--- Log opened Sun Sep 30 00:00:15 2018 20180930 00:01:57-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection] 20180930 00:35:38-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection] 20180930 00:35:44-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev 20180930 00:38:53<+wesdiscordbot> they should really remind people on GH that you need to append /github to the payload url 20180930 00:39:11<+wesdiscordbot> That's a Discord thing. 20180930 00:39:18<+wesdiscordbot> Why would it be GH's responsibility to remind you that? 20180930 00:39:26<+wesdiscordbot> fair point 20180930 00:49:51< irker170> wesnoth/wesnoth:master sigurdfdragon 75ab69d885 DM S08: Cleanup & improve ending sequenc AppVeyor: All builds passed 20180930 02:31:34-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection] 20180930 02:31:43-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev 20180930 03:07:04< irker170> wesnoth/wesnoth:1.14 Joseph Gelfand e73f10d47b Modified tests as per discussion with jy AppVeyor: All builds passed 20180930 03:44:27< celticminstrel> WTH does "append /github to the payload URL" even mean 20180930 03:59:23<+wesdiscordbot> celmin: I was setting up the webhook for the new repo for the godot prototype and I didn't realize you needed to append /github at the end 20180930 03:59:58< celticminstrel> That doesn't even make any sense but okay whatever. 20180930 04:00:29<+wesdiscordbot> ¯_(ツ)_/¯ 20180930 04:00:34<+wesdiscordbot> discord makes it so 20180930 04:01:11< celticminstrel> That's not an explanation. Not that I was asking for one mind you. 20180930 04:01:48<+wesdiscordbot> anyway, once development gets rolling on that (bitron has stated work) I'm probably gonna make a new channel here so if you want to join in the discusion you'll need to join discord 20180930 04:04:25< celticminstrel> Why would I want to join the discussion? 20180930 04:04:46< celticminstrel> Also I'm not getting Discord so if you want me in the discussion you'll need to make it accessible through some other channel. 20180930 04:04:48<+wesdiscordbot> who knows 20180930 04:06:02< celticminstrel> I still think this is the wrong path to take. 20180930 04:20:48<+wesdiscordbot> it is most certainly not the wrong path to take 20180930 04:21:16< celticminstrel> There's no certainty involved here for either side of the argument. 20180930 04:21:38< celticminstrel> If you're saying there 's certainty it just makes me even less confident that you know what you're doing. 20180930 04:21:44<+wesdiscordbot> it apparently took bitron only like an hour to get some terrains loaded up and running in godot. 20180930 04:21:54< celticminstrel> Fun. 20180930 04:22:11< celticminstrel> But in order for it to be viable you still need about 90% compatibility with old Wesnoth. 20180930 04:22:23<+wesdiscordbot> nope 20180930 04:22:31<+wesdiscordbot> can we not have this argument again 20180930 04:22:37<+wesdiscordbot> we literally will never say anything different 20180930 04:22:50<+wesdiscordbot> I say no compatibility, you say a lot of compatibility 20180930 04:22:52< celticminstrel> Your project is going to fail if you don't have compatibility. 20180930 04:23:11<+wesdiscordbot> and I say otherwise 20180930 04:23:23< celticminstrel> You can't replace an existing popular thing with something new and not have compatibility and still succeed. This is well tested. 20180930 04:23:38<+wesdiscordbot> If you're saying there's certainty it just makes me even less confident in your argument. 20180930 04:24:27< celticminstrel> I've cited specific examples to back up my claim though. 20180930 04:24:38< celticminstrel> Last time the argument came up, that is. 20180930 04:24:55< celticminstrel> I don't recall you citing any examples to show that it could succeed. 20180930 04:26:28<+wesdiscordbot> I’m trusting it will be attractive enough for players to port their basic stuff and to encourage new modders to make new and better stuff using new capabilities 20180930 04:27:21< celticminstrel> Your trust is misplaced. 20180930 04:28:30<+wesdiscordbot> @Vultraz You attempted one ambitious thing(a_r), and for all intents and purposes it failed. Now you're trying something even more ambitious, even if it is just a prototype for now. Like I said on the forums, I'll help however I can, but I also can't say this approach creates a great deal of confidence. 20180930 04:30:27<+wesdiscordbot> What would be helpful for the current version of wesnoth, which is released and being played by people, would be to have its master branch finally in a working state, one way or another. 20180930 04:30:51< celticminstrel> It's bad enough that addons haven't been compatible between stable versions, but at least they only required a few tweaks here and there. Now you're telling addon devs that all their old work is wasted and they need to rewrite the addon from scratch in the new API? Many of them will just quit. 20180930 04:31:35< celticminstrel> I'm not going to say you have to be 100% compatible, but if you don't have a layer of compatibility, you'll lose a ton of people. 20180930 04:31:45< celticminstrel> You won't lose everyone, I'm sure, but... 20180930 04:32:08<+wesdiscordbot> This new ambitious thing is meant to sidestep the complications with the other ambitious thing 20180930 04:32:36< celticminstrel> "meant to" 20180930 04:32:41<+wesdiscordbot> Whereas we can focus on implementing actual improvements like a new game UI and better lighting effects and everything without spending 10 years writing the backend 20180930 04:34:33< celticminstrel> Did you ask Jyrki if it would actually take 10 years, because that seems pretty high for an estimate. 20180930 04:34:51<+wesdiscordbot> also, I think part of the compatibility question should involve asking UMC authors what they think. otherwise how many people will or won't leave is pretty speculative at best. 20180930 04:35:00< celticminstrel> True. 20180930 04:35:36< celticminstrel> I'm basing my analysis on things that I've read about DOS and Windows 3 and why they were so successful when other similar OSes weren't. 20180930 04:35:49<+wesdiscordbot> This is not a discussion I want to have with the general modding community until we have something concrete 20180930 04:36:10<+wesdiscordbot> You've seen the shitstorms that ariss when things like this are discussed in public 20180930 04:36:15< celticminstrel> But ... compatibility doesn't need to be the first priority. 20180930 04:36:15<+wesdiscordbot> ie: IIRC you've mentioned inferno8 as an example of how a brand new API could make things easier for him. but if he says he wouldn't want to start over from scratch anyway, even if it would be 1/4 the total effort or whatever the number would be, then what? 20180930 04:36:24< celticminstrel> Get a prototype working first, then add a compatibility layer. 20180930 04:36:38<+wesdiscordbot> I do not trust the forums to offer constructive suggestions 20180930 04:36:49< celticminstrel> That's unfortunate though. 20180930 04:37:28< celticminstrel> Mind you, the forums is a much more general audience than you'd need. 20180930 04:37:45< celticminstrel> Because it also includes players who haven't even considered modding. 20180930 04:38:05<+wesdiscordbot> PMing people would be better, or at least something in that format. 20180930 04:38:06<+wesdiscordbot> if anything, I would consult only with specific modders like inferno8 and shadowm 20180930 04:39:09< celticminstrel> That's not a good approach. 20180930 04:39:26< celticminstrel> If you only talk to a few people, you won't get a good overview of people's views. 20180930 04:41:41<+wesdiscordbot> I actually sort of agree with Vultraz on this one but for a different reason 20180930 04:41:58<+wesdiscordbot> while I'll always advocate UMC to new players 20180930 04:42:16<+wesdiscordbot> the base game should be prioritized over mods imo 20180930 04:42:35< irker170> wesnoth/wesnoth:1.14 josteph 8603729d16 Update changelog AppVeyor: All builds passed 20180930 04:42:58<+wesdiscordbot> The majority of modders won't maintain their UMC long-term. Likewise, those people are not likely to know the modding API in-depth enough to offer in-depth analysis of what's needed and how difficult it would be to port 20180930 04:43:08<+wesdiscordbot> it's more important to have a high quality mainline game, and if it really means sacrificing some UMC, then that's the cost 20180930 04:43:42<+wesdiscordbot> whereas people with huge mods have significant knowledge of what is and is not possible, how they work around these limitations, and likely want to maintain their mods long-term 20180930 04:43:53<+wesdiscordbot> and would therefor have a desire to port to series 2 20180930 04:44:00< celticminstrel> Yumi does have a point about different levels of modders, I guess. 20180930 04:44:14<+wesdiscordbot> you mean vultraz? 20180930 04:44:17<+wesdiscordbot> 😮 20180930 04:44:20< celticminstrel> No I mean Yumi. 20180930 04:44:21<+wesdiscordbot> if an author wasn't even going to port to a new stable version they won't port to 2.x anyway 20180930 04:44:57<+wesdiscordbot> my only annoyance is that porting to 2.x means rewriting all WML 20180930 04:45:09<+wesdiscordbot> but again, a high quality mainline is more important than UMC to me 20180930 04:45:16<+wesdiscordbot> that is also true 20180930 04:45:22< celticminstrel> Exactly, that's a huge annoyance, and might well be a deal-breaker. 20180930 04:45:33<+wesdiscordbot> but from the player perspective 20180930 04:45:43<+wesdiscordbot> having a base game where we're equipped to make even better mainline content will be a huge benefit 20180930 04:45:45<+wesdiscordbot> the first thing they see is mainline, and most will stick to it without touching umc 20180930 04:46:08<+wesdiscordbot> we've already discussed and agreed that much of mainline is significantly lacking 20180930 04:46:18< celticminstrel> True. But note that all the mainline campaigns would also need to be rewritten from scratch, so UMC is not the sole concern WRT compatibility. 20180930 04:46:24<+wesdiscordbot> in writing, gameplay, and map design. 20180930 04:46:25<+wesdiscordbot> I agree 20180930 04:46:25<+wesdiscordbot> but 20180930 04:46:32<+wesdiscordbot> I think they need to be rewritten anyway 20180930 04:46:48< celticminstrel> From scratch? 20180930 04:47:02<+wesdiscordbot> some do, yes 20180930 04:47:03<+wesdiscordbot> many of them, more or less 20180930 04:47:17<+wesdiscordbot> some, like TRoW, just need cosmetic and QOL udates 20180930 04:47:24<+wesdiscordbot> I'll disagree with you even on that one 20180930 04:47:25<+wesdiscordbot> but NR? neeeinnn 20180930 04:47:29<+wesdiscordbot> TRoW is too long 20180930 04:47:30< celticminstrel> I'm inclined to disagree here. Some need some rewrites, but they don't need to be rewritten from scratch. 20180930 04:47:31<+wesdiscordbot> and has too much filler 20180930 04:47:37< celticminstrel> I have no opinion on NR specifically though. 20180930 04:47:38<+wesdiscordbot> ah I misunderstood 20180930 04:47:55<+wesdiscordbot> rewritten from scratch is few, agree with celmin on that 20180930 04:48:01<+wesdiscordbot> also agre 20180930 04:48:09<+wesdiscordbot> probably one or two, that's it 20180930 04:48:15<+wesdiscordbot> only some need complete rewrites 20180930 04:48:27<+wesdiscordbot> that's what I meant when i say TRoW just needs cosmetic and QOL updates 20180930 04:48:30<+wesdiscordbot> it doesn't need a rewrite 20180930 04:48:38<+wesdiscordbot> I think you can get rid of some scenarios 20180930 04:48:39<+wesdiscordbot> and condense others 20180930 04:48:42< celticminstrel> Even what you're referring to as a complete rewrite, it wouldn't need to be entirely from scratch. 20180930 04:48:48<+wesdiscordbot> but I suppose you could consider that QOL 20180930 04:48:56< celticminstrel> You could probably reuse a lot of code if you kept compatibility. 20180930 04:49:19< celticminstrel> Might get shuffled around a bit, some might be removed, dialogue could well be entirely different, but... that's not the same as starting from nothing. 20180930 04:49:24<+wesdiscordbot> no mainline scenario is so complex that it requires a year to rewrite 20180930 04:49:30<+wesdiscordbot> no mainline campaign* 20180930 04:49:44< celticminstrel> How long did they take to write in the first place? 20180930 04:49:55<+wesdiscordbot> who knows 20180930 04:50:18< celticminstrel> Without that information it's impossible to gauge whether your statement is reasonable. 20180930 04:50:36<+wesdiscordbot> I'm saying that we still have the old code as a reference 20180930 04:50:39< celticminstrel> Though I'll admit a year does seem a little long for writing a campaign from scratch, I'd estimate more like three to four months, probably. 20180930 04:50:43<+wesdiscordbot> and 20180930 04:50:58< celticminstrel> But it also depends on the length - a campaign like HTTT for example might take quite a bit longer. 20180930 04:51:00<+wesdiscordbot> given that the scenarios in the mainline campaigns are mostly simple converting them should not take too much time 20180930 04:51:27<+wesdiscordbot> there are also possibly cases where complexity is due to the current engine's limitations 20180930 04:51:28<+wesdiscordbot> celmin: Genesis I took me about a year total to finish 20180930 04:51:46<+wesdiscordbot> you also can't consider simply the original amount of time it took to create them 20180930 04:51:53< celticminstrel> You're speculating now. That aside, not having to convert them because there's a compatibility will take even less time, and then you get all of UMC for free as a bonus. 20180930 04:51:55<+wesdiscordbot> since the maps and dialog were being designed from scrarch 20180930 04:52:00< celticminstrel> Or at least most of UMC. 20180930 04:52:13<+wesdiscordbot> whereas even with a rewrite, we're starting with a good outline and an idea of where to go 20180930 04:52:25<+wesdiscordbot> dialogue still probably needs to be rewritten 20180930 04:52:34<+wesdiscordbot> however a plot at least will be there 20180930 04:52:43<+wesdiscordbot> indeed 20180930 04:52:45< celticminstrel> @Vultraz - If you'd given me an actual estimate of how long they took intially, I was going to come back with "so assume maybe around half that for a rewrite" but since you didn't know I didn't get the chance to say that. 20180930 04:53:06< celticminstrel> So yeah, I'm not saying a rewrite from scratch takes the same amount of time as writing a new campaign from scratch. 20180930 04:53:14<+wesdiscordbot> I think t/2 is relevant whatever t is 20180930 04:53:58<+wesdiscordbot> and I don't know because many were written years, even over a decade, ago 20180930 04:54:01< celticminstrel> But how long would it take to program a compatibility layer? Two or three months, maybe? And then you don't have to convert them at all, or more likely you can convert them leisurely as you find time. 20180930 04:54:11<+wesdiscordbot> by people of varying skill in writing and coding 20180930 04:54:25< celticminstrel> And most UMC would continue to Just Work. So are there really any strong disadvantages to having a compatibility layer? 20180930 04:55:33<+wesdiscordbot> given the scope of Wesnoth's current API, and frankly how much of it seems to break between stable releases, I wouldn't be overly confident that a compatibility layer would be that simple either. 20180930 04:56:32< celticminstrel> Well, Vultraz was already saying he might well keep a lot of the Lua API, in which case the compatibility layer would be just the WML. 20180930 04:56:59< celticminstrel> So I don't think it would be that hard. There might be some complications though, so I can't be certain. 20180930 04:58:49<+wesdiscordbot> there are all kinds of WML corner cases 20180930 04:58:51<+wesdiscordbot> and inconsistencies 20180930 04:59:12< celticminstrel> Inconsistencies are easy TBH. Corner cases might not be though. 20180930 04:59:51< celticminstrel> FTR I'm thinking of inconsistencies in key names, like some places use red,green,blue and others use color; I suppose there could be other types of inconsistencies that are less easy. 20180930 05:00:10<+wesdiscordbot> maybe this would be an opportunity to add automated unit tests for these sorts of things, compatibility layer or not. 20180930 05:00:32<+wesdiscordbot> the current unit tests are tiny compared to the API 20180930 05:05:36< celticminstrel> It would indeed be useful - that kind of thing is why the WML unit test system was added in the first place. 20180930 05:06:13< celticminstrel> But the thing is, it's a lot of work to build decent unit tests for an already-mature system. 20180930 05:06:32< celticminstrel> So that's why there aren't really enough tests yet. 20180930 05:06:57<+wesdiscordbot> well 20180930 05:07:17<+wesdiscordbot> part of wesnoth's problem is also that it's impossible to add too many more tests 20180930 05:07:29<+wesdiscordbot> add another 50, and you'll start getting travis time outs 20180930 05:07:36<+wesdiscordbot> especially for -O0 builds 20180930 05:08:14<+wesdiscordbot> having a policy of "if you add to the API, you must also add unit tests for it" would also be necessary to enforce. 20180930 05:09:20<+wesdiscordbot> WML unit tests, to clarify 20180930 05:15:43<+wesdiscordbot> another thing to consider too, is that it's not necessarily a guarantee that Wesnoth 1 will end when Wesnoth 2 is ready. If current UMC authors don't port their stuff, and current developers don't want to move over either, you could well end up with a situation where Wesnoth 1 and Wesnoth 2 are actively being developed at the same time. 20180930 05:16:27< celticminstrel> Yeah, and if that does happen I'll probably stick to Wesnoth 1, and if there's no compatibility in Wesnoth 2 I definitely will. 20180930 05:18:20<+wesdiscordbot> I've said before series 1 would remain in bugfix-only status for at least a few years 20180930 05:18:24<+wesdiscordbot> before being sunset 20180930 05:19:00<+wesdiscordbot> pretty much anyone could fork it and keep adding features though, if they wanted to. 20180930 05:19:15<+wesdiscordbot> sure, but we wouldn't publish it 20180930 05:20:49<+wesdiscordbot> well, how many current developers have said they'd be willing to work on Wesnoth 2? celmin won't, based on what you've said on compatibility, and jyrkive has said he's contributing to Wesnoth so he can get more familiar with C++. 20180930 05:21:02<+wesdiscordbot> have any current devs said they would move to Wesnoth 2? 20180930 05:21:16<+wesdiscordbot> i suppose if there's really a group who wants to improve series 1, they can 20180930 05:22:41< celticminstrel> Jyrki has said he won't. I don't recall anyone specifically saying they will. 20180930 05:22:48< celticminstrel> I think Tad said he won't. 20180930 05:23:13<+wesdiscordbot> honestly we'll just have to see how things go 20180930 05:23:54< celticminstrel> I'm sure they won't go exactly how you want them to. 20180930 05:24:48<+wesdiscordbot> who knows 20180930 05:26:56<+wesdiscordbot> I think, along with UMC authors, you should really reach out to current developers and see what they would want to have happen in order for them to consider moving to Wesnoth 2. It won't do you much good to have the perfect, new version of Wesnoth 2.0, and only 2 devs to maintain the whole thing. I realize that doing the prototype is mostly how you're going to be getting information on all of this and how it would 20180930 05:26:57<+wesdiscordbot> work, but the surety with which you're already saying things and the absolutes you're using to say them raise a giant red flag. At least to me. 20180930 05:27:42< celticminstrel> Indeed. 20180930 05:27:44<+wesdiscordbot> even if some of those things turn out differently, you're already driving people towards not contributing by saying them that way. 20180930 05:29:29<+wesdiscordbot> in that case, it might be worthwhile making a forum post stating that there is an intention to move 2.x to a new engine and that it would involve getting rid of WML 20180930 05:29:39<+wesdiscordbot> one that's visible to the public, I mean 20180930 05:29:54<+wesdiscordbot> that would only happen after a successful prototype 20180930 05:29:58<+wesdiscordbot> I would hope 20180930 05:30:46<+wesdiscordbot> so then reaching out to UMC authors just means PM? 20180930 05:31:17<+wesdiscordbot> I don't think that's something that really needs to be decided right now. 20180930 05:31:31<+wesdiscordbot> maybe I'm not understanding 20180930 05:31:50<+wesdiscordbot> above, you said that it would be worthwhile to reach out to current devs and UMC authors to see what it would take for them to move to 2.x 20180930 05:32:16<+wesdiscordbot> obvs devs can see the forum thread, but most UMC authors can't 20180930 05:32:33<+wesdiscordbot> right. maybe PMs, maybe DMs, maybe some non-public UMC-dev working group sort of thing, etc. 20180930 05:33:43<+wesdiscordbot> really it's the big UMCs to worry about 20180930 05:33:44<+wesdiscordbot> right now though it doesn't sound like there are anywhere near enough specifics to actually start doing that. 20180930 05:33:53<+wesdiscordbot> okay, sure 20180930 05:47:54< celticminstrel> BTW, just because you dislike him doesn't mean you can just ignore Dugi (for example). 20180930 05:48:08< celticminstrel> Or his addon. 20180930 05:48:16< celticminstrel> (Dislike him or his addon) 20180930 05:48:51< celticminstrel> It is after all one of the bigger and more popular addons. 20180930 05:49:18<+wesdiscordbot> on iOS especially, it seems like 20180930 05:49:21< celticminstrel> (I dunno whether it's him or LotI that Vultraz dislikes, or both.) 20180930 05:50:39<+wesdiscordbot> LotI is definitely an example of something that stresses the current Wesnoth engine/API, if nothing else 20180930 05:50:47< celticminstrel> Indeed. 20180930 05:51:02< celticminstrel> FWIW I've never played it and am not even in the least bit interested in doing so. 20180930 05:51:29< celticminstrel> ... 20180930 05:51:45< celticminstrel> I just realized September is almost over already. 20180930 05:52:13<+wesdiscordbot> I really don’t care about dugi 20180930 05:52:26<+wesdiscordbot> At all 20180930 05:52:27< celticminstrel> Yes, that's the problem right there. 20180930 05:52:31<+wesdiscordbot> He’s an asshole 20180930 05:52:44< celticminstrel> Oh? 20180930 05:53:09<+wesdiscordbot> I can’t even remember what happened anymore but I know he was an asshole 20180930 05:53:18< celticminstrel> I can't say I've interacted with him enough to judge that, but I've interacted with him enough to feel it might not be inaccurate. 20180930 05:54:02<+wesdiscordbot> And his campaign is bloated, messy, and painful to even look at. 20180930 05:54:21<+wesdiscordbot> Now, would he be one of those who benefits from new engine features? Absolutely 20180930 05:54:26< celticminstrel> But it's popular. If you tell people LotI isn't going to work on Wesnoth 2, you'll lose many people. 20180930 05:54:35<+wesdiscordbot> But I have 0 interest in making him some sort of consultant on the effort 20180930 05:54:50-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev 20180930 05:55:04< celticminstrel> And as a larger campaign it would be a lot more work for him to port, so I'd totally understand if he felt it's not worth the effort. 20180930 05:55:31< celticminstrel> It costs nothing to ask him what he thinks. You don't have to actually take his advice. Or if you don't want to talk to him you could ask someone else to ask for you on your behalf. 20180930 05:55:45< celticminstrel> Of course, if no-one wants to talk to him that's another story... 20180930 05:57:28<+wesdiscordbot> I could be a go-between, if that's really necessary 20180930 05:57:30<+wesdiscordbot> >.> 20180930 05:58:19< celticminstrel> TBH I think it would be better for someone who isn't Vultraz to do the asking. Someone who can attempt to spin it neutrally. (Which also means someone who isn't me.) 20180930 05:58:44<+wesdiscordbot> A PMed a bit with him a while ago after some incident or another, and my impression is that he's very blunt while also pretty bad at judging how to use tone and considering how people might interpret things. 20180930 05:59:15< celticminstrel> I think English isn't his first language, either? Which might not help with judging how people might interpret things? 20180930 05:59:19< celticminstrel> Not sure though. 20180930 05:59:29<+wesdiscordbot> nevertheless though, he has stuck around and maintained something as massive and popular is LotI. 20180930 05:59:35<+wesdiscordbot> that I'm not sure about. 20180930 06:02:02<+wesdiscordbot> *not sure about any language barrier, I mean 20180930 06:02:35< celticminstrel> ISTR something about that but I can't remember any details so I could be imagining it. 20180930 06:04:01< celticminstrel> Anyway I'm going to bed now. G'night. 20180930 06:04:01<+wesdiscordbot> his user notes are... interesting, to say the least. not sure if devs have the ability to read them though. 20180930 06:04:17<+wesdiscordbot> cya 20180930 06:04:27< celticminstrel> I do believe I'm able to read notes, yes. 20180930 06:04:46-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has quit [Quit: And lo! The computer falls into a deep sleep, to awake again some other day!] 20180930 06:12:47<+wesdiscordbot> take a look at those if you're able to then, if you want some of the backstory 20180930 06:13:18<+wesdiscordbot> though nothing too recent, since he's mostly been keeping to his own threads for a while now. 20180930 06:31:40<+wesdiscordbot> I can talk to Dugi 20180930 06:31:44<+wesdiscordbot> if it comes to it 20180930 06:33:05<+wesdiscordbot> iirc LotI is popular on the iOS port, probably because it's very rpg-esque 20180930 06:36:53<+wesdiscordbot> but that may not be a problem if iOS stays as 1.14 20180930 06:43:12-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has joined #wesnoth-dev 20180930 07:42:13-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Remote host closed the connection] 20180930 07:45:13-!- irker170 [~irker@wesnoth/umc-dev/bot/shikadibot] has quit [Quit: transmission timeout] 20180930 09:56:54-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection] 20180930 09:57:07-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev 20180930 10:04:14-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection] 20180930 10:04:20-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev 20180930 10:49:07-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev 20180930 11:40:08-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has joined #wesnoth-dev 20180930 12:12:21-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection] 20180930 12:12:27-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev 20180930 13:04:53-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has joined #wesnoth-dev 20180930 14:32:09-!- timotei__ [~timotei@wesnoth/developer/timotei] has joined #wesnoth-dev 20180930 14:34:38-!- timotei_ [~timotei@wesnoth/developer/timotei] has quit [Ping timeout: 245 seconds] 20180930 15:26:36<+wesdiscordbot> Anyone know where on the forums the world map of Irdya is? (ie old continent, great continent, morogor, green isle, and how they relate to each other) I remember seeing it, but can't find it. 20180930 15:34:40<+wesdiscordbot> maybe it's in here: https://forums.wesnoth.org/viewtopic.php?f=23&t=46477 20180930 15:44:32< celticminstrel> I don't recall a world map, only a great continent map... 20180930 15:47:00<+wesdiscordbot> thanks Bitron, I found it in a link in that thread https://forums.wesnoth.org/viewtopic.php?p=578990#p578990 20180930 16:31:54-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection] 20180930 16:32:00-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev 20180930 16:39:01-!- ToBeCloud [uid51591@wikimedia/ToBeFree] has quit [Quit: Connection closed for inactivity] 20180930 16:39:30-!- boucman [~rosen@wesnoth/developer/boucman] has joined #wesnoth-dev 20180930 17:03:28-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection] 20180930 17:03:34-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev 20180930 18:26:48<+wesdiscordbot> ^also fixes offset issues with the larger forest terrain sprites. 20180930 18:36:54-!- celmin [~celmin@unaffiliated/celticminstrel] has joined #wesnoth-dev 20180930 18:36:55-!- celticminstrel [~celmin@unaffiliated/celticminstrel] has quit [Disconnected by services] 20180930 18:36:55-!- celmin is now known as celticminstrel 20180930 18:48:15<+wesdiscordbot> shouldn't the commit notifications be sent to the #project-haldric channel, by the way? 20180930 18:53:16-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection] 20180930 18:53:22-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev 20180930 19:01:52<+wesdiscordbot> I agree 20180930 20:17:11< mattsc> celticminstrel: @Vultraz @All : about the backwards compatibility discussion, I haven't followed everything that was said, so maybe this was beaten to death already ... 20180930 20:17:20< mattsc> What if there were a "rough" converter of some sort from WML to whatever the new format is going to be. I mean, I understand that having a script that does a perfect job is a pipe dream, but ... 20180930 20:17:21< mattsc> If you told me that I have to rewrite Grnk from scratch, I don't know whether I'd do it. (I honestly don't know, could go either way.) But if there were a script that did 80% of the job for me, I'd probably go for it. 20180930 20:17:22< mattsc> Just a thought — and as I said, sorry if this has been discussed and I missed it. I've been traveling a lot again recently. 20180930 20:17:57< mattsc> Having said that, I’ll continue with my AI work no matter what. That’s more important to me than my campaigns at this time anyway. 20180930 20:18:01< celticminstrel> I don't think there's any reason why a "near-perfect" converter could not be made, though I guess it depends on what the prototype ends up with. 20180930 20:18:27< celticminstrel> And if it's possible and it exists then there's really no reason for it to be a standalone program rather than integrated into the engine. 20180930 20:19:09< celticminstrel> But all that aside... if there was a converter that was just "rough", even as a standalone program, I think that would do a lot to prevent people who don't want to redo their work to give up and leave. 20180930 20:19:16< celticminstrel> It might even convince me. 20180930 20:19:47< mattsc> Well, from my personal selfish perspective, it’d probably convince me. 20180930 20:20:16< mattsc> And it would mean we don’t have to carry the double standard (and roughly double maintanance) in the engine itself. 20180930 20:23:58< celticminstrel> I don't think that's the best course of action. I also don't think it would be double the maintenance, though it would be more maintenance. 20180930 20:25:26< mattsc> Well, the difference is that a stand-alone converter would only have to get most of it right, while built-in WML support needs to work perfectly. 20180930 20:26:27< celticminstrel> But if you do things right I think it could end up being almost a fire-and-forget sort of thing needing barely any maintenance, though you might have to update some things if deprecated stuff that the compatibility layer depends on is removed. 20180930 20:27:33< mattsc> Well, I’m not really arguing one way or the other. My own opinion on this is split, I see the advantages of both approaches. I’m just throwing the idea out there (if it has not been done before). 20180930 20:28:33< celticminstrel> I might've mentioned it before, but I can't remember. 20180930 20:45:17< mattsc> I think another advantage of having a stand-alone script would be that you can draw from a different pool of people. 20180930 20:45:28< celticminstrel> ??? 20180930 20:45:32< mattsc> Take me, for example. There’s no way I could be of significant help with the engine work. 20180930 20:45:35< celticminstrel> Oh, you mean someone else could maintain it? 20180930 20:45:51< mattsc> But I think I _could_, in principle, write a python script converter. 20180930 20:46:22< mattsc> I am not volunteering for it, mind you. Or at least no until I have written the perfect AI. ;-) 20180930 20:47:21< mattsc> But I think there are “others like me” out here. 20180930 20:48:28<+wesdiscordbot> Someone take over and fix AtS for me thanks 20180930 20:49:13<+wesdiscordbot> I wish I could pay someone to fix that single bug that's bothering me lol 20180930 20:49:14< mattsc> @shadowm: you take over Fred and I’ll do it. ;-) 20180930 20:49:32< mattsc> What’s that bug? 20180930 20:50:06<+wesdiscordbot> Something really convoluted that I can't properly explain without the full context, which includes major spoilers 20180930 20:50:22<+wesdiscordbot> But which boils down to developers pulling the rug from under my feet as usual 20180930 20:50:58< celticminstrel> Which is unfortunately almost inevitable when all the edge cases aren't properly documented/tested. :/ 20180930 20:51:15<+wesdiscordbot> Quality game engine design 😉 20180930 20:51:29<+wesdiscordbot> It's kind of like C++ in here 20180930 20:51:49-!- sevu [~sevu@p548543A7.dip0.t-ipconnect.de] has joined #wesnoth-dev 20180930 20:52:20< celticminstrel> Nothing to do with C++. 20180930 20:52:25<+wesdiscordbot> Except that instead of having to read a 5,000 pages spec to learn about situations where you must look out for undefined behaviour, undefined behaviour comes to you when you least expect it 20180930 20:52:29<+wesdiscordbot> Yes it has all to do with C++ 20180930 20:52:32<+wesdiscordbot> Not the language, but the analogy 20180930 20:53:03< mattsc> Well, is it related to gameplay, animations, … ? 20180930 20:53:04< celticminstrel> Oh, you're comparing it to the C++ spec, okay. 20180930 20:53:17<+wesdiscordbot> I'm comparing the engine to C++ 20180930 20:53:52<+wesdiscordbot> It could be written in Python and the analogy would still work because the end result is still a labyrinthine mess where no-one has the full picture and "edge cases" keep cropping up like weed 20180930 20:54:00<+wesdiscordbot> mattsc gameplay 20180930 20:54:03< mattsc> I am reasonably good at finding Heisenbugs and the like, but not very much with anything roughly related to “appearances” … 20180930 20:54:37<+wesdiscordbot> It's a weird interaction between the events engine and the gameplay code 20180930 20:54:48<+wesdiscordbot> Which worked fine for 1.10 and 1.12 but then changed in 1.14 20180930 20:55:16<+wesdiscordbot> I'll figure it out in time, I just need to completely rewrite my code to unravel it to its core 20180930 20:55:26< mattsc> In other words, if you can describe it reasonably well and provide a rough test case, send me a PM (so that there are no spoilers) and I am offering to have a look (without making any promises). 20180930 20:55:28< celticminstrel> ...probably because a certain someone decided to rewrite the events list at the last minute even though everyone said that was a bad idea. >_> 20180930 20:56:01<+wesdiscordbot> I wouldn't be surprised if that was the case but also I wouldn't rule out the cause being some other refactoring even earlier in 1.13.x 20180930 20:56:14< celticminstrel> True, there are probably other possible causes. 20180930 20:56:15<+wesdiscordbot> Incidentally, any opinions on this? https://github.com/project-ethea/After_the_Storm/commit/eda548ba30b9c0a16c3b89dea0864745ac829325 20180930 20:56:35<+wesdiscordbot> Vultraz is blaming the attribute's merger 20180930 20:56:43<+wesdiscordbot> Since they used to be share_shroud and share_map or so 20180930 20:56:53< celticminstrel> Didn't they both default to false though? 20180930 20:57:01<+wesdiscordbot> But I swear they both defaulted to faulse, so in theory the new attribute should default to none and it seems it does not 20180930 20:57:37< celticminstrel> I bet the ordering here is the culprit: https://github.com/wesnoth/wesnoth/blob/master/src/team.hpp#L67 20180930 20:58:03< celticminstrel> Those should probably be in reverse order. 20180930 20:58:07<+wesdiscordbot> Wouldn't be surprised. 20180930 20:58:16<+wesdiscordbot> Only problem is that it's like 2 years too late to fix that. 20180930 20:58:16< celticminstrel> Though I guess the relative order of all and shroud don't actually matter. 20180930 20:58:23< celticminstrel> Is it? 20180930 20:58:26<+wesdiscordbot> Yes. 20180930 20:58:38< celticminstrel> Because it's an API change and would cause OOS errors? 20180930 20:58:46<+wesdiscordbot> Yes. 20180930 20:59:01<+wesdiscordbot> Although I was looking at it from the "it's an API change" perspective and didn't even factor in synchronization. 20180930 20:59:01< celticminstrel> Right, that makes sense. 20180930 20:59:09<+wesdiscordbot> But if you put it that way yeah, that's even worse. 20180930 20:59:43<+wesdiscordbot> Who knows why I'm the one who keeps running into this stuff when I'm not even the most prolific content author. 20180930 20:59:51<+wesdiscordbot> Or the one who makes the most complicated crap. 20180930 21:01:20< mattsc> Other people, who make even less complicated stuff, run into things like this too (e.g. I have many times). You just notice it for yourself more than for others. :) 20180930 21:01:30<+wesdiscordbot> I did not notice it 20180930 21:01:39<+wesdiscordbot> Someone reported it to me on my bug tracker 20180930 21:01:54<+wesdiscordbot> can we add a regression test for this now, so it doesn't break again ? 20180930 21:02:03< mattsc> You notice it for your own add-ons more than for other people’s add-ons ... 20180930 21:02:42<+wesdiscordbot> I mean I swear people must've run into this and just shrugged and patched it in their own add-ons 20180930 21:03:02<+wesdiscordbot> So if they never reported back here it probably means they're used to this kind of thing by now 🤔 20180930 21:03:20< mattsc> I am sure that happens, yes. I am sure I have done so myself at times, esp. in the “early days”. 20180930 21:03:24<+wesdiscordbot> maybe they just wanted their add-on to run on 1.14.0, without waiting for .1 20180930 21:03:52<+wesdiscordbot> Except this kind of fix doesn't belong in 1.14.x to begin with as we already mentioned above 20180930 21:04:16< mattsc> Anyways, @shadowm , if you’d like to have a second set of eyes on it, send me a PM. 20180930 21:04:18<+wesdiscordbot> In theory someone should've seen it during 1.13.x alpha or beta and reported it and a dev would've fixed it for the next release 20180930 21:05:03<+wesdiscordbot> I don't know, just food for thought. It's a relevant issue to think about regardless of whether Wesnoth gets ported to a new engine or not 20180930 21:05:43< celticminstrel> I'm not sure if I ran into this. I might've? 20180930 21:06:47< celticminstrel> The problem with relying on people to detect these things in 1.13.x is a) a lot of people don't use dev series and b) a lot of add-on devs don't update their addons to dev series. 20180930 21:06:54<+wesdiscordbot> Exactly 20180930 21:07:20<+wesdiscordbot> And you'll keep relying on people until you have 100% test coverage (I feel like I said something to this effect back in 1.11.x) 20180930 21:07:44<+wesdiscordbot> How to achieve 100% test coverage though? The strategy thus far seems to be "get bug, fix bug, add test case". 20180930 21:07:59<+wesdiscordbot> But if you don't get a bug report then how do you find out you need to add a test case? 20180930 21:09:03< celticminstrel> Yeah, getting 100% test coverage is hard even if you were making the tests as you went; adding them after the fact... 20180930 21:09:38<+wesdiscordbot> The "how to get people to use dev releases" problem is not new either. 20180930 21:09:50<+wesdiscordbot> I was trying to deal with it as far back as 1.9.x. 20180930 21:10:25< sevu> test coverage… I came today among this link https://github.com/auchenberg/volkswagen/ 20180930 21:10:33<+wesdiscordbot> I never managed to come up with a good solution and nobody else was particularly interested in trying to understand the community. 20180930 21:10:40<+wesdiscordbot> ee 20180930 21:10:46<+wesdiscordbot> That's great xD 20180930 21:11:17<+wesdiscordbot> But yeah back in the day the dev team was like "we don't need no community to make dev releases" 20180930 21:11:24<+wesdiscordbot> It got us to this point 20180930 21:12:05<+wesdiscordbot> Someone could make this project's mistakes the subject of their thesis or something 20180930 21:14:04< sevu> I'm looking for a thesis, but … 20180930 21:23:01-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection] 20180930 21:23:07-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev 20180930 21:46:37-!- stikonas [~gentoo@wesnoth/translator/stikonas] has quit [Remote host closed the connection] 20180930 21:48:35-!- stikonas [~gentoo@wesnoth/translator/stikonas] has joined #wesnoth-dev 20180930 21:50:25< sevu> I tried to build wesnoth 1.8 … it crashes at 98% (where else), the error message indicates that it is soing sth it isn't allow to do in C++11 20180930 21:50:45< sevu> what was there before C++11? 20180930 21:51:06< sevu> In case you wonder, I thought it would be a good idea tio have old releases around to watch replays 20180930 21:53:12< sevu> c++03… a 8 year gap 20180930 21:54:08< celticminstrel> You can't even do it by explicitly specifying the language standard? 20180930 21:54:29< celticminstrel> Like -std=c++03 in gcc 20180930 21:55:13< celticminstrel> This isn't possible with Microsoft's compiler but should work with both gcc and clang, I would think... but I have no idea if it would solve that kind of crash. 20180930 21:55:22< celticminstrel> What is the error message anyway? 20180930 22:01:10< sevu> I try rebuild with that option now, I lost the error message already though 20180930 22:11:13-!- boucman [~rosen@wesnoth/developer/boucman] has quit [Remote host closed the connection] 20180930 22:36:31-!- gfgtdf [~Daniel@x4dbafc8f.dyn.telefonica.de] has joined #wesnoth-dev 20180930 22:45:57<+wesdiscordbot> celmin, @mattsc if there were areas where a simple syntax conversion were possible, I wouldn't be opposed to having an external tool to do so 20180930 23:03:30< sevu> celticminstrel, I was wrong about the c++11 thing, that's just a warning. it's lua.cpp with something about "scope" https://bpaste.net/show/b2afdb8f4b00 20180930 23:07:59-!- hyp3rbor3ax [~hyp3rbor3@p57A6ECDB.dip0.t-ipconnect.de] has joined #wesnoth-dev 20180930 23:14:05-!- zookeeper [~lmsnie@wesnoth/developer/zookeeper] has quit [Ping timeout: 268 seconds] 20180930 23:25:05-!- gfgtdf [~Daniel@x4dbafc8f.dyn.telefonica.de] has quit [Remote host closed the connection] 20180930 23:37:05-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has quit [Remote host closed the connection] 20180930 23:37:11-!- janebot [~Gambot@unaffiliated/gambit/bot/gambot] has joined #wesnoth-dev --- Log closed Mon Oct 01 00:00:17 2018