#ossasepia Logs for 03 Jan 2020



April 21st, 2020 by Diana Coman
lobbes: diana_coman: just letting you know I am back from holidays and at home terminal. Catching up and getting reorganized [00:13]
whaack: diana_coman: EOD Report: I did the act-first-think-second comp building this morning we discussed for 3hr. Then I sent 2hr on writing my article with the type of my disassembly notes. I did find a missing part, i do not have screws to mount the ssd and hdd (pcgamingcr did not send the case's assembly kit that contains them.) I ordered them for my mother to bring when she visits this Sunday. For TheFleet I setup basic join/disconnec [01:42]
whaack: t logging. I looked into logging join/part and it should be something I can get done quickly tomorrow. I have not forgot about the comment and took some time to draft a response. [01:42]
diana_coman: lobbes: wb, hope you had a refreshing break. [05:28]
diana_coman: whaack: hopefully you get the test at least started so it can have a run over the next 2 weeks when it seems you'll not have much time on this. [05:28]
diana_coman: dorion_road: how's the cruise? [05:34]
whaack: diana_coman: Yes I am trying to get some bots connected by tonight / tomorrow midday. [11:55]
whaack: in scams: when I dropped the "no manuals? then the parts are not new?" message, i got no reply from pcgamingcr until the next day (this morning) "Todo es nuevo si no no dariamos garantia en los artículos" to which i responded "Cuales partes son nuevos y cuáles tienen garantia?" [11:58]
diana_coman: lolz. [12:00]
whaack: i think some of the parts actually are new, and they decided to loot the accessories. because for example with the case they gave me the manual, just not the screws that come with it [12:00]
diana_coman: quite possibly or for that matter even "lost/dropped as unimportant the accessories" wouldn't surprise me much; but at the end of the day, this sort of details really don't matter, as the mess is still the mess, there's not much point to differentiating now between flavours of mess based on...intentions or something. [12:04]
diana_coman: whaack: did they actually give you any papers for that "garantia"? and even if papers - what exactly are they worth? will they replace the part if it's faulty or what? (I can imagine how that would go,lolz) [12:05]
whaack: diana_coman: right. i don't care much about figuring out the intentions, i want to play my cards so that i just recuperate as much of my loss as possible. that's why i have not gone full PUTA MADRE ME ENGAÑÓ. [12:07]
whaack: diana_coman: no i think the garantia is an empty statement but i could ask for some papers for that i guess [12:07]
diana_coman: sure, you should ask for the papers anyway. [12:11]
adlai: this channel is now unsubscribed from cat facts; whaack may send a private message to request a link to the unsolicited advice [12:41]
whaack: diana_coman: I have two categories of messages I need to log. One is a standard message to channel, like the one i'm sending right now. The other type of message I need to log is a special message from fleetbot giving some information, there are currently 4 special messages: a msg when fleetbot connects/disconnects from a channel, and a join/part message. Currently I am using the same table "log" for both types of messages. The ben [13:04]
whaack: efit of this design is you can do a simple select query on the log ordered by received date and see all the information about a channel you'd typically want in order. The bad part of this design is that the 'special messages' are distinguished from normal messages via ~ 'magic values' [13:04]
diana_coman: whaack: what do you mean by magic values? [13:45]
diana_coman: it's a different category and you need categories anyway so supposedly there's already a column for that, isn't there? [13:46]
whaack: diana_coman: I don't have a category column, would be a good addition. as for magic values - there are columns like "host" that I set to "127.0.0.1" for the "special messages" [13:49]
whaack: er "magic values" [13:49]
diana_coman: whaack: neah, add the category column; you need it *anyway*, see http://ossasepia.com/2020/04/21/ossasepia-logs-for-31-Dec-2019#1014347 [13:51]
ossabot: Logged on 2019-12-31 11:45:30 diana_coman: I'd rather it did; ideally marked as a different category (so you can filter them out/in later, at analysis time) [13:51]
diana_coman: and moreover, it IS the sane approach there, what; no 127.0.0.1, why would you even think of that? [13:52]
diana_coman: whaack: in general: do NOT blindly overload existing parts with whatever new meaning you happen to need on the spot, it's unsanitary, quite disgusting and the start of one full set of clusterfucks. [13:54]
whaack: diana_coman: As for not adding the column category immediately - I was being stupid for a second. However there still needs to be values put into fields (such as host) that don't have meaning for special messages, even if those values are just null / emptry string. [13:56]
whaack: My nose was at least working as I could smell the unsanitary code [13:57]
diana_coman: whaack: the DB has default values, use them; this being said, the fact that you find yourself in the situation where you need to fill in fields that have no meaning for some rows is essentially a sign that you are stuffing in there what doesn't quite belong; whether it's a problem or not, it depends. [14:04]
diana_coman: whaack: strictly speaking, you are indeed abusing the log table there since the fleetbot "messages" are not irc messages, are they? [14:05]
diana_coman: so yeah, that "have to fill this which has no meaning" is precisely the flag of your abuse. [14:06]
diana_coman: whaack: also, the ordering is usually saner by id of message rather than the timestamp which can easily run in all sorts of trouble; is there an id per line? [14:07]
whaack: diana_coman: yes correct the fleetbot messages are not irc messages. [14:08]
diana_coman: (in turn, this can further complicate the trouble when you start inserting in there non-message messages); like all design, db design will end up quite a mess if you "just do it". [14:08]
whaack: diana_coman: each row has a unique id [14:09]
diana_coman: whaack: basically your fleetbot messages belong in a table of their own and you can always do a join, dbms is good at that; have it log chan id or whatever for link + timestamp or something. [14:09]
diana_coman: whaack: so myeah, not very helpful to then mess up the sequence of message ids by inserting those non-messages in there. [14:09]
diana_coman: whaack: I don't recall how much xp you have with sql/similar/db. [14:10]
whaack: diana_coman: ah well the unique ids are random hexes, i will see to it that they have incrementing ids [14:10]
diana_coman: uhm, why random hexes? [14:11]
whaack: diana_coman: that is the design of logbot that I have not changed [14:11]
diana_coman: ah, huh; I can't see a reason for that; trinque , why did you use random hexes as unique ids for each log line in db? [14:12]
diana_coman: whaack: just add an auto-incrementing id to the table and the dbms will do the rest for you. [14:13]
whaack: diana_coman: ack [14:14]
diana_coman: I somehow rather doubt there's any danger of overflowing the data type there + if it's at that, there's potential trouble in theory with random hexes too via clash. [14:15]
diana_coman: will bbl [14:15]
lobbes: http://logs.ericbenevides.com/log/ossasepia/2020-01-03#1014489 << ty, it was very refreshing! Good to be back though. [18:37]
ericbot: Logged on 2020-01-03 09:22:37 diana_coman: lobbes: wb, hope you had a refreshing break. [18:37]
lobbes: http://logs.ericbenevides.com/log/ossasepia/2019-12-24#1013903 << anything specific by Popper you recommend I start with? I started chewing through Kant a bit last year, but didn't get too far (was indeed a bit over my head). [18:37]
ericbot: Logged on 2019-12-24 17:13:33 diana_coman: ha, forgot lobbes ^ [18:37]
diana_coman: lobbes: he's very readable and quite enjoyable really in ~all I read; the trouble with starting points is that they make sense or not depending on what you already have (or not); otoh, one starts wherever one can anyway, so there isn't any "wrong" starting point either. [18:42]
lobbes: diana_coman: I think I gotcha. So I kinda have to start chewing through ~something~ to really get the rest anyways [18:44]
diana_coman: fwiw (and iirc) I think I read first his essays (grouped I think in "in search of a better world") - kind of made me want to listen to what he had to say; the open society and its enemies draws upon a lot of references so on one hand it can be overwhelming if you have no idea of those references, on the other hand you could perhaps use it as a starting point to learn about those too. [18:44]
diana_coman: lobbes: yes re chewing and also keep in mind that first read is just that - the *first* read, not like you have to attempt the impossible task of "getting everything from this one read" [18:48]
diana_coman: lobbes: sure, you could in principle go at it historically ie start with the ancients and work your way as they did; it makes for a fun project but I wouldn't call it a very practical approach tbh. [18:50]
lobbes: this is true re: first reads. Looking back, anything I found worth reading I ended up reading more than once usually [18:51]
lobbes: lol, yeah starting with the ancients sounds fun but also I think I would perish from this earth before making it too far [18:52]
diana_coman: that is 1; then reading is re-reading is 2; and usually one needs at least 3 passes to fully get something so that's 3 and we'll stop there as it's quite enough anyway. [18:53]
diana_coman: lobbes: heh, if you also add the requirement to read them in original, certainly; and that's precisely how one gets both very happy ("will never run out of stuff to do!!!") and totally useless (will *also* never get to *do* anything either) [18:54]
diana_coman: lobbes: did you enjoy Kant though? [18:55]
lobbes: diana_coman: I did enjoy (what little I grokked) of Kant. I need to get an english version though, it was an extra layer of confusion trying to see if I was getting the German translated correctly [18:58]
lobbes: It was an enjoyable exercise, but then I think that rolls into your 'never be bored, but never get to do anything' point just above [18:59]
diana_coman: well, let me translate for you and the logs, this tidbit that might shed some light on my take on philosophy overall so what I say on it gets a bit of context too: I was once before sworn at by a philosopher as he said I was a philosopher myself. [19:02]
diana_coman: lobbes: but re Kant, I asked because on reflection, initially I hadn't included you in the recommendation for Popper mainly because I thought you were quite likely to enjoy Kant possibly even more. [19:03]
diana_coman: basically I spent quite a lot of years considering all philosophers idiots :D [19:05]
lobbes: haha. For what reasons though? Too much opining about and not enough practical bits? [19:06]
diana_coman: lobbes: well, I was very young indeed (about 15-20 pretty much); looking back at it, I'd say that philosopher guy I mention in the comment had seen something ie it was a combination of "hah, I'll make my OWN if I need one, thank you very much" + they'll talk of everything and do nothing + some idiots as "teachers" of the subject in school (+ possibly more tbh). [19:11]
diana_coman: will bbl [19:13]
lobbes: that combination does resonate with me in a familiar way, I can say, esp. at that age. [19:18]
lobbes: will also bbl [19:18]
trinque: diana_coman: UUIDs are cheaply globally unique for tiny datasets like human-produced logs [23:04]
trinque: I didn't want to be in the business of worrying about renumbering log lines if I merged logs from two sources, and had previously said all manner of things about that log-line by foreign-key reference [23:05]

Comments feed: RSS 2.0

Leave a Reply