MWSE, Lua and OpenMW compatibility

[psi29a]

viewtopic.php?f=6&t=5198&p=54877

Reflecting back on this, I'm wondering if Blue is a sockpuppet account. It certainly would be in the benefit of NullCascade to see OpenMW forked to be Windows only so that his work on MWSE wouldn't become irrelevant/obsolete. They seem to be both pointing out that the reason OpenMW isn't done yet is because the project is dragged down by being multi-platform and that the only real solution to this is for it be Windows only.

I for one, welcome a bit of competition.

[NullCascade]

viewtopic.php?f=6&t=5198&p=54877

Reflecting back on this, I'm wondering if Blue is a sockpuppet account. It certainly would be in the benefit of NullCascade to see OpenMW forked to be Windows only so that his work on MWSE wouldn't become irrelevant/obsolete. They seem to be both be pointing out that the reason OpenMW isn't done yet is because the project is dragged down by being multi-platform and that the only real solution to this is for it be Windows only.

I for one, welcome a bit of competition.

lol. God damn. Deleting my posts, and thinking I'd make alt accounts? I don't even agree with Blue on half that shit. If I were to remake Morrowind's engine it wouldn't be based on OpenMW, it'd be based on something like Unreal, Lumberyard, or Unity. Like 80% of the work in OpenMW is redundant, and the upgrade cycle would be sooooo much easier. We'd get free technology every year just by building off another engine. We wouldn't have to spend years implementing shadows. But that wouldn't be FOSS, which is what this project is about. I wish Blue well, but think that he is biting off more than he realizes. Maintaining a fork is not a trivial matter. Plenty of people have forked OpenMW. Does anyone use those forks? Not really, because just having a GitHub repo isn't going to get end users to use your software, and the choices he is making will make pulling improvements in OpenMW's main repo more difficult.

But, no. I know MWSE will become obsolete. There's an interview I did with DarkElfGuy that should air this weekend where I think I touch on the topic. My hope was that the project leads here could be convinced to not alienate many years of mods and modders from their platform, and to make modders deal with two conflicting scripting systems. But that's where we're headed. ~15-20% of mods released for last month's modathon are built using MWSE. That's almost 1 in 5 mods released last month that OpenMW users will never be able to use. Those will never work on OpenMW, but will instead have to be remade.

My priority isn't to ship my own tool, my priority is to empower developers/modders. That's what I do professionally, what I did in other games, and what I'm doing now with MWSE. I originally wanted to contribute to OpenMW, but the project was painfully catering to one operating system for developers. It took me less time to completely remake MWSE than to deal with its Linux-focused build environment, which, if I fixed, would never get merged -- because the fix is to do things platform-agnostic, rather than the Linux way. So after talking with people on the Morrowind Modding Discord, it became clear the best way to empower modders was to work on MWSE, not OpenMW. That's what the modders wanted, was viable in the short term, and allowed the biggest bang for my time investment. To those ends I've tried to make their lives easier by trying to improve OpenMW compatibility in MWSE 2.1, since OpenMW doesn't care about compatibility the other way around. I'm not mad because my "baby" will stop being used in 2091. I'm mad because of how many years of modding and how many mods are being thrown out the window when adding MWSE compatibility would be trivial.

But yeah, no. Blue is totally my secret alt to disturb you. That makes way more sense.

[psi29a]

This made me smile, thank you.

I honestly think if you could tone down the rhetoric and open hostility (seriously, software justice warriors?) then you might still be able to persuade Zini to support more of MWSE or that you could help lead that with your own merge requests. I mean, you can even help guide/shape what that means and even do the work yourself, if as you said, it is trivial to add.

Not everyone reacts the same to your style of communication and being aggressive doesn't help MWSE, nor the modding community's cause. Replying with also doesn't help anyone. I think there is a difference between being mad at Zini (OpenMW's future) and then posting on discord about the awesome things you can without OpenMW. It gives the impression that you're now hostile towards OpenMW and using anything that happens or effects OpenMW in anyway as a springboard to attack it. This is the perception you have given others, doesn't mean that it's true. I hope not, I'd rather see a compromise between MWSE and OpenMW and/or see you as helping OpenMW.

There is somehow a miscommunication that implied OpenMW was leaving GitHub which isn't going to happen, at least not in the foreseeable future. I've created a mirror on GL, but that's it. It gives those who prefer not to use GH a way to contribute if they want to. It doesn't detract from development on GH since those people wouldn't have been on there to begin with. That hasn't happened yet, that's fine.

If you want, I and others would be more than willing to help, in any way we can, to make development on OpenMW easier for you or any other developer. You just have to ask. I openly admit that I'm out of my depth when it comes to development on Windows, but there are others here who can help.

Believe it or not, I also agree that Lua, AngelScript or Python is a better way forward than sticking with and expanding mwscript. Then again, perhaps Zini is also planning on revamping mwscript to make it suck less so that there really isn't a need for the rest. That part isn't yet clear.

[lysol]

It took me less time to completely remake MWSE than to deal with its Linux-focused build environment, which, if I fixed, would never get merged -- because the fix is to do things platform-agnostic, rather than the Linux way.

This is a straw man's argument. OpenMW is multi-platform and has always been. I have already stated why the project is easier to build on Linux in my previous post. If you're willing to have a (sober) discussion about this, I bet many of your suggestions actually might get merged. But you've never even discussed them here, at least not while also not being disrespectful.

Why wouldn't the OpenMW project want good compatibility? It's just stupid to suggest this. Zini has his ideas for how the project should evolve, and I bet he would gladly accept your input on how stuff could be done, but then you need to be open for a friendly discussion about it. I'm sorry man, but you've always since the start come out as "Oh guys, what are you thinking?! You need to do it THIS way!". An attitude like that just makes people get defensive, and this is exactly how the situation is right now. We can't always agree 100 % with each other, but if we at least discuss with respect for each other, we might get a good compromise.

So again, no one wanted a linux vs windows situation. OpenMW has never been hostile towards windows.

[akortunov]

persuade Zini to support more of MWSE.

It would be great. In this case I'd even help to implement MWSE script commands.

[NullCascade]

I'll try not to derail this thread too. Here's my $0.02, and I probably won't be replying further. If you want to chat about it further with me, take it offline. Or, well, online, but not here. The people that want to deny that there are issues with the project will keep denying that there are issues. But I already offered. Zini wants to keep his plans secret for now, and has come out hard against some of the fundamental aspects that have made Elder Scrolls modding what they are. It just isn't worth doing without forking, and even though modders are asking me to fork the project, I don't think that's healthy for OpenMW or the community. We'd go from having 2 competing ways to mod to 2.5-3 competing ways to mod. How does that help modders? They're already pissed at having to reply to users saying that they can't make their mod work in OpenMW (even if it doesn't use MWSE), they don't want to deal with splitting support between 3 engines.

Zini has said no. Go back and visit the thread. I didn't start off pissy. It went:
> Someone asked about MWSE support.
> Zini said no, never.
> I suggested that if Zini was planning on Lua, we should totally make our APIs compatible to make modders sane.
> Zini said no, never.

I can't work with that. People have added MWSE and additional scripting support before. It never got merged, because Zini isn't interested in it. We want different things. He wants OpenMW to be the focus. I want modders to be the focus. He wants pull requests and forks and centralized features. I want a loose, open platform to let TES do what TES does best -- become anything its players wants it to be, without the need for merges and recompiles. Zini closed that door. I can't open it, he has to.

I love the idea of OpenMW, but it just... isn't there. I followed the project from its initial D days. I provided the first domain/web hosting. Then I left high school, went to college, moved around, got jobs in the gaming industry, etc. In that decade, OpenMW has offered nothing new to everyday modders. Every release I'll download it. Every release I run into mod incompatibilities, or general instability. With 0.43 (as well as tes3mp), it only works on 1 of the 3 devices I tried on. When I go to try to contribute I see a tangled mess that takes 5 seconds to get working on my Ubuntu install, and hours if not days to get working on Windows, and even then is a jumbled mess with no project organization in the IDE. Every year, for the past few years, I hear "we're so close to 1.0, it's probably our next release!" But it never comes. At some point OpenMW stopped being the exciting promise of the future in Discord, and instead something modders complained and memed about. I get things like "lul OpenMW #2090" and "Man OpenMW is still super wonky. I tested my house mod & it does not work with it. .. a simple house mod! The mannequins float up & down & fly away if you get close to them. The lighting is horrible & my teleport animations do not work. Wow I can't believe there are folks that think it is ready for prime time." and "i'm so disappointed in ZiniMW after reading this topic where you were talking about using luaJIT". Those aren't end users. Those are quotes from prolific Morrowind modders, what should be the backbone of the community.

I hoped that I could work on MWSE to empower modders this year, while working with OpenMW developers so that we come to meet one day, and everything just works in harmony. That was shot down, and my decade-long growing disappointment in the project transitioned to jaded assholery. Sorry for spilling that onto the forums. I'll of course continue to follow the project but at this point I'd rather spend my free time talking in Discord with modders, adding new features they want to MWSE to enable cool mods than being constrained by OpenMW's chosen direction. It's just not worth getting riled up arguing with people on the internet about issues that they refuse to believe in.

[AnyOldName3]

Zini has said no. Go back and visit the thread. I didn't start off pissy. It went:
> Someone asked about MWSE support.
> Zini said no, never.
> I suggested that if Zini was planning on Lua, we should totally make our APIs compatible to make modders sane.
> Zini said no, never.

He didn't say never at first. He said he didn't think it was a good idea and wasn't planning on it. He also never said Lua wouldn't be a thing, just that he'd rather focus on making MWScript not suck first and didn't want Lua to be able to call native code directly. You then got stroppy and he got defensive and then he took a firmer stance.

and even then is a jumbled mess with no project organization in the IDE.

You've said this several times but it's simply not the case. The folder structure is exactly the same both inside and outside of the IDE. Here's a screenshot showing /components/sceneutil/ being exactly the same in both places:

[lysol]

He didn't say never at first. He said he didn't think it was a good idea and wasn't planning on it. He also never said Lua wouldn't be a thing, just that he'd rather focus on making MWScript not suck first and didn't want Lua to be able to call native code directly. You then got stroppy and he got defensive and then he took a firmer stance.

I have to say this is exactly how I interpreted too. It might have been due to some misunderstanding somewhere, but then we need to straighten that out.

[Zini]

Fine. One clarification coming up:

* There are some aspects in MWSE that trigger my security alert. That makes these aspects a complete no go. This is the one point where I take a hard stance. I do not compromise on security and there is nothing more to discuss.
* Some parts of MWSE seem to be incompatible with current OpenMW or enhancements planned for the future. I don't remember what the problems are anymore. Its been a long time when I last looked at MWSE.
* Some other parts of MWSE would require large scale enhancements that go beyond the plan for OpenMW 1.0. Also a no-go for now, but may reconsider after 1.0.
* And finally there remains a subset of MWSE that could be added without causing us problems now or further down the line. I just don't know how useful such a subset would be. I don't have a firm opinion on this point.
* Lua is a completely different topic and I would prefer to discuss it another time (because that could be a lengthy discussion for which I really don't have the time right now)

[Chris]

Ninja'd.

Yeah, that. Since OpenMW's source is created from scratch, it does things differently internally than vanilla, which means replicating certain functionality MWSE exposes could be difficult. And since the source is fully open, it doesn't have to resort to the kinds of hacks and workarounds that may have been necessary for MWSE to provide certain functionality. But there's nothing to say someone couldn't add support for some MWSE functions that won't interfere or cause future issues. In either case, nothing will happen before 1.0 anyway, since we need to focus on getting vanilla stuff working correctly before looking at how it can be extended further. Once 1.0 is squared away, we can start collaborating on how to extend the engine and its modding capabilities.