IRC logs for #openttd on OFTC at 2020-03-15
⏴ go to previous day
00:00:23 <andythenorth> they go in a list called 'later' :P
00:02:19 <TrueBrain> which you store in /dev/null, I assume? :D
00:05:08 <rotterdxm> Reality: Worst Game Ever. Get your FREE COVID DLC
00:10:50 <TrueBrain> and I fixed another TODO! :D
00:15:54 <TrueBrain> and 2 moreeeeeee :D
00:35:11 *** rotterdxm has joined #openttd
00:40:56 *** andythenorth has joined #openttd
00:47:55 <TrueBrain> and one more, done. Scenarios and heightmaps work too, w00p :D
01:02:14 <frosch123> yay, ugly template is ugly
01:03:42 <TrueBrain> There is no such thing :p
01:07:58 <DorpsGek_III_> [OpenTTD/OpenTTD] glx22 commented on pull request #8035: Fix: Non-applicable goal in global goals list is no longer clickable https://git.io/JviVR
01:46:05 <ac> any good resources to see nml examples? openttd/nml/examples is extremely limited, and first source is in pynml
01:47:05 <frosch123> not all are nml, but i think most
01:47:34 <glx> you can check the regression tests too (but they are mostly syntax examples)
01:48:51 <ac> frosch thanks. haha will take a look. right now my cargos are replacing the original ones for some reason, so probably an obv fix once i see someone else's
01:52:15 <ac> thanks. i was using 60 and 61 and still had the issue. nml5.0 and openttd1.9.3. perhaps i need to use <32
01:53:03 <ac> from the page: "It is recommended to set the 'number'-property to the same value as the item ID. " does this mean: "item(FEAT_CARGOS, item_cargos_charcoal) { property {
01:53:03 <ac> number: item_cargos_charcoal;...
01:54:09 <ac> so: item(FEAT_CARGOS, item_id) {
01:54:16 <ac> and then i set number: item_id; ?
01:54:24 <ac> chooses the first valid new number?
01:54:52 <glx> The first argument is the feature of the item, the second (optional) argument is the name used to identify the object throughout the NML file and the third (optional) argument is the numerical id you want to use for this item
01:56:04 <ac> aha that makes sense. so i can put item(FEAT_CARGOS, item_cargos_charcoal, 12) and use number: 12
01:56:46 <glx> some parts of the doc are not very friendly :)
01:58:11 <ac> the part of that where it is just <expression> <expression> <expression> is a bit odd haha
01:58:25 <ac> especially when the first is always the feature
01:58:48 <frosch123> if you do cargos/industires, then yeti is probably the easiest to read
01:58:52 <glx> yeah the author of this part must be a coder :)
02:01:00 <ac> gls: must be haha. fr: is there an important diff of pnmlvs. nml?
02:02:22 <frosch123> pnml is run through the c-preprocessor to procude a single nml
02:03:17 <glx> pnml are subsections of a nml file
02:03:19 <ac> so they are using #defines etc ?
02:03:35 <frosch123> yes, some use macros
02:03:53 <frosch123> andy switched to generating nml with python, that's pynml then
02:03:54 <ac> but mostly to modularize id guess
02:04:24 <frosch123> it's about project size :) small -> nml, medium -> pnml, huge -> pynml
02:05:05 <glx> don't dare looking at firs or iron horse :)
02:05:43 <ac> i was reading andrew's firs repo
02:06:04 <glx> even him has hard time to follow its code :)
02:08:36 <glx> anyway when you do a massive industry/vehicle set it's way easier to use one file per industry/vehicle
02:09:20 <ac> the docs have not given me a very clear idea of the lifetime of STORE_TEMP vs STORE_PERM variables or what shares the same registers. are STORE_PERM values stored for each industry and will last thru the whole game, whereas TEMP is shared by everything and only should be used if directly read again?
02:09:39 <ac> anyway when> i would like to start having multiple files already, with one industry and two cargos:)
02:11:22 <glx> temporary is valid during the current callback run
02:12:40 <ac> ok thats much more clear. when a callback sets run_again, the next run is considered the same ch ain?
02:15:17 <ac> setting the number:11 and 12 and putting a 3rd argument to item worked. got new cargos now:)
03:35:51 *** Wormnest_ has joined #openttd
03:57:22 <ac> haha is dev.openttdcoop.org hg server mega slow?
03:57:35 <ac> might be here all week trying to clone yeti
04:45:54 *** rotterdxm has joined #openttd
05:31:36 *** ChanServ sets mode: +v tokai
07:26:18 *** Progman has joined #openttd
08:12:54 *** andythenorth has joined #openttd
08:49:59 <andythenorth> is it coffee time?
09:45:35 <andythenorth> alternative history nonsense
09:46:19 *** sla_ro|master has joined #openttd
09:49:29 <nielsm> andythenorth if you do that you really should also have proper manufacturer names
09:49:38 <andythenorth> that is one of my questions
09:56:48 <andythenorth> do I write the story from the perspective of:
09:57:02 <andythenorth> 1. the train company you're running ('your chief engineer has designed x')
09:57:23 <andythenorth> 2. a vehicle manufacturer 'we designed this x'
09:58:08 <andythenorth> 3. wiki-style / history book / vehicle catalogue 'manufacturer foo designed x'
09:59:26 <nielsm> hm I think I prefer 2, manufacturers designing vehicles in response to technological developments and demand trends in the market
10:08:01 <andythenorth> nielsm: I am going to do this in web docs initially
10:08:36 <andythenorth> I'd like to have an 'extended info' window for vehicles in game though, just a simple window opened from buy menu / vehicle window
11:00:48 *** rotterdxm has joined #openttd
11:20:33 *** gelignite has joined #openttd
11:29:58 <DorpsGek_III_> [OpenTTD/OpenTTD] SamuXarick commented on pull request #8035: Fix: Non-applicable goal in global goals list is no longer clickable https://git.io/JviXj
11:34:24 <Samu> this new version of edge kind of sucks
11:34:45 <Samu> feels like it's not well integrated with windows
11:36:30 <andythenorth> stores to temp register, then reads it, stores it to permanent register
11:36:41 * andythenorth is lol programmer
11:38:18 <nielsm> I'm wondering if it would be possible to write callbacks in a more procedural manner, though not with NML
11:39:32 <Samu> looks like glx was right...
11:40:43 <Samu> s->company or this->window_number or this->owner, doesn't matter
11:41:58 <andythenorth> nielsm: well there are varact2 procedures...
11:42:03 <andythenorth> which are now supported in nml
11:42:12 <andythenorth> they reduce the weirdness
11:42:28 *** qwebirc77939 has joined #openttd
11:42:35 <andythenorth> writing everything as case->branch->[repeat]->return is weird
11:42:47 <Eddi|zuHause> andythenorth: the presence of procedures doesn't make anything procedural, though
11:43:42 * andythenorth reading wikipedia
11:45:08 <andythenorth> I am always hazy on the distinctions between different programming paradigms
11:46:41 <DorpsGek_III_> [OpenTTD/OpenTTD] SamuXarick commented on pull request #8035: Fix: Non-applicable goal in global goals list is no longer clickable https://git.io/Jvi1m
11:47:29 <andythenorth> what paradigm best describes newgrf varact 2?
11:47:36 <DorpsGek_III_> [OpenTTD/OpenTTD] SamuXarick updated pull request #8035: Fix: Non-applicable goal in global goals list is no longer clickable https://git.io/JvrFj
11:53:22 <nielsm> newgrf va2 syntactically looks like ML pattern matching, except it does range matching rather than patterns
11:54:49 <nielsm> maybe syntactically is the wrong word, structurally?
12:35:51 <DorpsGek_III_> [OpenTTD/OpenTTD] glx22 approved pull request #8035: Fix: Non-applicable goal in global goals list is no longer clickable https://git.io/JviMn
12:42:01 *** HerzogDeXtEr has joined #openttd
12:50:19 *** frosch123 has joined #openttd
14:09:30 <andythenorth> the official government sources tend to lack enough detail, which is causing certain dramatic blog posts to spread virally
14:30:11 <TrueBrain> right, so BaNaNaS .. what was I doing ...
14:30:57 <TrueBrain> search, github stuff and authentication
14:33:36 <michi_cc> nielsm: NML abstracts some of the most weird GRF stuff away (like the strange action 6 self-modifying grf parameter stuff), but other than that it is still a direct 1:1 low level structure.
14:34:32 <michi_cc> nielsm: A high-level language could represent it much simpler, but I'm not sure if trying to squeeze it into NML would make any sense.
14:35:25 <nielsm> yeah a higher level language would probably invite more complex code that would make everything even slower
14:51:45 *** tokai|noir has joined #openttd
14:51:45 *** ChanServ sets mode: +v tokai|noir
15:02:47 <spnda> I see on train action3 callbacks that under comment for the "speed" callback it says "Units are not (yet) available". SOo how would I use this then?
15:11:29 <frosch123> TrueBrain: what is your intended usecase for "search"?
15:11:32 <frosch123> i do not use it yet
15:11:58 <TrueBrain> to find dependencies
15:12:09 <TrueBrain> if you know a better way, I am open for it :)
15:12:26 <frosch123> ctrl+f on package list?
15:12:52 <TrueBrain> you download all the packages? As that are pretty long result-pages
15:13:01 <TrueBrain> I most likely even have to paginate them
15:13:09 <TrueBrain> (NewGRF is 14 MiB of JSON data)
15:13:44 <TrueBrain> I was considering removing "/package/{content_type}"
15:13:48 <TrueBrain> it is just too darn much data
15:14:34 <TrueBrain> what do you think would be the best approach?
15:17:05 <frosch123> i want a "all packages of content-type, but only with the newest version that is available for new savegames"
15:17:12 <frosch123> ok, maybe "search" covers that
15:17:20 <frosch123> then i do not need /package/content-type
15:17:38 <TrueBrain> I think I prefer that, as even your query is a very huge result :P
15:17:52 <TrueBrain> and for search it is I guess okay to only return the first N results
15:18:07 <TrueBrain> would that cover adding dependencies for you?
15:18:17 <TrueBrain> (don't really know what you are using /package/content-type for atm :D)
15:30:48 *** Progman has joined #openttd
15:34:38 <frosch123> TrueBrain: essentially i want the same list as ottd in-game
15:35:01 <TrueBrain> what are you trying to do with it? :)
15:35:41 <TrueBrain> for the manager part?
15:35:51 <TrueBrain> ah, you are building that too?
15:36:08 <frosch123> i link between the stuff
15:36:17 <frosch123> you can click on a package, and then see all versions
15:36:28 <TrueBrain> k, use the /package/content-type for that for now. I will have to figure something out .. not sure yet
15:36:46 <frosch123> maybe the api calls could have another parameter, which lists the wanted result items
15:36:52 <frosch123> then i can skip "description" etc
15:37:32 <TrueBrain> pretty sure we can remove more data from it .. possibly good indeed to minimize the /package/content-type endpoint
15:37:39 <frosch123> for adding deps, i am thinking: there are pages for every version, which lists some id that you can c&p into the upload form
15:37:40 <TrueBrain> but okay, build first, improve later
15:38:28 <TrueBrain> so that meansI don't have to implement search endpoint for now :P
15:45:48 <frosch123> user clicks frontpage -> package_list -> package_info -> version_info
15:46:12 <frosch123> these are the public pages, there are equivalent pages with entries instead of plain text, for editing
15:47:29 <TrueBrain> guess we really need to store custom licenses somewhere too :D
15:47:39 <TrueBrain> you can now only see them after downloading the binary :P
15:47:50 <TrueBrain> nice work frosch123 :) andy can do the styling I am sure :D
15:48:33 <frosch123> yeah, if the api could give me urls for license/readme/changelog/download, including download for savegame-only versions if authenticated as author
15:48:53 <TrueBrain> I am going to write that down on the wishlist :)
15:49:34 <TrueBrain> as I think those are good improvements
15:49:42 <TrueBrain> but atm: finish first, improve after :)
16:09:58 <TrueBrain> why did I make 2 types of "Content Type" .. 1 for the API, and one for on disk ..
16:10:02 <TrueBrain> ai-library vs "AI Library"
16:10:09 <TrueBrain> in an URL, the first looks so much better
16:10:16 <TrueBrain> hmm .. maybe I should just also change it on disk
16:10:31 *** stefino has joined #openttd
16:10:32 <TrueBrain> ugh, shitty job incoming
16:10:41 <stefino> Eddi|zuHause: hi Eddie are you here?
16:10:44 <TrueBrain> (just nasty to make such changes, but okay .. it is for the better)
16:10:50 <Eddi|zuHause> stefino: partially
16:11:18 <stefino> Eddi|zuHause: do you have experience with stations in NFO?
16:11:27 <Eddi|zuHause> not really, sorry
16:11:46 <stefino> Eddi|zuHause: np, thanks :)
16:11:56 <Eddi|zuHause> i tried looking into it once, but did not get very far
16:12:04 <stefino> hi all, has anyone experience with stations in NFO?
16:13:54 <stefino> I need help with "cargo aware function" for multitile stations and cargo aware function for 1x1 tile station but with random cargo graphics
16:14:52 <frosch123> then you need to look at action3
16:15:10 <frosch123> there is no much flexibility. you can do what action3 provided, but nothing more
16:15:52 <frosch123> that implies that the stationtype already decides what cargotype matters
16:16:00 <frosch123> you cannot dynamically change it or something
16:19:09 <frosch123> however, the problem is that cargo amount is 0 when a train is waiting, so iirc some station grfs just do not care about waiting cargo, and display a fixed sprite of cargo piles
16:22:10 <stefino> I have to confess that I only can use Quast's tutorials but to do something new is impossible for me :D my ideas are: to add passangers for island platform. So it has 2 tracks and it is 1 tile length. Classic platform is OK but here is a problem because I think it will be strange to have only 1 type of passangers.
16:23:07 <stefino> it would be good to have 4-5 types /groups of passangers and add it into random function to make it diverse
16:23:46 <Eddi|zuHause> shouldn't a RandomAction2 work for that?
16:24:54 <stefino> the last thing is similar like island platforms - station buildings so 1 track or non-track but 2-3-4 long
16:26:52 <stefino> whole ideas mean one cargo graphics for all cargo types.
16:29:52 <stefino> I saw some station sets where was one type of station and cargo graphics is different for coal, iore, ... but this is so complicated I think
16:29:59 *** Wormnest_ has joined #openttd
16:30:20 <Eddi|zuHause> i think you should look at how they did it
16:33:25 <stefino> I tried but decompiled grf was a little bit unreadable :D
16:41:08 <stefino> but still really crazy :D btw it there still ID limit for FF value? So I can have only 256 stations in one GRF?
16:44:26 <frosch123> there is a maxium of 256 station classes (all grfs combined), and 256 station types per class
16:44:38 <frosch123> don't spam the players, and it will be enough
16:45:41 <frosch123> and yes, also a limit of 255 stations per grf
16:45:46 <frosch123> but again, don't spam the player
16:46:25 <stefino> okay okay :) I'm looking on chips and it is really crazy
16:47:03 <stefino> I think I will be happy when I'll code basic stations
16:49:18 <TrueBrain> meh, seems I walked into an utf-8 issue ... for some reason I think I am not going to like this :)
16:52:46 <TrueBrain> "Korean train set\n\nMade by Las\n\n\uB77C\uC2A4\uB2D8\uC758 \uD55C\uAD6D \uC5F4\uCC28 \uC138\uD2B8 1.6\uBC84\uC804\uC785\uB2C8\uB2E4."
16:52:53 <TrueBrain> that ... is not what it should been outputting
16:54:03 *** argoneus has joined #openttd
16:56:34 <TrueBrain> fucking yaml.dump .. grr
16:58:07 <TrueBrain> 2048\\xD71024 <- that doesn't look like valid utf-8, does it?
16:59:08 <frosch123> no, if the highest bit is set, it must be set on multiple bytes in sequence
16:59:49 <TrueBrain> pyyaml is not the best library out there, to put it midly
17:00:00 <TrueBrain> I tell it to output utf-8 .. and that is the result :P
17:02:11 <TrueBrain> it needs allow_unicode=True
17:02:16 <TrueBrain> why .. is that not the default?!
17:03:00 <TrueBrain> I am currently reading the whole BaNaNaS database, and writing it back on disk again; there should be no diff ofc :)
17:03:25 <frosch123> sounds like there is :p
17:03:36 <TrueBrain> most surrounding description :)
17:03:54 <TrueBrain> people did some crazy shit in that field
17:07:04 <TrueBrain> someone managed to get \ts in there too :D Nice
17:08:30 <TrueBrain> "Kondor Zoltán<U+FEFF>"
17:08:34 <TrueBrain> pretty sure that is invalid UTF too
17:09:26 <frosch123> ZERO WIDTH NO-BREAK SPACE
17:09:34 <nielsm> yes also known as the byte order mark
17:10:19 <frosch123> then it would be at the beginning?
17:10:39 <TrueBrain> I do not really understand how this got loaded in the system, as it should fail on utf-8 decoding, but okay :P
17:11:19 <nielsm> U+FEFF is valid, technically anywhere in text
17:11:24 <TrueBrain> it fails to encode .. hmm .. so why does it ...
17:11:34 <nielsm> unless you mean the byte sequence FE FF which is not valid UTF8
17:11:55 <TrueBrain> Zoltán<feff> <- is what "vim" tells me
17:12:09 <TrueBrain> every editor highlights it as "THIS IS WEIRD" :P
17:12:54 <TrueBrain> so indeed, it is valid UTF .. so why is this a problem for pyyaml ..
17:13:22 <frosch123> hmm, there is something fishy. kcharselect does not list an utf-8 representation either, only utf-16
17:13:23 <nielsm> by convention text files in a unicode encoding are allowed to have FEFF as the first character, because the encoding of it can be used to detect the UTF format
17:13:32 <TrueBrain> sorry, I assumed what it presented me meant there was a bytecode of FEFF, but ther eis an UTF-8 character FEFF
17:13:58 <TrueBrain> well, this is in the middle of nowhere :P
17:14:36 <TrueBrain> okay, lets not fuzz too much over this, lets just remove the character :)
17:14:51 <nielsm> stripping it out should be permitted yes
17:15:17 <nielsm> well technically it can affect shaping of text in some complex scripts
17:15:20 <TrueBrain> I was just a bit surprised this was allowed in the system to start with. Owh well :)
17:15:36 <nielsm> prevent ligatures from happening between the two codepoints it separates
17:16:19 <nielsm> but there's a different codepoint that's recommended for that purpose, ZWNBSP is not supposed to be used anywhre but the first character in a file/text field
17:16:24 <TrueBrain> I would nearly ban you for being a spam bot SpComb :P
17:17:44 <TrueBrain> okay, reading and writing everything results in the same content .. w00p :) Found tons of other unrelated issues, but that can be fixed :)
17:21:14 *** Speedy` has joined #openttd
17:30:04 <TrueBrain> so ... we are on lockdown too, I guess .. schools closed, bars closed, everything closed till 6th of April
17:30:50 <TrueBrain> owh, we cannot call it lockdown, but yeah .. it kinda is :P
17:37:06 <frosch123> so, is lordaro stranded in rotterdam?
17:37:08 <TrueBrain> could have done without living it, but at least we can say we lived this part of history :P
17:38:50 <TrueBrain> okay, so content is now stored on disk as expected .. now I only have to make the commit and push it, I guess .. but that is for another day :D
17:38:55 <rotterdxm> hey, I just noticed something... the farm fields in my openTTD games have disappeared. this has happened somewhere in the past 2 months or so. no idea why. someone told me this was a FIRS thing, but I have been using the same version for a while now
17:39:27 <DorpsGek_III_> [OpenTTD/content-server] TrueBrain updated pull request #1: Add: first version of the content_server rewritten in Python https://git.io/Jvi4P
17:41:59 <TrueBrain> frosch123: if you like, I also pushed our super-secret-bananas-github-repository with the latest
17:42:09 <TrueBrain> with that checked out under BaNaNaS in content-api
17:42:28 <TrueBrain> not sure if your code already talks with an API, but that would allow you to test real-life data :)
17:42:43 <TrueBrain> don't look in the history of BaNaNaS, it is horrible :P
17:42:48 <frosch123> nah, i am pushing the api to the end, makes it easier
17:42:49 <TrueBrain> (I will squash it soon)
17:43:20 <TrueBrain> it should also work in terms of updates, uploads, etc
17:43:29 <TrueBrain> authentication is just "Bearer <username>"
17:44:03 <TrueBrain> I am going to remove the "search" endpoint for now, and indeed limit the output of /package/content-type
17:44:16 <TrueBrain> otherwise, it is nearing completion ... it is ~2000 lines of code ...
17:44:20 <TrueBrain> it is so much bigger than I expected :P
17:45:34 <TrueBrain> after fixing flake8 :P
18:28:24 *** andythenorth has joined #openttd
18:41:53 <frosch123> do you know where the eints vm lives these days?
18:42:18 <frosch123> i am tried of getting a mail every day that it is broken, so i want to shut it down
18:49:33 <planetmaker> oh, it should kinda still live where before, just different hardware
18:50:12 <planetmaker> I can just shutdown the VM, if you want
18:50:25 <planetmaker> so no eints / translations anymore for openttd?
18:50:26 <frosch123> yes, that's the easiest
18:50:29 <planetmaker> that'd be kinda sad
18:50:52 <frosch123> planetmaker: in a few weeks we have github-logins on openttd.org
18:51:44 <planetmaker> but... that's missing all the hints etc... no UI for it as now, yes?
18:52:20 <planetmaker> or is it a plugin for eints?
18:52:39 <frosch123> just you login with your github account
18:52:46 <frosch123> and translator roles are managed via github teams
18:52:55 <planetmaker> ah. so that then lives on github?
18:53:14 <frosch123> eints lives somewhere on a ottd server
18:53:18 <frosch123> it is not yet cloud ready
18:53:37 <frosch123> that's after bananas
18:53:51 <planetmaker> it's no problem to keep it on a VM on my machine
18:54:26 <frosch123> i want to use the dorpsgek infrastructure
18:55:01 <planetmaker> means it has to be on that machine or is it just having a VM or...?
18:55:28 * peter1138 fiddles with OpenTTD in vscode.
18:55:29 <frosch123> it has dorpsgeks private keys
18:56:35 <planetmaker> well. It wouldn't matter, there's two admins on this server, spike and me. And then whoever has access to the VMs
18:56:45 <frosch123> for now it will just feature openttd. next are basesets, then maybe andy stuff
18:56:58 <frosch123> that's probably everything that may be somewhat alive
18:57:48 <planetmaker> seems the coop server becomes completely obsolete, eh?
18:58:43 <frosch123> it has 344 repositories, which are all dead
18:59:11 <frosch123> it's not particulary exciting to work on something that noone uses
18:59:43 <frosch123> ah wrong, the finnish guys use it, so 2 alive
19:19:08 <glx> hmm maybe #7990 should have "backport requested" label
20:04:39 *** arikover has joined #openttd
20:10:25 <andythenorth> coop still has bundles?
20:12:56 <LordAro> frosch123: well the fery appears to be moving away from the continent
20:13:48 <frosch123> local trains are free now :)
20:13:55 <frosch123> but glad you made it :)
20:20:27 <andythenorth> except where it wasn't
20:21:12 <planetmaker> coop still has bundles, yes
20:23:01 <Wolf01> So, I think people here are allergic to rules... yesterday about 7000 in entire Italy charged for not respecting quarantine or simply the best practices to avoid the virus, yes 7000 in a single day. Only 18 here btw.
20:24:39 <andythenorth> planetmaker: don't shut the server yet then :)
20:24:45 <andythenorth> but bundles could be a set of s3 buckets tbh
20:25:15 <andythenorth> Wolf01 has the confirmed case growth rate changed?
20:25:26 <andythenorth> it's very weird how much news-but-no-news there is
20:28:47 <planetmaker> andythenorth, the server won't be shutdown. It's migrated to a new one... and no need to change a thing as-is
20:45:26 <Wolf01> <andythenorth> Wolf01 has the confirmed case growth rate changed? <- no, still exponential
20:46:30 <Wolf01> South Korea nailed it, we aren't able to do it
20:50:35 <frosch123> oh, it's so localized to the north?
20:50:44 <frosch123> (or was, 2 weeks ago)
20:52:07 <Wolf01> No, it's still localized in the north, but a lot of people fleed to south, we expect them to go extinct in 2 weeks, the south is not well covered with our healthcare system
20:52:43 <Wolf01> They have only some cases, and locked down about everything
20:55:15 <frosch123> well, when i checked germany's numbers on friday, they were about the same in all states
20:55:45 <frosch123> only a variation of factor 2 between different states wrt "cases per population"
20:56:09 <frosch123> and that's well within measurement accuarcy, since small states had very few cases
20:56:50 <Wolf01> North Italy is more like a big city, a lot of connections, a lot of people moving from all the north continuously, so it spread fast
21:00:30 <TrueBrain> in the netherlands they said: we seen 1000 cases, we estimate there are 6000 people carrying this virus
21:00:37 <TrueBrain> I was thinking: that sounds like a low number
21:00:54 <Wolf01> Yep, we have 24k cases, estimate is 120k
21:01:26 <TrueBrain> means 1 in 6 has sympthons severe enough to mention it
21:01:32 <TrueBrain> I guess they know, but it feels low
21:01:51 <frosch123> that other blog, that andy hates so much, used 800 cases per death, assuming that that is measured more accurately
21:02:40 <TrueBrain> but death strongly depends on the effectiveness of the healthcare system
21:03:23 <TrueBrain> well, in a few years: remember that spring of 2020?
21:03:32 <SpComb> Finland currently has the capacity to test 500-600 samples per day, so it's going to be severely underdiagnosed here
21:03:32 <frosch123> it's the combination of the spread rate over 2 weeks, and the death rate after 2 weeks when there is no overload
21:04:00 <TrueBrain> SpComb: in the netherlands they stopped testing most people
21:04:03 <frosch123> TrueBrain: some people say that next season will be worse :)
21:04:12 <TrueBrain> "do you think you will die? No? Just sit it out, you will be fine"
21:04:55 <SpComb> yeah, the current policy is to not even call unless you have severe symptoms
21:04:56 <TrueBrain> I like the reasoning: tested or not, you will be at home, sitting it out .. so knowing if you have it yes/no doesn't change anything
21:05:02 <frosch123> i have no idea what influenza does during its summer vacation
21:05:36 <SpComb> positive tests and official quarantine would have an impact on people's behavior
21:05:51 <TrueBrain> knowing for sure you have it, means you are even more careful
21:06:39 <frosch123> are you sure? maybe you take the initiative to visit the anti-vaccination club?
21:06:51 <TrueBrain> that we do anyway :P
21:07:13 <TrueBrain> I like the meme: I haven't heard anything about anti-vaxxers lately? - No, they are all dead!
21:07:27 <TrueBrain> morbid humour etc :P
21:07:48 <frosch123> TrueBrain: it's better :) there was a anti-vaxxer demo planned in munich next week. it got canceled
21:09:58 <SpComb> but ATM the hope is that people will take the initiative to stay at home with even mild flu symptoms
21:10:12 <frosch123> hmm, usually pm watches the same channels as i do, so weird that is news to him
21:10:47 <frosch123> SpComb: that's a weird hope when people go to work
21:11:03 <SpComb> someone sneezes in the cafeteria, and then everyone goes silent for a couple seconds, etc :)
21:12:52 <SpComb> allergy season is going to be really awkward
21:29:44 <frosch123> did you link that earlier today?
21:30:12 <andythenorth> it fills a gap between WHO etc who are publishing infrequently and without much explanation
21:30:21 <andythenorth> and the bloggers, who honestly know fuck all
21:30:37 <andythenorth> one has gone round the UK today explaining that herd immunity only follows mass public vaccination
21:30:43 <andythenorth> which is bollocks, but eh FB loves it
21:32:13 <andythenorth> if the number of people who are vectors for onward transmission falls sufficiently, then r0 diminishes to 1 or less
21:32:29 <andythenorth> the question is how do we lower r0, and how long can we hold it there
21:33:20 <frosch123> question is, how much is r0 affected by weather
21:36:48 <andythenorth> someone observed there are no southern hemisphere cases, but eh, who knows
21:37:16 <frosch123> australia is southern hemisphere
21:37:59 <andythenorth> I doubt that has ever happened in all history ^
21:41:46 <milek7> UK still doing nothing?
21:44:18 <andythenorth> perception is that yes
21:44:27 <andythenorth> relative to other European nations, yes
21:45:31 <andythenorth> social media suggests some people would like a lot more to be done
21:47:24 <andythenorth> but my social media bubble is very left wing so eh
22:08:56 *** arikover has joined #openttd
23:07:57 *** andythenorth has left #openttd
23:22:27 <DorpsGek_III_> [OpenTTD/OpenTTD] alistaircarscadden commented on issue #8034: Text in cheat menu window is excessively dramatic https://git.io/Jvrdw
23:28:23 <Samu> i think i found a bug with transfers
23:29:02 <Samu> i have a train doing unload and leave empty
23:29:40 <Samu> the station doesn't accept the cargo anymore, i get a transfer money, which prevents the train from getting into negative profit
23:30:40 <Samu> train remains in the positive profit, while in reality, it didn't get any real money. is that intended? cargodist isn't even on
23:32:01 <glx> that's how transfer works
23:32:47 <Samu> :( i can't detect when a vehicle isn't making money
23:34:42 <glx> basically never use unload on final destination
23:35:41 <glx> if the cargo is accepted it will be unloaded anyway
continue to next day ⏵