I mean, I’m working on something now too based on my “IMO” post above.
It’s distracting me heavily from the phone I’m trying to get done. 
| 
I’m calling it Frame’s.
One for each language currently supported, and will grow as FiveM does.
That’s about all I’ll let out for now.
[L-Frame] | [J-Frame] | [C-Frame] etc etc…
I’d love to see a framework personally that is lean on servers, saving resources, has some good staff and moderation commands, something that isn’t esx, is modular and easy to program new stuff into it and has a jobs like system, almost like gmod dark rp except way more feature complete and powerful in many areas that dark rp lacked. But that’s just me.
I know this topic is called “Roleplay Frameworks”, but it would be cool if the base of the framework can be used outside RP too.
Like the base of the framework is just plain simple framework with basic functionality for example for data managment, than people can add our modules which adds more “roleplay” functionality
I think the job system shouldn’t be in the framework itself but be another module for the framework so people can decide if they want it or not without having useless code
Yeah that could be a thing for sure. But definitely there should be a base that can be modular like the framework I listed earlier.
I have spent the last 3 or so months poking and proding at ESX. I have been programming for somewhere on a dozen years in various languages (Java, JS, C++, a dozen or so proprietary ones, and a few others). I dont like lua, but ESX while not great, isn’t the worst. I give the people that did it credit, but it just doesnt work right at times. It doesnt do what I want it to do. I’m trying to optimize resources around a clunky framework. I want something newer, better documented, maybe doesnt use mysql, lightweight, better documented, doesnt make you want to claw your eyeballs out, and did i mention well documented.
Clean UI Elements, Built for varying levels of RP (from the jokes to the serious ones), a decent and customizable phone, integrated syncing, and exploit protection. This will sell me easy.
And as much as I would love a great RP framework, a good non rp framwork would be incredible. If someone does this I would recommend a very module based framework. I would love to see what people would make.
I’ve been planning the data save stuff lately, and I’ve made a few prototypes that will save character data for multiple characters etc etc.
Trying to make it backend agnostic so people can implement a “data catcher” for Redis or SQLite if they wanted to.
Limbo is billed as an “RP Framework”, but nothing will stop anyone for using it for non-RP stuff. I mean, joining a “group” to get access to certain things is also at the core of a team deathmatch server or a CTF server, right? “Money”, with a lick of paint, becomes “Points”.
At the end of the day, what I aim to provide is just a toolkit to abstract away a bunch of the boring work so anyone wanting to make something can get straight to the cool stuff without worrying about implementing “get the model of the vehicle the player is in, if any” in 5 different resources.
Things like implementing a garage or apartments or jobs is not the area of operation for the framework. The framework should simply provide the most commonly written and rewritten functionality as a common base, and then you can put whatever window dressing on that (calling your team a “job” or a “gang” or whatever you want) on that using a resource.
That said, I will of course be working on the most commonly used resources and release that, too. Car dealer, police jorb, ambulance jorb, parking garage, etc etc etc. That is secondary to working on the actual core, though, and I want that flowing like a [inappropriate] in an [inappropriate] before I move on to any of the other stuff.
Rugged, well-tested functions that provide useful tools, like octree and easy data storage. That’s what’s needed first, not garages before there is even a character editor.
I mean… FiveM is the framework… I don’t understand the problem honestly. If you want something, add it. Anything you can do with a framework, you can do without. Albeit you’ll need to find natives but still…
The point of adding another layer on top of FiveM is so that you don’t have to replicate code again and again in different resources. This makes writing resources easier and faster, and cuts down on maintenance complexity.
It’s also a handy way to have resources interoperate.
FiveM, by design, does not have concepts like “player apartment”, “parking”, “character customization”, “persistent player location” or “money” in it. That’s up to the different resources. Frameworks help bring that sort of thing in so your resource can focus on what is unique to it.
Why don’t we start a github repository that is public along with a Dcord guild and a collective of us can come together to create a new up-to-date framework?
Scratch part of the above. I took the liberty of creating them anyway:
GitHub: https://github.com/Starystars67/Framework-V
.gg/ACdFEGU
Limbo already has both, but if you want to compete that’s no problem for me.
I think pooling our efforts is a better plan, but you might not agree with what is already done for Limbo, and I can respect that.
Best of luck.
I don’t have a problem with esx, but it needs to be re-written and updated to fix it’s flaws. You could get esx and just modify it alot, or change the names of the events and such to pretend like it’s a custom framework, but it’s still esx.
What about some of the fundamental design flaws?
Lots and lots of problems with ESX come with the message Attempt to index nil value (upvalue xPlayer)
You’d have to track down a few race conditions, add a bunch of error handling, rewrite some of the core logic. In short, I think it’s more work than it’s worth.
The only upside is that it might end up compatible with existing ESX scripts. With the vast majority of the public ESX scripts being of questionable quality, however, I don’t really know if this is a good thing. Competent developers that have written good ESX scripts can probably port to a new framework very easily, so we’re only really hurting the newbies.
Hurting newbies is a bad thing, obviously, but I think it might be worth it to get rid of some of the absolute crap that is very common at the moment. How many servers run that old version of esx_policejob that hardly functions and just barely scrapes under the limit for a resource warning? Probably thousands of servers!
ESX has had it’s run. Now it’s time to replace it with something built on what we learned from using ESX.
ESX is good, but yeah. Needs something to replace it, maybe some sort of like esx v2 or something. With previous bugs fixed and improved performance, and all that.
I’m really surprised no one has mentioned NFive in this thread.
Because it’s utter garbage which tries to circumvent as many built-in systems as possible for no reason other than that the author doesn’t like working with the community to improve said built-in systems.
Tell me how you really feel @stannum lol. So, it’s garbage in your opinion because it doesn’t use said built-in systems? But at the same time insult the author for not working with the community to improve said build-in systems… I mean it is an open source project, so the Collective, could use the code if it is an improvement over whats currently in FiveM.
I’m just trying to figure out why the hostility. It seems like you have some personal issues with the guy or something.
no, we could not, since the ‘built-in systems’ are generally written in C++ (and not C#), and replacing said systems entirely would 100% break compatibility with resources that do use them. also licensing concerns are a thing.
that’s why ‘improving’ was explicitly stated. replacing entirely with something incompatible is the opposite of ‘improving’.
No, we have issues with anyone who decides they can do ‘better’ and then rewrites a whole ton of systems for their own release, without even asking if they can contribute to the core project let alone stating what their issues are with said built-in systems so that both the core project can be better (and not only their own release) and their releases can interoperate with any other releases.
The same goes for the author of this ‘txAdmin’ shit, who when asked if he wanted to contribute his code to the core project (before we resumed work on built-in webadmin) responded with nothing but ‘contribute to the project? you high?’, and the vRP author who went ‘hey, there’s a bug in this, but I won’t say what the bug is or work around it in my resource, I rewrote the entire component and the bug doesn’t happen’, and when we fixed the bug just went ‘yeah no I’ll still force people to use my rewrite as 
’ even though his rewrite has yet another bug added to it compared to the official fix.
I’m not trying to argue but oddly enough I know about both those situations. And I remember imagic putting in a PR and it was rejected dont remember why off the top of my head.
As for txadmin I know tabarra did say he would work with the collective but you guys refused the way he had it setup already.
I’m not trying to be on any side as it doesn’t benefit anything.
I was just suggesting something that wasn’t mentioned already that seems to be a very viable option.
I didnt know the backstory of NFive and such but to each their own I suppose and as long as it’s making the platform grow than I’m all for it.
The platform is open source so I’m glad its left up to the server owners to decide what can be run for now.
Maybe I’m missing something, but NFive seems to mostly re-implement things that already exist?
It has some database/storage interaction that seems novel, but other than that it looks a lot like what the C# API already in FiveM already does.
Granted, of course, that my cursory glance at it might have missed something like 90% of what it provides.