Friday Facts #391 - 2023 recap

[Hares]

The solution was simple: In 2.0, I added a new Manage mods button that allows full access to the mod manager GUI, including all mod portal functionality. You can see it in the screenshot above.

can I ask the same thing but for mod settings?

This one is a bit trickier because mod settings rely on mods being fully loaded past the settings stage (because not only mods can define their settings, they can modify other mods' settings), and fast-reloading is not an option there.

OMG does this mean there will be a factorio 3.0 ?!?!

(or 2.1)

[Wechselstrom]

I've always been too lazy to try out mods at all, despite how easy it was even before. I might have to give it a go!

[thedoh]

Starting with 2.0, we will ship the game with multiple built-in mods

I, for one, welcome the new mod features. I hope there's a toggle for "base game" (no mods) and "base game + space age" since there's going to be the "1.1+ experience" and the "2.0+ experience" with mixed and matched mods.

And then we get blown away with...

2.0 features an all-new explore tab that does just that.

yes what a great feature -- and nice looking, too!

Quite interesting to see Golang getting some love at Wube. Was it only for prototyping, or is there now Golang in the tech stack? Go is Great!

[AileTheAlien]

Quite interesting to see Golang getting some love at Wube. Was it only for prototyping, or is there now Golang in the tech stack? Go is Great!

I'm also curious, since it should be a pretty good language for all the non-game server stuff. (It's garbage-collected, plus probably some other stuff I'm forgetting, so your time between game-updates wouldn't be consistent, which is a big problem for Factorio since it's tick-accurate / synced across players in multiplayer.) I haven't used C++ professionally, but used it back in college - the extra boilerplate and verbose syntax made everything slower to read than say Java, or recently Typescript or Golang. Looking at this "modern C++" site, it seems like classes / objects / etc are still quite verbose. ^^;

[Kadet123]

Glad to see mod management updates, good work.

Speaking of 2.0 and mods, are there any plans closer to time of release to publish updates to the Lua APIs for mod authors to look at before launch? So they can work on updating their mods before Space Ages's release?

Or having a demo version of the 2.0 base game engine (without the Space Age mod) available before Space Age?

[MaxAstro]

I would like to second the request for "mod groups".

The biggest pain point in setting up mods for me right now is manually switching between i.e. my Krastorio setup and my Space Exploration setup, which often have different secondary mods also because different mods complement each other differently...

I have desperately wished for a long time for some way to "save all active mods as a group" and then "activate all mods in a saved group".

Currently you can kinda do this in a janky way by creating a save game with the set of mods you want and then loading that save just to change your active mods, but that's really janky.

[ManaUser]

Just in case this is a good time to mention it, my #1 wish for the mod manager is an easy way to update currently enabled mods only. Over the years I've downloaded a lot of mods and "mod packs", but of course I only use a limited set of them at any one time. e.g. maybe right now I'm running Warptorio2, the extra planets for that and a few QoL mods. I'd like to have quick way to update only these (if they need it) without manually finding them on the Update tab or updating everything, which would include Space Exploration, Seablock, etc. etc. Updating mods I'm not currently using is mostly a waste of time (and Wube's bandwidth) since they'll probably need updating again by the next time I actually want to play them.

[mmmPI]

The second chart shows all the 2023 player downloads grouped by mod category. It confirms the cliché that people are only here for the Content and I should've prepared these chart for mod tags instead.

To me there are Qol and content, once you have a set of Quality of life mods that adress your wishes, then you don't "need"/"want" to dig further, it's only content then. But it doesn't mean i'm not interested in Qol or tweaks, just that those are somewhat exclusive or overlapping, say mods to plan your factory, or to browse receipes you are less likely to need 3 of them at the same time, or to change after you got familiar with one. Whereas you could play with different content mods, say changing 3 times, for 3 games, but all 3 games would also have the same Qol mod. That's more download for content because it changes everytime.

Occasionnaly some "content" mod makes it so that some additionnal tweaks are wished, but those are bound to be low download, when they are tweak of a content mod since only a subset of the player could be potentially interested. The "bigger" the overhaul, the more numerous and diverse the tweaks it generate. And this create a ratio of "content with high download" and its attached "tweak with limited download". Making it always in favor of the "content" mod.

Does that mean people are not interested in Qol ? I have another reserve, the download count is an average, maybe 90% of player download 1 Qol mod, and the 10% left download all the content mods. That would be "most people download tweaks or Qol". But "most downloaded" is "content". You could survey players or ask if some of them allow to have their download count public like opt-in, so you could know if there are such profile of download per user which would act similar as poor-mini pc downloading MK3 armor over and over.

This year we continued our quest of improving discoverability . First we added the automagical Highlights page, every week it tries to be smart about picking an interesting selection of mods. It works okay but is still in a "experimental" stage where we haven't decided where to take it next.

Well the mod portal doesn't have a good way to improve its selection because i think it doesn't have feedback on its previous recommandations. It should be possible to activate the webcam of players to see if they are enjoying the recommended mods, or to listen to their microphone, to hear if they laugh when playing maybe make a profile of different players, by spying on their computer and looking for the other game or software installed, so that an AI could recommend different mods to some players based on their profile looking similar or not to other players. Ever heard of player retention ? You could measure the time played by players of similar profile with different recommendations, so as to know which recommandation makes players plays more or are more likely to be downloaded

On the other hand, it already takes some times to get used to all the tick boxes, and it does indeed works okay when their meaning is deciphered. sorry for the dystopian ideas, i've been doing too much youtube recently and i'd hate to see a mod "factorio but easier " recommended to me, just because i downloaded one to make space exploration faster ? comon at least recommend me the one called " factorio for busy player that has limited time to chill " that's a smart algorithm !

What i 'dislike' at the moment is that sometimes several mods with the exact same image are recommended at the same time and although it would make sense because the mods can be differents the image is just the cover, it makes duplicate information on the page and for discoverability, it makes the page like a tree with similar branch, or less branch.

Sometimes i feel like it could also be nice to highlight authors, not just mods, but it's difficult to make a fair algo in such case and it being unfair is detrimental. Thinking some mods aren't updated because they are not maintained, sometimes i recommend a mod i played with, but i usually don't when i encountered a bug, i report it, after some times it becomes obvious that some authors are very active and try to advertise their creations and incorporate feedbacks and some made a mod once for themselves which everyone is free to use but that's about it sorry for the simplistic nature of the example taken as illustration that's the profile thing from marketing. I think it's a good opportunity for discovering new mods presented in a consistant format, with the same standards, that are sometimes meant to work together when browsing someone's mod and it's easily overlooked when those are not super shiny but a set of Qol and tweaks. I suppose some "players profile" could match some "modder profile", not sure how obvious this is for everyone, it took me some times to realize and it was a pleasant experience to use someone's set of tweaks for a game with thoughts put into it, rather than trying to reinvent the wheel everytime quickly before launching a game or taking long time to prepare a modpack.

Picture this: You have just configured your mod list for a new playthrough of Krastorio 2, and you start a new game. Unbeknownst to you, you accidentally enabled the Power Armor MK3 mod, which Krastorio 2 has marked as incompatible. It takes until an hour into the playthrough for you to realize that the Krastorio 2 content is missing, and now you have to throw out the save and start over from scratch.

That happened to me but it was not just 1 hour in the game so instead i cheated hahaha. Happy to hear it is adressed, because i wasn't happy at the time.

Also the fact that mods have versions, and you can have several versions downloaded but not active ?, and you can naively update the mod, and launch a game, and unbeknownst to you, the mod's version is not set to the latest. ( and a mod is then not compatible and is silently not being active... same end result ). Maybe i made a mistake, i don't want the recommandations to suggest i need mod to make the game easier, i am more tempted to click on the very big overhaul modpack but then there is this marketing thing where everytime you have to make a decision before purchasing, everytime a % of potential customer will make the decision that bring them away from the purchase. That apply to me when choosing mods, everytime there is a incompatibility i missed, or i need to reload the game, and change a setting, and reload the game there is the risk i would get distracted instead of being playing some factorio. I think it's the one of the main critics people who don't like mods do to them : the risk of technical complications, a bad experience (even from other games) can damage the will to do some discovery. ( the other being cheating/betraying the original game ).
Very happy to read this part was given some attention !

I have settled into a workflow of searching for a mod on the website and only searching for it in-game after I have found the mod I need. Wouldn't it be nice if the game had the same experience, with all of the same searching and filtering capabilities?

It would be nice but what about the pictures ? because that's mostly the kid inside me that feel like in a toy shop where i can select any mod that you are trying to convince to use some logic filter before playing the game ? Good luck with that Really appreciating the efforts to make the search match i'm relying on that I will use the bookmarks and anything that would make managing a long list of mods in which only a few are active for several game easier, makes me curious to see what's the rest of the bucket of laundry list, i will have to stay tuned , should we expect a 2024 roadmap ?

[FuryoftheStars]

Great to hear about the mod manager improvements! Keep it up!

--------------------------------------------------------

OMG does this mean there will be a factorio 3.0 ?!?!

Um, no...? This means there will be 2.x.x versions....

--------------------------------------------------------

since there's going to be the "1.1+ experience" and the "2.0+ experience" with mixed and matched mods.

I don't think there's going to be a "1.1+ experience" except for those that downgrade/don't update. 2.0 in and of itself is a free update they are releasing just like every other update. It's the 2.0 mods (aka Expansion) that will need to be paid for.

So, for the most part, we'll essentially have a "plain" or Expansionless 2.0+ experience and a 2.0+ mixed and matched with Expansion mods experience.

[plepper1]

Thank you all for another year of amazing work on what was already a fantastic game! Here's to 2024, 2.0 and admiring your hard work hands-on!

[aka13]

Compared to other sandboxes, I think that factorios modding is already uncomparable good! Cheers to it getting even better.

[Karamel]

Please add mod grouping to the mod explorer. For example, when I want to play Space Exploration, I have to browse through the huge list of mods which I have installed and one by one select the ones I use for my SE map. It would be nice and easy to have a SE group, to which I would put all the SE mods, K2 group and then group containing the two named K2SE so I can turn the mod lists all at once.

Related to this, if a mod depends on another mod, why do both need to be explicitly enabled? It should be enough to enable, say, Industrial Revolution 3, since Factorio knows that it needs Industrial Revolution 3: Assets 1 to function. Of course this dependency should be transitive, so that if an explicitly enabled mod requires other mods and they require yet others everything gets loaded. Since Factorio already traces the dependency graph, this should be easy enough to implement.

This would then allow mods to be marked as libraries and removed from cluttering the main list altogether. It would also all mods which do nothing except depend on other mods to facilitate easy installation, such as Sea Block, to both do a better work - just enable or disable Sea Block itself and let it worry about the rest - and be properly marked as mod packs for potentially putting at top of list.

Mod groups could then be implemented as auto-generated mods which simply depend on all of the mods in the group.

[Pockets]

I like the new mod interface ingame.

However, if you like to play different saves with different modpacks, there is no easy way out of the box to switch one mod configuration to another.

It would be nice if, when you open Factorio, instead of start loading the game, there is a small window with a list of "mod configurations", so you can add entries to this list and each would be a different mod combination. Then you select one and the game starts with it.

I have a workaround for this but it is a bit tedious to configure:

[KeithFromCanada]

Thanks for the new info! Some thoughts:

1. When updating/managing mods, I don't always want to play the game after I'm done, so I would like to see 2 confirmation buttons on the in-game mods page: 'Confirm & Restart' and 'Confirm & Shut Down'. (Yes, I can--and do--press <ALT-F4> while the game is restarting if I'm done playing, but not everyone knows that they can do so.)
2. It occurred to me that setting the system to only use one (hardcoded) mod repository is an unnecessary limitation. What if someone also wants to set up a separate repository with their own private set of mods and use both in-game? ...or set up a backup repository on a LAN to hold mods for hundreds of game servers that only falls back to the main repository if something isn't in the local one, to reduce external traffic? (This would be the same sort of setup Linux distros use.)

Does that sound reasonable?

[KeithFromCanada]

I like the new mod interface ingame.

However, if you like to play different saves with different modpacks, there is no easy way out of the box to switch one mod configuration to another.

It would be nice if, when you open Factorio, instead of start loading the game, there is a small window with a list of "mod configurations", so you can add entries to this list and each would be a different mod combination. Then you select one and the game starts with it.

I have a workaround for this but it is a bit tedious to configure:

I have the same issue and fully agree. To make things much easier, I create taskbar shortcuts that jump me straight into the save for my current run + any multiplayer servers I play on. Here is a sample 'Target' string for the shortcut:

Code: Select all

"C:\Program Files (x86)\Steam\steam.exe" -applaunch 427520 --mod-directory %APPDATA%\Factorio\mods_20230828-ff --load-game 20230828-ff-e

(Notice that I set up a separate mod folder for each run/server, so no mods have to be swapped. Also, you can set a custom icon for the shortcut right from Factorio's .exe file.)

Cheers!

[KeithFromCanada]

Please add mod grouping to the mod explorer. For example, when I want to play Space Exploration, I have to browse through the huge list of mods which I have installed and one by one select the ones I use for my SE map. It would be nice and easy to have a SE group, to which I would put all the SE mods, K2 group and then group containing the two named K2SE so I can turn the mod lists all at once.

Related to this, if a mod depends on another mod, why do both need to be explicitly enabled? It should be enough to enable, say, Industrial Revolution 3, since Factorio knows that it needs Industrial Revolution 3: Assets 1 to function. Of course this dependency should be transitive, so that if an explicitly enabled mod requires other mods and they require yet others everything gets loaded. Since Factorio already traces the dependency graph, this should be easy enough to implement.

This would then allow mods to be marked as libraries and removed from cluttering the main list altogether. It would also all mods which do nothing except depend on other mods to facilitate easy installation, such as Sea Block, to both do a better work - just enable or disable Sea Block itself and let it worry about the rest - and be properly marked as mod packs for potentially putting at top of list.

Mod groups could then be implemented as auto-generated mods which simply depend on all of the mods in the group.

Even handier would be to be able to expand/collapse dependency hierarchies right in the 'Explore' list with full descriptions/etc., so players could quickly and easily decide for themselves which optional dependencies they might want to include.

[raiguard]

@raiguard, would Factorio CLI also be improved? It would be nice to have FactorioModManager functionality bundled, as well as have options to check/download/install Factorio updates from CLI.

It is on my laundry list to make the mod management logic available via CLI. It probably won't make it into 2.0 though.

[raiguard]

Quite interesting to see Golang getting some love at Wube. Was it only for prototyping, or is there now Golang in the tech stack? Go is Great!

I'm also curious, since it should be a pretty good language for all the non-game server stuff. (It's garbage-collected, plus probably some other stuff I'm forgetting, so your time between game-updates wouldn't be consistent, which is a big problem for Factorio since it's tick-accurate / synced across players in multiplayer.) I haven't used C++ professionally, but used it back in college - the extra boilerplate and verbose syntax made everything slower to read than say Java, or recently Typescript or Golang. Looking at this "modern C++" site, it seems like classes / objects / etc are still quite verbose. ^^;

Unfortunately Go is not used in our stack - all of the web stuff is in Python. If I was at Wube from the beginning and if Go had existed back then I would have chosen it. It's simply phenomenal for web stuff. That said, the lack of sum types is endlessly frustrating.

[raiguard]

Glad to see mod management updates, good work.

Speaking of 2.0 and mods, are there any plans closer to time of release to publish updates to the Lua APIs for mod authors to look at before launch? So they can work on updating their mods before Space Ages's release?

Or having a demo version of the 2.0 base game engine (without the Space Age mod) available before Space Age?

We will have more details on all of the Lua API changes closer to release. We are not discussing this now because things are still in flux and could change.

[Svip]

Unfortunately Go is not used in our stack - all of the web stuff is in Python. If I was at Wube from the beginning and if Go had existed back then I would have chosen it. It's simply phenomenal for web stuff. That said, the lack of sum types is endlessly frustrating.

As someone who maintains a large Go codebase, the lack of sum types isn't as big of a deal, once you've learnt to "program around them". Sum types are - at least in imperative languages - a relatively new feature. Honestly, I didn't find the lack of generics frustrating either, but now they are there. If Go adds sum types, it would also be a welcome addition. But it is getting harder to keep the Go 1 compatibility promise in such cases. Though their loop iteration backwards compatibility solution could bode well for potentially breaking changes in the future, so who knows. But I love that Go - as a language - is relatively small; it's possible to keep the entire grammar in your head.