--- Log opened Fri Nov 16 00:00:36 2018
20181116 01:09:54-!- hk238 [~kvirc@unaffiliated/hk238] has quit [Quit: KVIrc 5.0.0 Aria http://www.kvirc.net/]
20181116 01:58:42-!- celmin|away is now known as celticminstrel
20181116 03:19:46-!- mattsc [~mattsc@wesnoth/developer/mattsc] has joined #wesnoth-umc-dev
20181116 03:30:18-!- sevu [~sevu@p54855FAA.dip0.t-ipconnect.de] has quit [Remote host closed the connection]
20181116 04:45:19-!- celticminstrel is now known as celmin|sleep
20181116 04:50:56-!- mattsc [~mattsc@wesnoth/developer/mattsc] has quit [Quit: So long and thanks for all the fish.]
20181116 09:00:08-!- vn971 [~vasya@office.zivver.org] has joined #wesnoth-umc-dev
20181116 11:05:38-!- hk238 [~kvirc@unaffiliated/hk238] has joined #wesnoth-umc-dev
20181116 11:05:40< hk238> hi
20181116 11:05:53< hk238> if I want to include default into a custom era, how would I go about doing that?
20181116 11:06:02< hk238> I mean is there a shortcut for that?
20181116 11:08:47< vn971> hk238: don't think so. At least if you mean units.
20181116 11:57:09<+wesdiscordbot> <josteph> look at the {ERA_DEFAULT} macro
20181116 11:59:57-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-umc-dev
20181116 12:12:54< hk238> okay thanks
20181116 12:13:02< hk238> I was doing that but did it incorrectly apparently
20181116 12:14:14< hk238> ahh this isn't working but I don't know why
20181116 12:18:25< hk238> when I try to select the era in the era drop down menu, it doesn't get selected but instead default era gets selected
20181116 12:18:36< hk238> Any idea what might cause that?
20181116 12:19:14< vn971> hk238: same ID or name?
20181116 12:19:42< vn971> // or lack of any of those fields?
20181116 12:19:44< hk238> hmm I don't think so
20181116 12:20:27< vn971> I remember having this for maps, but in the end it's always an ID that has been forgotten.
20181116 12:20:53< vn971> or accidentally left untouched, equal to another map id.
20181116 12:22:01< hk238> okay that has to be something along the lines I guess... tried disabling everything in the era tags so it must be the unit files have some error
20181116 12:22:47< hk238> Can units have the same name as long as they have different id?
20181116 12:22:52< hk238> coz I just kept the names for the units
20181116 12:25:45<+wesdiscordbot> <josteph> yes units can have the same name if they have a different id
20181116 12:25:51< vn971> hk238: IDK, personally. I think there's some bug for that regarding built-in help (F1). Ravana_ might know on that one.
20181116 12:26:12< vn971> but generally yeah, same names **should** be allowed.
20181116 12:26:13<+wesdiscordbot> <josteph> but in general you should tell us what error message you get, or pastebin your code etc
20181116 12:26:26<+wesdiscordbot> <josteph> same name is definitely allowed, mainline campaigns do that
20181116 12:26:46< vn971> @josteph: modulo the bug I described (and Ravana found) though...
20181116 12:26:56<+wesdiscordbot> <josteph> link ?
20181116 12:29:21< hk238> I don't get any errors 
20181116 12:29:35< vn971> @josteph: https://github.com/wesnoth/wesnoth/issues/1774
20181116 12:29:48< hk238> maybe the problem is that I'm trying to use the default images without providing them as part of the era?
20181116 12:29:59< vn971> hmm, the issue is closed tho.
20181116 12:30:35< vn971> maybe it's solved by now then. I may be remembering outdated stuff. Though I think It actually applied to releases made a few months as well... Could be another issue thin.
20181116 12:30:37< vn971> then.
20181116 12:30:48<+wesdiscordbot> <josteph> vn971, that's for weapon specials
20181116 12:30:52<+wesdiscordbot> <josteph> not units
20181116 12:31:07<+wesdiscordbot> <josteph> see also https://github.com/wesnoth/wesnoth/issues/3355
20181116 12:31:21< vn971> something else indeed then.
20181116 12:32:08< vn971> anyway, gotta back to work. /me having an office job now.
20181116 12:33:45< hk238> I'll make a repository on github (if I manage) so it's easier to show what I got here
20181116 12:36:51< hk238> agh 
20181116 12:38:39< hk238> if I already have a folder how do connect a repository to the folder? like agh 
20181116 12:39:07< vn971> go to the directory
20181116 12:39:10< vn971> git init
20181116 12:39:17< vn971> this will initialise a repository
20181116 12:39:35< vn971> then use:  git remote add origin addres://copied-from-github
20181116 12:39:52< vn971> then you can try `git push`
20181116 12:40:31< hk238> okay I tried that though i'm using this software called gitkraken, but I did try the 'init' but now it's an empty repository and not linke to the folder
20181116 12:40:33< vn971> it'll complain because you haven't defined the binding to this remote yet, but it'll say you the correct command for your case.
20181116 12:40:55< vn971> hk238: nice. Now add the remote with a command like I wrote above
20181116 12:41:05< vn971> hk238: and don't forget to make your very first commit
20181116 12:41:29< vn971> committing should be easy. You can go do it straight away even, before adding a remote repo.
20181116 12:44:12< hk238> agh
20181116 12:45:00< hk238> it seems the problem with this gitkraken is that it can't initialize a repository from a folder
20181116 12:45:14< hk238> and also github webinterface is terrible so I'm having trouble
20181116 12:45:16< vn971> hk238: just open a terminal and type in `git init` ?
20181116 12:45:25< vn971> in the right folder
20181116 12:45:35< hk238> yeah but I'm trying to use this gitkraken so I dont have to use the command line
20181116 12:45:43< hk238> but okay I gues I could do that
20181116 12:45:52< hk238> except now I already have 2 folders which dont work so I need to delete them
20181116 12:47:36< hk238> I'm amazed that this well made software is incapable of doing something so basic as this :D
20181116 12:48:22<+wesdiscordbot> <jyrkive> It's not that Git would be "incapable" of doing it. It's just that Git's UX design is terrible.
20181116 12:48:32< hk238> I mean this gitkraken 
20181116 12:48:43< hk238> you can initialize a repository, you can open a repository, you can clone a repository
20181116 12:49:03< hk238> but if you have a folder which files already, then you cant do anything you need to initialize an empty repository, then copy the files into it manually
20181116 12:49:21< hk238> but okay let's do that then but now I need to rename the folder I actually the files in
20181116 12:49:26< hk238> so that i cna use that as the name for the init
20181116 12:49:37<+wesdiscordbot> <jyrkive> GitKraken probably can also initialize a repository from a directory, or clone a repo into an existing directory. Most likely you just don't know how.
20181116 12:50:45< hk238> yeah it has an init function but it doesnt do it that way it say sthat it has to be an empty folder
20181116 12:51:07< hk238> also it automatically makes the path into the name of the repository
20181116 12:51:24< hk238> You're probably right that it's probably possible
20181116 12:51:30< hk238> but this isn't done in a logical way
20181116 12:52:23<+wesdiscordbot> <jyrkive> ...so, even worse UX design than in Git itself. "Hmm, I'm sure that no one will ever need to create a Git repository out of an existing directory. Let's make sure that no one can do it! 😈 "
20181116 12:55:52< hk238> I'll have some coffee now. :D
20181116 13:00:58< hk238> omg
20181116 13:01:26< hk238> it's even worse I couldn't do the rename trick while the software was open
20181116 13:03:05< hk238> https://github.com/n832sv/dunefolk_modified
20181116 13:03:06< hk238> :D
20181116 13:03:16< hk238> so I renamed the folder
20181116 13:04:09< hk238> initialized an empty folder, which was automatically linked to a folder with the same name as the repository, while the software was open I couldnt delete the folder and replace it with another one, because then software would fail and lose the repository or create a new folder.. But now I just copied the files fro mthe renamed folder
20181116 13:04:19< hk238> it's just kind of frustrating that something so trivial can cause a problem 
20181116 13:05:09< hk238> in the meanwhile I still haven't understood why this doesn't work
20181116 13:07:31< hk238> I also included the server.pbl file apparently
20181116 13:07:42< hk238> but whatever 
20181116 13:07:44< hk238> :D
20181116 13:09:25< hk238> oh
20181116 13:09:40< hk238> can the #textdomain line cause a problem somehow? I forgot to modify those 
20181116 13:10:12<+wesdiscordbot> <jyrkive> If your server.pbl has ever been public (like it is now), you need to change your add-on passphrase.
20181116 13:10:26<+wesdiscordbot> <jyrkive> Otherwise anyone can send an "update" to your add-on.
20181116 13:10:58<+wesdiscordbot> <jyrkive> And make sure that you don't commit the changed server.pbl (remove it from the repository and add the filename to .gitignore)
20181116 13:11:31< hk238> Yeah I do understand that luckily at least I use a different password for all the things I publish 
20181116 13:11:43< hk238> and this is just a quick overhaulf of the dunefolk faction I did in an hour or two 
20181116 13:11:53< hk238> but yeah you're of course right 
20181116 13:12:00< hk238> how do you change the password though?
20181116 13:12:26< hk238> oh well, I haven't actually uploaded it yet so  I guess I can just.. well change it
20181116 13:12:34< vn971> hk238: I know there-s a CLI way with wesnoth_addon_manager.
20181116 13:12:40< vn971> not sure how to do it via GUI though
20181116 13:12:47<+wesdiscordbot> <jyrkive> Yeah, if it's not yet uploaded, you can just change the password locally.
20181116 13:12:50< vn971> hk238: yup:)
20181116 13:13:09< vn971> don't forget to add the file to .gitignore AND remove it from repository though.
20181116 13:13:16<+wesdiscordbot> <jyrkive> By the way, I recommend using a password manager ( https://en.wikipedia.org/wiki/Password_manager ) and fully randomly generated passwords (like I do).
20181116 13:13:32< hk238> I don't even understand what .gitignore means
20181116 13:13:43<+wesdiscordbot> <jyrkive> It's a special filename.
20181116 13:13:46< hk238> I like to use passwords that are phrases, depending on the context they're easier or harder
20181116 13:13:52< hk238> it's a good way of doing passwords 
20181116 13:14:07< hk238> sort of easy to remember, but hard to crack
20181116 13:14:15<+wesdiscordbot> <jyrkive> No, it's not. That kind of passwords are fairly easy to crack with brute force.
20181116 13:14:36< hk238> actually as far as I know they're not :D
20181116 13:14:46< hk238> simply because the length of the phrase 
20181116 13:14:53<+wesdiscordbot> <jyrkive> With a password manager, I don't even need to remember my passwords. My PC remembers them for me.
20181116 13:14:59< hk238> so if you have like an 8 letter password 
20181116 13:15:05< hk238> with symbols and what not 
20181116 13:15:08< vn971> jyrkive: <offtopic> personally, I prefer 3-4 random words. About as safe this way, and more easy to remember. Like explained in the XKCD thingy.
20181116 13:15:12< hk238> it's much easier to crack than a phrase that's 24 letters
20181116 13:15:37< hk238> plus you can always add one random symbol in there if you want to avoid dictionary based bruteforce search, although I guess that can still work
20181116 13:16:11<+wesdiscordbot> <jyrkive> Attackers don't do it in such a stupid way as to try every possible character combination. Automated password cracking tools pick words from a dictionary and stick them together. (They also try common substitutions such as replacing letters with numbers, so that won't help you either.)
20181116 13:16:23< vn971> hk238: create .gitignore file and put this inside:
20181116 13:16:23< vn971> *.pbl
20181116 13:16:23< vn971> .build
20181116 13:16:23< vn971> target/
20181116 13:16:24<+wesdiscordbot> <jyrkive> As a general rule, anything you can remember is also easy to crack.
20181116 13:16:37< vn971> or maybe just the first line.)
20181116 13:16:58< hk238> yeah I guess that's partially true, but thankfully you can use finnish phrases 
20181116 13:17:04< hk238> since you got the inflection 
20181116 13:17:23<+wesdiscordbot> <jyrkive> Here's a good article about the subject: https://arstechnica.com/information-technology/2013/05/how-crackers-make-minced-meat-out-of-your-passwords/
20181116 13:17:34< vn971> @jyrkive: it's all about the entropy. English words have a surprisingly high number on this. I think it's about 64k on the relatively popular ones.
20181116 13:18:00<+wesdiscordbot> <jyrkive> That's assuming you really pick truly random words.
20181116 13:18:30<+wesdiscordbot> <jyrkive> In practice, people who use this method choose from a couple thousand common words (and almost exclusively nouns at that).
20181116 13:18:33< hk238> yeah it's not that phrase based passwords are impossible to crack, but it's just not easy depending on details of course
20181116 13:18:51< hk238> you can argue athat in somecases using a word is equivalent to using a single letter
20181116 13:18:52<+wesdiscordbot> <jyrkive> Case in point is "dunefolkmodifiedpasswordisverylong". "Dune" is the only uncommon word here.
20181116 13:19:05< hk238> although a letter which happens have several thousand possible symbols
20181116 13:19:13< hk238> so if you string to gether 5 words 
20181116 13:19:20< hk238> even if you dont modify the words a lot 
20181116 13:19:30< hk238> if the pool your drawing the words from is like 60.000 words
20181116 13:19:35< vn971> jyrkive: so you basically chose between log(64000)^word_count and log(40)^letters_count
20181116 13:19:37< hk238> you still got a lot of combinations to go with 
20181116 13:19:43< hk238> meanwhile if you use 8 letters that are random symbols
20181116 13:19:50< hk238> you're only drawing them from a very small pool
20181116 13:20:09< vn971> @jyrkive: if you use password words generator, then it's indeed random.
20181116 13:20:13< hk238> and if you mix in even a single random symbol that breaks a word, it's just impossible to guess that
20181116 13:20:18<+wesdiscordbot> <jyrkive> vn971: realistically, it's more like log(1000)^word_count.
20181116 13:20:24< vn971> well... No less random then `cat /dev/urandom`, anyway.
20181116 13:20:27< hk238> I mean I agree with you I'm just also interested in the matheamtic side of this
20181116 13:20:41< vn971> @jyrkive: again, depends on your source of words
20181116 13:20:49< hk238> I guess there's common words you can prioritize which can speed up the bruteforce search
20181116 13:21:11<+wesdiscordbot> <jyrkive> The truth is that human beings are terrible at choosing random words (or randomness in general).
20181116 13:21:47< hk238> yeah I'm pretty sure my main passwords are difficult to crack, but then lower priority passwords are easier
20181116 13:21:53< vn971> @jyrkive: and yet again, depends on your source of words.)))
20181116 13:22:06< hk238> finnish language is good here 
20181116 13:22:13< hk238> even though finnish has a lot less words than english, 
20181116 13:22:23< hk238> english has like 300.000 although most people have a vocabulary probably more like 60.000 words
20181116 13:22:31< hk238> and finnish has maybe 100.000 words
20181116 13:22:43< hk238> the fact that if you make a phrase out of them introduces inflections
20181116 13:22:49< vn971> hk238: unless you tell everyone you use finnish words, and the attacker targets you specifically )
20181116 13:22:53< hk238> and that just increases the effective number of words 
20181116 13:23:14< hk238> right I forgot.. Actually I'm swedish passwords with some german mixed in. Never finnish. Forget everything
20181116 13:23:14< hk238> :D
20181116 13:23:20< vn971> :D
20181116 13:23:51<+wesdiscordbot> <jyrkive> Seriously... piling on hacks with the phrase method is just a band-aid, and makes the passwords harder to remember. Fully random passwords together with a password manager are the only real solution.
20181116 13:24:47< hk238> the phrase method is pretty good because it increases the length of the string, of course depending on the phrase. If it's "mypasswordisverysecret" then it might not be agood phrase
20181116 13:24:58< hk238> but if it's "mygrandmotherlikeschocolatecakeswithvanillapuddingontop"
20181116 13:25:07< hk238> then it's not the kind of thing you typically have
20181116 13:25:20<+wesdiscordbot> <jyrkive> That's still not good. Automatic cracking tools are smart enough to try sentences.
20181116 13:25:25< vn971> @jyrkive: the counter-argument is that humans are good with remembering words, much better than at remembering random alphanums. This theory suggests that it's easier to get the same entropy using words than random alphanums.
20181116 13:25:32< hk238> yeah except they're not actuallly going to be able to crack that
20181116 13:25:40< vn971> I mean, requires less effort to remember and get the same entropy
20181116 13:25:56<+wesdiscordbot> <jyrkive> vn971: That's the point... you're not supposed to use passwords you can remember.
20181116 13:26:08< hk238> right like that phrase there has 9 words
20181116 13:26:11<+wesdiscordbot> <jyrkive> With a password manager, you don't need to remember.
20181116 13:26:45< vn971> @jyrkive: the password manager could be a weak point then. A **very** weak point. I don't believe you have a fully offline phone that you carry around.
20181116 13:26:55< hk238> if you assume that's drawing a from a pool of 30.000 common words, there's still 30.000^9 number of combinations ?
20181116 13:26:57 * vn971 wonders how soon we'll get to U2F.
20181116 13:27:20<+wesdiscordbot> <jyrkive> Your example is drawing from like a couple thousand words.
20181116 13:27:24< vn971> hk238: for 9 words, yes
20181116 13:27:32< hk238> if you have regular alphabet you have 27 letters or whatever
20181116 13:27:56< hk238> I think it's more than a couple of thousand
20181116 13:28:02<+wesdiscordbot> <jyrkive> vn971: how exactly is the password manager a "weak point"? An attacker would need to get my password database AND its master password.
20181116 13:28:06< hk238> but you might be right that 30.000 is a too high of an estimate
20181116 13:28:16< vn971> hk238: exponent grows really fast though, so the trade-off is not that obvious. Still, I switched to words some years ago and I believe it's more secure presuming you have to remember your passwords.
20181116 13:28:17<+wesdiscordbot> <jyrkive> It's hard to imagine that they'd get even one of them, let alone both.
20181116 13:28:35< hk238> but even then you can see that a string of 9 random letters is inferior 
20181116 13:28:58< vn971> @jyrkive: or inject a virus. Or get physical access. Depending on your threat model. Anyway, really gotta get back to work soon..
20181116 13:29:17< vn971> *physical access + video-recording master password.
20181116 13:30:39<+wesdiscordbot> <jyrkive> I am not aware of even one virus that targets password managers (ones with master passwords, to be precise).
20181116 13:31:01<+wesdiscordbot> <jyrkive> Whereas cracking password databases of websites is extremely common.
20181116 13:32:23< vn971> @jyrkive of course I never proposed to use same passwords anywhere ;-)
20181116 13:32:44<+wesdiscordbot> <jyrkive> Yeah, but that makes it even harder to remember all your passwords.
20181116 13:33:02<+wesdiscordbot> <jyrkive> I have 206 passwords in Password Safe.
20181116 13:33:20<+wesdiscordbot> <jyrkive> It would be hopeless for a human being to attempt to remember that much random data.
20181116 13:34:03<+wesdiscordbot> <jyrkive> At 9 words per password, we'd be asking for you to remember well over 1800 random words.
20181116 13:36:12-!- celmin|sleep is now known as celmin|away
20181116 13:36:31< hk238> yeah except they don't have to be random, since you can make a phrase that makes sense, such as the one above, which is easier to remember. But I agree if you need to remember 200 passwords then it sounds like way too much
20181116 13:39:05< hk238> if you assume a 1000 word pool, a 3 word password is as good as a password which is just 9 numbers
20181116 13:39:28< hk238> but 3 words are easier to remember than 9 numbers
20181116 13:39:42< hk238> and you can use larger word pool
20181116 13:39:56< hk238> and you can also introduce some symbol 
20181116 13:40:23<+wesdiscordbot> <jyrkive> Your example above used some extremely common words such as "my", "likes" and "with".
20181116 13:40:24< hk238> in the end, finnish phrases that are somewhat personal, are really good passwords
20181116 13:40:38<+wesdiscordbot> <jyrkive> Each of those words doesn't add anywhere near 10 bits of entropy.
20181116 13:40:38< hk238> yeah that's true 
20181116 13:41:04< hk238> sure they do 
20181116 13:41:21< hk238> coz even words that are that common, there's a lot of them too
20181116 13:42:22< hk238> but you're right in that a well made bruteforce dictionary search can do a weighted search in which it prioritizes words based on how common they're 
20181116 13:43:36< celmin|away> Which means that if you're using uncommon words (or even easy-to-remember yet made-up words) it's harder to guess.
20181116 13:43:46<+wesdiscordbot> <jyrkive> And also based on order of the words. Password attacks are likely to try e.g. "two adjectives followed by noun" or "[name] is [adjective]".
20181116 13:43:52< celmin|away> Maybe try mixing languages too for extra points.
20181116 13:44:05< hk238> yeah
20181116 13:44:38< hk238> the point I have is basically just that even though there are ways to guess at words, if you make a non-obvious phrase, it's really hard to guess, because the length of the phrase increases 
20181116 13:45:01< hk238> so like
20181116 13:45:22< hk238> a phrase that's 25 letters long might be the equivalent of a truly random string of 15 letters
20181116 13:45:42< hk238> but one of them is much easier to remember 
20181116 13:45:46<+wesdiscordbot> <jyrkive> My point: your intuition draws you towards common sentences and terms, and they have WAY less entropy than pure "three words out of a pool of thousand = 30 bits of entropy!" thinking suggests.
20181116 13:46:39<+wesdiscordbot> <jyrkive> Let's go back to "dunefolkmodifiedpasswordisverylong", for example. Two nouns, adjective, noun (a very common one in passwords), is, and two adjectives.
20181116 13:46:52< hk238> yeah but that's not a good password
20181116 13:47:01< hk238> I use those for the pbl files 
20181116 13:47:18< hk238> like my bios password for an example is pretty hard to crack
20181116 13:47:39< hk238> my linux password is also a bit harder, although I can't say it's great
20181116 13:48:04< hk238> and then these random things where I sign up somewhere online, they're not very good 
20181116 13:48:35<+wesdiscordbot> <jyrkive> BIOS and Linux passwords are less important, they can't really be brute-forced.
20181116 13:48:46<+wesdiscordbot> <jyrkive> And the online passwords are the ones which should be better.
20181116 13:48:54<+wesdiscordbot> <jyrkive> Your priorities are reversed.
20181116 13:49:15< hk238> yeah except I do think that if someone actually manages to get the linux administrator password that would be much more significant
20181116 13:49:55<+wesdiscordbot> <jyrkive> Plus physical access to your PC, or at least arbitrary shell code execution...
20181116 13:50:01< hk238> besides I did tell you that I have this mental health thing related to the russians, and I thought they hacked my smartphone somehow on root level 
20181116 13:50:19<+wesdiscordbot> <jyrkive> Without either of them, knowing your Linux password isn't really useful.
20181116 13:50:44< hk238> yeah I'm not an expert on software security most definitely, I can barely use linux 
20181116 13:50:59< hk238> but I just appreciate the mathematical point of passphrases
20181116 13:51:27< hk238> that even though you can do dictionary based searches, phrases especially in finnish, are very efficient passwords considering how easy they're to remember vs how hard they're to crack
20181116 13:51:31<+wesdiscordbot> <jyrkive> The "mathematical point" only works in theory, as far as I'm concerned.
20181116 13:51:51< hk238> yeah but someone has a 10 mixed symbol string it's not a good password because it's short
20181116 13:51:57<+wesdiscordbot> <jyrkive> "The specific type of hybrid attack that cracked that password is known as a combinator attack. It combines each word in a dictionary with every other word in the dictionary. Because these attacks are capable of generating a huge number of guesses—the square of the number of words in the dict—crackers often work with smaller word lists or simply terminate a run in progress once things start slowing down. Other
20181116 13:51:57<+wesdiscordbot> times, they combine words from one big dictionary with words from a smaller one. Steube was able to crack "momof3g8kids" because he had "momof3g" in his 111 million dict and "8kids" in a smaller dict."
20181116 13:52:05< hk238> if you've a phrase that's long, it's better because it's harder
20181116 13:52:26< hk238> right 
20181116 13:52:32< hk238> but the problem with that password is that it's short
20181116 13:53:20<+wesdiscordbot> <jyrkive> It's five words. "mom", "of", "three", "great" and "kids" - together with letters-to-numbers substitutions.
20181116 13:54:22< hk238> I don't even know what momof3g is ?
20181116 13:54:47<+wesdiscordbot> <jyrkive> "three" -> 3, and "g8" -> "great"
20181116 13:55:45< hk238> Well I find it a bit surprising that this phrase would be so common as to be included in such a dictionary
20181116 13:56:22< hk238> but note that it wasn't cracked because 5 words is bad but because this whole thing "momof3g" was included as a single word in the dictionary
20181116 13:57:34<+wesdiscordbot> <jyrkive> The article doesn't specify how it ended up in the dictionary. Most likely it had been found in a plain-text password leak earlier (and possibly shortened from a longer password).
20181116 13:58:44<+wesdiscordbot> <jyrkive> It's true that if you go for less common phrases than "mom of N kids", you're going into the range of acceptable entropy... although just barely.
20181116 13:59:05< hk238> or to put it inanother way that person has a dictionary with 111 million words
20181116 13:59:15< hk238> even if you pick words from that dictionary 
20181116 13:59:23< hk238> it's not going to be easy to crack as long as you have several words
20181116 13:59:50<+wesdiscordbot> <jyrkive> Still, password managers are the obvious solution. They not only store your passwords regardless of how much entropy you have, but also generate passwords which truly have as much entropy as they're supposed.
20181116 14:00:01< hk238> yeah but
20181116 14:00:19<+wesdiscordbot> <jyrkive> Whoever came up with "momof3g8kids" definitely believed that her password had more entropy than it actually did.
20181116 14:00:19< hk238> sure okay I'm not goign to argue against using password managers
20181116 14:00:25< hk238> because I dont know they work
20181116 14:00:33< hk238> I'm surprised by that too
20181116 14:01:13< hk238> i'm pretty sure that the grandmotherchocolatecakes password mentioned earlier though would take years for that bruteforcealgorithm to crack
20181116 14:01:23< hk238> if they kept it running on a home pc around the clock
20181116 14:02:53<+wesdiscordbot> <jyrkive> I'd say that the grandmotherchocolatecakes password has only five words with halfway reasonable entropy.
20181116 14:03:08< hk238> yeah and it's too much 
20181116 14:03:27<+wesdiscordbot> <jyrkive> "momof3g8kids" has three words with somewhat acceptable entropy. That one was cracked in something like 10 minutes.
20181116 14:04:09< hk238> yeah I'm surprised by that, but I think it's because it's a relatively common phrase, that women who are mothers use that as their password
20181116 14:04:10<+wesdiscordbot> <jyrkive> Each additional word multiplies the time by about a thousand, I think. So we're talking 10 million minutes, or 116 days.
20181116 14:04:16< hk238> so even it's personal, it's something that's very common 
20181116 14:04:42<+wesdiscordbot> <jyrkive> 116 days is easily long enough to be safe. That much I can agree with.
20181116 14:05:19<+wesdiscordbot> <jyrkive> Still... asking human beings to remember a couple hundred passwords like that is ridiculous.
20181116 14:05:22< hk238> yeah but if we're talking finnish phrases with 1 random symbol mixed in
20181116 14:05:36< hk238> then it's just not going to happen
20181116 14:05:55< hk238> because 
20181116 14:06:14<+wesdiscordbot> <jyrkive> Eh, I'd say that Finnish phrases are safe only because there aren't that many of us.
20181116 14:06:44< hk238> no but it's also because they increase the effective word count
20181116 14:06:59< hk238> although I guess if you would use some natural language search based on finnish
20181116 14:07:06< hk238> then you could get the inflections automatically 
20181116 14:07:19< hk238> I mean like if it was able to do them
20181116 14:07:29<+wesdiscordbot> <jyrkive> Finnish nouns have 15 grammatical cases, which in theory adds four bits of entropy per noun. But I bet that in practice the words are almost always in fairly predictable cases. I think declesion adds at best something like one bit per word.
20181116 14:08:11< hk238> yeah if you've a software that first of all can do the inflections correctly 
20181116 14:08:22< hk238> then I guess it doesn't add much 
20181116 14:08:35< hk238> but at least it's not possible to use a simple dictionary search then
20181116 14:08:38<+wesdiscordbot> <jyrkive> The "protection" we have at the moment is that no one is using Finnish natural language processing for password cracking, and thus each grammatical case is being treated like a separate word.
20181116 14:09:12< hk238> yeah 
20181116 14:11:50< hk238> hmm how long do you think it would take to crack the password
20181116 14:12:15< hk238> carcatdogpasswordmousetrain
20181116 14:12:18< hk238> :D
20181116 14:12:57<+wesdiscordbot> <jyrkive> This one is significantly better, actually. A string of six nouns is something that attackers are unlikely to try.
20181116 14:13:09<+wesdiscordbot> <jyrkive> ...of course, it's also difficult to remember.
20181116 14:13:37< hk238> yeah although  'cat' 'dog' 'password' 'mouse' are all included in the 'default password words' category, since a lot of people make passwords out of them
20181116 14:14:02< hk238> and 'car' and 'train' are extremely common words most people who barely know english understand
20181116 14:14:26< hk238> but if you assume the size of the super common words table is like 100 words and those 4 fall in them
20181116 14:14:37< hk238> an then you have like barely english vocabulary, size 1000 
20181116 14:15:14< hk238> you still have 100 000 000 000 000 combinations
20181116 14:15:24< hk238> I think?
20181116 14:15:24< hk238> :D
20181116 14:15:44< hk238> and that's still bruteforceable I guess, compared to grandmother password
20181116 14:16:30<+wesdiscordbot> <jyrkive> 1,010^14? That's cute. The passwords Password Safe generates are 12 random characters, with approximately 3,210^22 combinations. 😛
20181116 14:16:52< hk238> yeah 
20181116 14:17:13< hk238> 12 random characters?
20181116 14:17:19< hk238> that's pretty weak
20181116 14:17:31< hk238> I mean the grandmother password is much stronger compared to that
20181116 14:17:41<+wesdiscordbot> <jyrkive> More than eight orders of magnitude better than your example above.
20181116 14:18:01< hk238> yeah but it was a mockexample made of bad words 
20181116 14:18:02<+wesdiscordbot> <jyrkive> And your grandmother example is much weaker than pure mathematical calculations would suggest.
20181116 14:18:22< hk238> well it has 9 words so let's say let's give them all a number
20181116 14:18:54< hk238> mygrandmotherlikeschocolatecakeswithvanillapuddingontop, and the number represents the size of the vocabulary that contains it
20181116 14:19:22<+wesdiscordbot> <jyrkive> In the grandmother example, I consider that it has five words from a pool of 1000, and four from a pool of 32.
20181116 14:19:42<+wesdiscordbot> <jyrkive> Calculation gives me 1,0*10^21 combinations.
20181116 14:19:52< hk238> pool of 32? where did you get that?
20181116 14:21:05<+wesdiscordbot> <jyrkive> 1,0*10^21 is acceptable... but there's still the obvious problem that it requires a lot of effort to remember (whereas I'm not spending any effort to remember my passwords, except the master password and the ones I need to use without access to the password manager).
20181116 14:21:43<+wesdiscordbot> <jyrkive> I came up with 32 because it has 5 bits. Half of 10 bits, which is 1024 possibilities.
20181116 14:21:51< hk238> there's more than 32 prepositions in english language
20181116 14:22:04< hk238> so you can't argue that "top" and "on" and "my" come from a pool of 32
20181116 14:22:15< hk238> althoug they're no doubt very common
20181116 14:22:18<+wesdiscordbot> <jyrkive> Yes, but some of them are much more common than others.
20181116 14:22:32<+wesdiscordbot> <josteph> Those words are correlated.
20181116 14:22:41<+wesdiscordbot> <jyrkive> I think I can't really give more than five bits per word for those.
20181116 14:22:50<+wesdiscordbot> <josteph> Nobody's going to say "with vanilla pudding under" in a sentence about cakes
20181116 14:23:18<+wesdiscordbot> <josteph> It's not a bunch of random words, it's a sentence.
20181116 14:23:34< hk238> yeah it's true but this isn't possible with mere dictionary search either
20181116 14:23:46< hk238> you'd need a 'guess the next word' neural network
20181116 14:23:58<+wesdiscordbot> <josteph> You'd need something for sentences, yeah
20181116 14:24:06<+wesdiscordbot> <jyrkive> Such neural networks already exist, I think.
20181116 14:24:11< hk238> yeah it's possible
20181116 14:24:44< hk238> but are we using dictionary based cracking, or neural network based cracking, or combining dictionary and neural networks?
20181116 14:24:56< hk238> and how many results from the network you're going to include?
20181116 14:25:03<+wesdiscordbot> <jyrkive> It doesn't really matter what's used today.
20181116 14:25:06< hk238> coz you're not going to have the network generate the phrase
20181116 14:25:10<+wesdiscordbot> <josteph> how smart is the guy trying to crack your password ?
20181116 14:25:14< hk238> and you're not goign to crack it with dictionary
20181116 14:25:30<+wesdiscordbot> <josteph> you're not going to open a beer bottle with a philips screwdriver, either
20181116 14:25:32<+wesdiscordbot> <jyrkive> If phrase passwords become common, you can expect tools to crack them to be developed.
20181116 14:25:35<+wesdiscordbot> <josteph> so what? 😃
20181116 14:25:40< hk238> okay I admit
20181116 14:25:48< hk238> "ontop" is something that is more like a single word
20181116 14:25:59< hk238> they're definitely highly correlated and probably found in dictionaries
20181116 14:26:04< hk238> but those dictionaries have a lot more than 32 words
20181116 14:26:08< hk238> it's more like a thousand words
20181116 14:26:20<+wesdiscordbot> <jyrkive> In the end, only the pure entropy of the password matters. All "methods to generate passwords" only work until someone develops a countermeasure.
20181116 14:26:41<+wesdiscordbot> <jyrkive> So I strongly recommend against using such methods.
20181116 14:27:02< hk238> however it's also possible that your estimate of 32 words is way too low
20181116 14:27:13< hk238> and even this this method results to a much stronger password than your password manager
20181116 14:27:23< hk238> okay if you claim that the pool you draw these actual words from is roughly the size of the alphabet
20181116 14:27:30< hk238> then yeah okay it's not so great
20181116 14:27:37< hk238> but obviously there a lot more words than there are letters in the alphabet
20181116 14:27:43<+wesdiscordbot> <jyrkive> According to my calculations, your method loses by an order of magnitude.
20181116 14:27:53< hk238> yes but those calculations are not very credible
20181116 14:28:07<+wesdiscordbot> <jyrkive> Ignoring its actual problem, that you need to remember a very long sentence for every single password.
20181116 14:28:21< hk238> that's actually not very hard because it's easy to remember
20181116 14:28:32< hk238> okay maybe we should instead use this method for evaluating the difficulty
20181116 14:28:48<+wesdiscordbot> <jyrkive> It's easy to remember because it doesn't have anywhere near as much entropy as you think.
20181116 14:28:52< hk238> assume the bruteforce dictionary search is actually not random but based on having a ranking for the words
20181116 14:28:57<+wesdiscordbot> <josteph> Right
20181116 14:29:22<+wesdiscordbot> <josteph> If passwords are easy to remember you can expect people to write neural networks for predicting how easy to remember some sentence is...
20181116 14:29:44< hk238> coz that password is extremely difficult to crack if you're usign a 1000 words dictionary, since it's still 9 words, you have 1000^9 combinations
20181116 14:29:59< hk238> but let's assume that's not the case and instead the bruteforcers hackers have created a database
20181116 14:30:06< hk238> which ranks every english word by popularity
20181116 14:30:10< hk238> so they have like 'the most common word'
20181116 14:30:15< hk238> and 'the second most common word'
20181116 14:30:25< hk238> all the way down to the '300.000th most common word'
20181116 14:30:41< hk238> and then it does a selective search based entirely on the ranking of the words
20181116 14:30:59< hk238> now to claim that there's 5 words in my sentence that rank in the top 32 is ludicrous
20181116 14:31:14< hk238> you might be better off saying they're in the top 500 
20181116 14:31:19< hk238> and not all of them are either
20181116 14:32:01<+wesdiscordbot> <jyrkive> To be fair, I used it as an approximation for what the tool would actually do. It would try the correct sentence structure from the get-go.
20181116 14:32:23< hk238> right so adjective, noun? and so on?
20181116 14:32:31<+wesdiscordbot> <jyrkive> "My [words] likes [words] with words"
20181116 14:32:58< hk238> I don't think there are a lot of tools that figure out the 'with' part there
20181116 14:33:08< hk238> but I agree my [words] likes [words] does sound credible
20181116 14:33:28< hk238> but that's only after they have picked the 'my' words
20181116 14:33:30< hk238> and the 'likes' word
20181116 14:33:31<+wesdiscordbot> <jyrkive> "My, "likes", and "with" do buy you some entropy... and TBH, I think giving even 15 bits for them is too generous.
20181116 14:34:12< hk238> knowing this order for the two words does reduce the entropy a bit
20181116 14:34:13<+wesdiscordbot> <josteph> I don't think there are 32k sentence structures of which that is one, no.
20181116 14:35:10< hk238> I don't think 15 bits is too generous
20181116 14:35:18< hk238> you might say that for juts the 'my' and 'likes' part
20181116 14:35:31< hk238> but okay if we assume the hackers are using 
20181116 14:35:49< hk238> okay let's put it this way
20181116 14:36:02< hk238> even if the bruteforce tool was actually using a neural network to do all the guesswork
20181116 14:36:13< hk238> even running the network requires computations
20181116 14:36:26< hk238> and you'd need to run it for every word you're getting from the dictionaries
20181116 14:36:32< hk238> it's probably not even going to speed up the search
20181116 14:37:09<+wesdiscordbot> <josteph> Can you generate a password from driving directions or something? I make more than 30 turns on my way to the supermarket, can I get a password with 30 bits of entropy out of that?
20181116 14:37:13< hk238> oh chess world championship match starts in 20 minutes... 
20181116 14:37:24<+wesdiscordbot> <jyrkive> We can expect cracking technology to improve if using phrases as passwords becomes common.
20181116 14:37:35< hk238> yeah okay it can improve
20181116 14:37:40< hk238> but there's the thing with this that
20181116 14:37:45< hk238> longphrases are hard to crack because
20181116 14:37:55< hk238> for the very same reason long random strings are hard to crack
20181116 14:37:59< hk238> even if you can pick the words from dictionaries
20181116 14:38:06< hk238> the number of combinations tends to grow a lot
20181116 14:38:12<+wesdiscordbot> <jyrkive> The constant factor caused by NN computations may be optimized away, with things like the "trying common sentences" search I suggested above.
20181116 14:38:39<+wesdiscordbot> <josteph> hk, that's just a different way of saying they have a high entropy.
20181116 14:38:44< hk238> yeah but this requires extremely nice tools like if you combine all these approaches then okay yeah I agree
20181116 14:38:48<+wesdiscordbot> <jyrkive> Again, I think your grandmother example has only five words with acceptable entropy, and even they are fairly common words.
20181116 14:38:51< hk238> it's not the best possible password phrase
20181116 14:39:09< hk238> but it's still pretty strong 
20181116 14:39:20<+wesdiscordbot> <josteph> Phrases might work for now because most people don't use phrases, so hackers that don't target you specifically won't bother spending time on your phrase password
20181116 14:39:43< hk238> no the reason phrases work are because it's a memory efficient way of creating longer strings
20181116 14:39:55< hk238> even if you use a dictionary it doesn't help that much
20181116 14:40:14< hk238> because there are a lot of words
20181116 14:40:21<+wesdiscordbot> <jyrkive> Human memory isn't magic. The truth is that phrases are easy to remember because they don't have that much entropy in the end.
20181116 14:40:30< hk238> No that's not true
20181116 14:40:35<+wesdiscordbot> <josteph> hk, do you have a reference for your statement?
20181116 14:40:38< hk238> they're easy to remember because they encode information
20181116 14:40:49< hk238> and the world is complex, so there's a lot of information to encode
20181116 14:41:07< hk238> but it's true that the words when you create a phrase
20181116 14:41:14< hk238> are english language or finnihs language
20181116 14:41:22< hk238> and if you combine language capacity with dictionary searches
20181116 14:41:36< hk238> then theoretically that reduces the entropy a little bit
20181116 14:41:41< hk238> but not much 
20181116 14:41:50< hk238> the order of the words isn't a huge factor in the first place
20181116 14:42:02< vn971> @jyrkive I thought a bit about what you said:
20181116 14:42:02< vn971> > I am not aware of even one virus that targets password managers (ones with master passwords, to be precise).
20181116 14:42:02< vn971> Obviously there should be viruses targeting password managers. It's too damn straight approach to refrain from it. And indeed, there's at least something: https://malwarefixes.com/remove-advanced-password-manager/  Will you count it?
20181116 14:42:10<+wesdiscordbot> <jyrkive> Searches don't "reduce entropy". The entropy in your password is fixed.
20181116 14:42:25< hk238> I need to watch the chess world championships, but I also get pizza on fridays
20181116 14:42:26< vn971> // even if you won't, I'll still assume password managers to be an extremely high security risk area anyway.
20181116 14:42:28< vn971> TBH
20181116 14:42:35< hk238> but now if I get pizza, I might not get to see it from the beginning. Huge dilemma
20181116 14:42:36< hk238> :D
20181116 14:42:47<+wesdiscordbot> <josteph> Order isn't a factor? The factorial function is superexponential, it adds a lot of entropy
20181116 14:42:47< hk238> okay I got to go at least to buy some catfood, brb
20181116 14:42:57<+wesdiscordbot> <jyrkive> You might simply get temporary relief when search tools miss the lack of entropy and have to fall back to pure character brute-force.
20181116 14:43:08< hk238> yeah if you have 9 words you have a coefficient 9! 
20181116 14:43:15< hk238> if you guess all the words that is
20181116 14:43:20< hk238> if you just guess 'my' is the first word
20181116 14:43:22<+wesdiscordbot> <josteph> 9! is something like 20 bits
20181116 14:43:26< hk238> you're still left with 8!
20181116 14:43:38< hk238> but for 5 word phrases 5! is nothing
20181116 14:43:39<+wesdiscordbot> <jyrkive> For example, keyboard walks generate passwords which look strong at a first glance, but in reality they have hopelessly weak entropy and a tool that's aware of them can crack them in a fraction of a second.
20181116 14:44:10<+wesdiscordbot> <josteph> 5! is 7 bits
20181116 14:44:16< hk238> if you can find a bruteforce tool that breaks that grandmother password then I admit you're right, but I don't thin kthere's one currently used by anyone anywhere
20181116 14:44:19< hk238> maybe in the future there will be
20181116 14:44:28< hk238> anyway I got to it was an interesting discussion though!
20181116 14:45:31<+wesdiscordbot> <jyrkive> There isn't one at the moment because there are more than enough people around who use even weaker passwords. Targeting long phrases isn't worth it at the moment.
20181116 14:46:36<+wesdiscordbot> <jyrkive> Again: my main gripe with phrases isn't their insecurity. It's that it's too much cognitive load to attempt to remember a couple hundred phrases (which is how many you need in the real world, source: my password database).
20181116 15:26:43-!- Ravana_ [~Ravana@unaffiliated/ravana/x-2327071] has quit [Ping timeout: 245 seconds]
20181116 16:32:42-!- Ravana_ [~Ravana@unaffiliated/ravana/x-2327071] has joined #wesnoth-umc-dev
20181116 17:31:35-!- vn971 [~vasya@office.zivver.org] has quit [Quit: Leaving.]
20181116 19:45:44-!- hk238 [~kvirc@unaffiliated/hk238] has quit [Quit: KVIrc 5.0.0 Aria http://www.kvirc.net/]
20181116 19:50:22-!- hk238 [~kvirc@unaffiliated/hk238] has joined #wesnoth-umc-dev
20181116 20:58:04-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 250 seconds]
20181116 22:12:01< hk238> Chess WC match over.. So now back to figuring out why this addon isn't working
20181116 22:15:35< hk238> I still have the same problem that when trying to select era, it fails and default is selected instead
20181116 22:15:46< hk238> let's see then
20181116 22:18:00< hk238> okay I tried removing all the units from the addon so the unit files have no error. I also disabled the multiplayer side
20181116 22:19:18< hk238> that doesn't leave a lot to look at
20181116 22:20:00< hk238> https://github.com/n832sv/dunefolk_modified/blob/master/_main.cfg
20181116 22:20:06< hk238> is this somehow erroneous?
20181116 22:21:06< hk238> abilities.cfg is currently empty. Maybe that's what causes the error :p
20181116 22:21:40< Ravana_> is it intended that /units/dunefolk is not included?
20181116 22:22:38< hk238> oh wait can the .git stuff cause the addon not to work?
20181116 22:22:44< hk238>  ididnt have hidden files visible so I didnt realize they're there
20181116 22:23:00< hk238> no it's not intended what do you mean?
20181116 22:23:04< Ravana_> not included = not preprocessed into wml document
20181116 22:23:23< hk238> right line 11 should have the subfolder dunefolk?
20181116 22:23:34< Ravana_> yes
20181116 22:23:44< Ravana_> ageless handles that by having _main.cfg for subfolders too
20181116 22:25:06< hk238> okay but I tried this without the units so that wasn't the source for the addon not working at all, but of course that still has to be fixed. I'm guessing the .git subfolder causes the error, let's try that
20181116 22:25:52< Ravana_> unless you specifically {} include .git or similar, it won't have any effect
20181116 22:26:07< hk238> okay yeah that wasn't the reason either
20181116 22:26:19< Ravana_> maybe even then, I remember recent changes making more files invisible for API
20181116 22:26:20< hk238> well this is perplexing the only file I think can be the cause for this is the specials.cfg
20181116 22:26:53< hk238> yeah there's a typo in the specials.cfg that's probably it
20181116 22:28:03< hk238> nope that wasn't it, or at least not the only reason..
20181116 22:28:48<+wesdiscordbot> <Ravana> check if wesnoth --preprocess gives the output that you expect
20181116 22:29:14< hk238> how do I do that? I don't know what that means
20181116 22:30:00< Ravana_> command line option for wesnoth, --help gives description of it
20181116 22:31:23< hk238> I probably have to run that in terminal since when running that with launcher nothing happens?
20181116 22:31:44< Ravana_> yes
20181116 22:33:09< hk238> this seems complicated
20181116 22:33:22< Ravana_> example line how I can run it: ravana@ravana-think ~/wesnoth/userdata_1_14/data/add-ons/XP_Modification $ wesnoth14 --preprocess-defines MULTIPLAYER -p ~/wesnoth/userdata_1_14/data/add-ons/XP_Modification out
20181116 22:34:04< Ravana_> (good to mkdir something for output, otherwise I think it overwrites your _main.cfg)
20181116 22:37:43< Ravana_> removing ERA_DEFAULT include might be helpful to analyse result
20181116 22:38:23< hk238> there's like 3 cfg files here which of one has to be erroneous, _main.cfg dunefolk_modified.cfg and specials.cfg 
20181116 22:38:40< hk238> because I tried removing the dunefolk multiplayer side and the dunefolk folder and it still didnt wor
20181116 22:38:43< hk238> *work
20181116 22:38:56< Ravana_> what command you used?
20181116 22:40:03< hk238> command?
20181116 22:42:25< hk238> hm I have the era id before the era name, does that order matter?
20181116 22:42:51< Ravana_> no
20181116 22:43:38< Ravana_> I get "error preprocessor: Macro/file 'ERA_DEFAULT' is missing" when I cloned your repo and preprocessed it
20181116 22:44:45< Ravana_> if I comment that, then it succeeds and I see simple era structure
20181116 22:46:05< hk238> mm I tried deleting the era default part but it's still not working for mee
20181116 22:47:17< hk238> but I guess it's good to know that macro doesn't work either
20181116 22:47:43< Ravana_> it might work, but better to add it later when addons loads
20181116 22:48:51< hk238> this does work at least partially because the description shows, but when trying to select the era it fails
20181116 22:49:00< hk238> also there's no errors in console
20181116 22:49:05< hk238> just can't select it from the dropmenu
20181116 22:50:36< Ravana_> did you include unit files now?
20181116 22:50:53< hk238> I think so, maybe? 
20181116 22:51:05< hk238> but I already tried this after deleting the multiplayer side part and the dunefolk folder contents
20181116 22:51:11< hk238> so it's hard to see that thosecould cause the problem
20181116 22:51:45< hk238> but it could be there's several errors and I'm just toggling them on and off but not simultaneously
20181116 22:52:42< hk238> although
20181116 22:52:53< hk238> if the era ends up being empty, does that also mean it will fail to select?
20181116 22:53:17< Ravana_> when I clone your repo, comment era_default, then it works
20181116 22:53:21< Ravana_> I can start game with it
20181116 22:53:46< hk238> weird coz that doesn't happen to me
20181116 22:53:54< hk238> so 
20181116 22:54:12< hk238> maybe it's the .git files and the ERA_DEFAULT macro? Since I didn't try removing those at the same time?
20181116 22:54:22< hk238> well I'll try that next
20181116 22:54:32< Ravana_> since I cloned it, I have .git files just as you
20181116 22:55:24< hk238> yeah it still doesn't work for me though
20181116 22:56:05< hk238> ah actually it might not be this addon that's bugged
20181116 22:56:08< hk238> but eternal era instead
20181116 22:56:19< hk238> coz all the addons upto that addon in alphabetical order succeed 
20181116 22:56:34< hk238> no that's not it but they both probably have an error
20181116 22:56:52< hk238> I'll remove some addons anyway to see if there's something with them causing it
20181116 22:58:13< hk238> hm when I tried to delete dunefolk modified it failed
20181116 22:58:58< hk238> using remove addons so something fishy is going on with that
20181116 22:59:10< hk238> maybe it's the gitkraken that does something that causes problems?
20181116 22:59:19< hk238> the software I use for github?
20181116 23:02:27< hk238> it works now 
20181116 23:02:48< hk238> after removing some addons and deleting the folder and copying it again, and then deleting the git files, not sure if the git files had anything to do with it or not
20181116 23:02:54< hk238> but it might have beeen some kind of a conflict with some other addon 
20181116 23:03:54< hk238> okay so how do I include the default units?
20181116 23:08:06<+wesdiscordbot> <sevu> cour _main.cfg is empty
20181116 23:08:10<+wesdiscordbot> <sevu> *your
20181116 23:08:32<+wesdiscordbot> <sevu> it does what is written there – nothing
20181116 23:09:19<+wesdiscordbot> <sevu> (assuming we talk about https://github.com/n832sv/wesnoth_vampire_faction/)
20181116 23:34:19<+wesdiscordbot> <sevu> https://github.com/n832sv/dunefolk_modified/commit/24b28f8f469226622f05bbea79260a26fdd47a91 …
20181116 23:38:01< hk238> nope we're talking about the modified dunefolk thing actually 
20181116 23:38:16< hk238> I havent updated the vampire faction in a bit although it's missing so little work
20181116 23:38:25< hk238> I have some motivation problems with finishing it but I really need to do that at some point
20181116 23:38:34< hk238> it would take maybe 30 minutes as things are now (and hours of bug fixing)
20181116 23:49:59< hk238> ?
20181116 23:50:33< hk238> sevu yes I included the server.pbl file by accident at first, but since I had not uploaded it to the addon server I can change the password
20181116 23:50:46< hk238> and still havent uploaded 
20181116 23:54:03<+wesdiscordbot> <sevu> [units] {~add-ons/dunefolk_modified/units/dunefolk} [/units]
20181116 23:54:15<+wesdiscordbot> <sevu> you miss the dunefolk at the end
20181116 23:57:02-!- sevu [~sevu@p5485415E.dip0.t-ipconnect.de] has joined #wesnoth-umc-dev
20181116 23:57:25< sevu> hk238, ^
20181116 23:57:32< hk238> yeah that's that part is already fixed the github version is outdated
20181116 23:57:38< hk238> it works now and I*m just going to push the updated version
20181116 23:57:53< sevu> ah, okay
20181116 23:58:08< hk238> https://github.com/n832sv/dunefolk_modified
20181116 23:58:17< hk238> https://github.com/n832sv/dunefolk_modified/blob/master/units/specials.cfg
20181116 23:58:26< hk238> what do you think are these tasteless or unrealistic or problematic?
20181116 23:58:46< hk238> also I didn't modify the lvl2,lvl3 versions of the units, this is just for initial testing
20181116 23:59:09< hk238> the firearcher has the naphthaline special, and piercer is changed to.. ahh speaking of which I forgot to update the description in the server pbl
20181116 23:59:11< hk238> well anyway
20181116 23:59:25< hk238> piercer has 8-2 with the onslaught on the lance instead of 20-1
--- Log closed Sat Nov 17 00:00:21 2018