04/21/ossasepia-logs-for-12-Dec-2019#1012821 << btw I grok this point now in that there is a big difference between ~knowing~ what to do and actually doing it. ty for the context | [19:15] |
| ossabot: | Logged on 2019-12-12 18:04:56 diana_coman: lobbes: here, to help you re that rational point; note the apparently in there too. | [19:15] |
| lobbes: | http://ossasepia.com/2020/04/21/ossasepia-logs-for-13-Dec-2019#1012873 << ty for the congrats! | [19:15] |
| ossabot: | Logged on 2019-12-13 13:19:38 dorion_road: enjoyed the lobbes / diana_coman conversation, congrats to both! | [19:15] |
| lobbes: | http://ossasepia.com/2020/04/21/ossasepia-logs-for-13-Dec-2019#1012874 << unfortunately no real room to negotiate; I'm just a wage slave for a large bank. Though they do have jobs elsewhere in the world (Hong Kong, for example), so there is a potential path there. | [19:15] |
| ossabot: | Logged on 2019-12-13 13:20:28 dorion_road: http://ossasepia.com/2020/04/21/ossasepia-logs-for-12-Dec-2019#1012741 << is this a mandatory on premise saltmine ? or do you have room there to negotiate a contractor basis wrapped in a corporation, say deleware llc, and deliver the excel jockey work and administer the corporation from abroad ? | [19:15] |
| lobbes: | In addition, I know some other folks that I used to work with at my current job that were able to snag 100\% remote work doing similar things, so that is another path in the back of my mind. | [19:15] |
| lobbes: | http://ossasepia.com/2020/04/21/ossasepia-logs-for-13-Dec-2019#1012876 << this is definitely my ultimate goal (I certainly don't want to sit on this house forever). I figure I sharpen my mind a bit first while getting the house in a more sellable shape (I got to clear out from it the useless shit I've accumulated over the years, and do some minor repairs, etc.). If the Charlotte bubble doesn't pop in the meanwhile I ought to come out the other side with some | [19:15] |
| ossabot: | Logged on 2019-12-13 13:22:29 dorion_road: http://ossasepia.com/2020/04/21/ossasepia-logs-for-12-Dec-2019#1012755 << how much have you looked into selling the house and renting ? that'd at least provide you more liquidity. | [19:15] |
| lobbes: | quidity. | [19:15] |
| lobbes: | http://ossasepia.com/2020/04/21/ossasepia-logs-for-13-Dec-2019#1012878 << tbh you both have real-world experience in running a business, whereas I do not. So I'm mostly just following along and learning from both the two of you and others that have offered advice/perspective. | [19:16] |
| ossabot: | Logged on 2019-12-13 13:25:28 dorion_road: http://ossasepia.com/2020/04/21/ossasepia-logs-for-12-Dec-2019#1012811 << glad to hear it. do you have any questions/comments about | [19:16] |
| ericbot: | (trilema) 2019-12-11 mp_en_viaje: "soo... well done on the 50mn exit. now, why are you at walmart with a credit card, in the same outfit as the single mother over there with her govt-issued scrip ? you know the same social worker can cut your access that gives her the electronic meal ticket, yes ?" | [19:16] |
| lobbes: | As such, I have no real questions/comments as of yet (but will definitely chime in if I do) | [19:16] |
| diana_coman: | http://ossasepia.com/2020/04/21/ossasepia-logs-for-13-Dec-2019#1012908 - ugh, a sort of commute; I guess it's the price of living on the beach otherwise. | [19:23] |
| ossabot: | Logged on 2019-12-13 16:24:02 whaack: diana_coman: It is a ~1hr drive each way. The roads are 1 lane though and so problems can add 30mins+ to this. The bank takes about 30min-1hr depending on the line. Then I go grocery shopping which takes about another 40min. I also run around picking up miscellaneous items on every trip which takes another 30mins-1hr. | [19:23] |
| diana_coman: | interesting stuff out of testing jfw's keksum on mac at any rate; fwiw I tested it on centos 6 and didn't have any trouble at all (tests were fine too). | [19:24] |
| diana_coman: | http://ossasepia.com/2020/04/21/ossasepia-logs-for-13-Dec-2019#1012961 - this or even the Hong Kong version would still be an improvement over current situation really. | [19:25] |
| ossabot: | Logged on 2019-12-13 19:15:43 lobbes: In addition, I know some other folks that I used to work with at my current job that were able to snag 100\% remote work doing similar things, so that is another path in the back of my mind. | [19:25] |
| diana_coman: | lobbes: what does it take to switch to remote work? | [19:25] |
| diana_coman: | RubenSomsen: since you had some interest in labels | [19:33] |
| ossabot: | Logged on 2019-12-12 18:16:08 diana_coman: well, the label is either a burden or a recognition, depending on where you actually are. | [19:33] |
| lobbes: | diana_coman: to switch to full remote would require switching to another employer, the Hong Kong jobs could be done via current employer | [21:10] |
| lobbes: | as for what it would take beyond that, I'd need to do some further research | [21:11] |
Comments feed: RSS 2.0