I’ve had a few workshop items on Steam for the last six or so years, and if I have learned anything, it’s that people sadly don’t read, and they don’t understand either. I couldn’t count how often I was asked something that was stated in the very first line of the description. After a while, I just stopped answering, which isn’t a nice thing to do I believe, but I just couldn’t be arsed anymore.
No disagreement about the difficulty - like I’ve written in the original post or so, Stonehearth still has a very long way to go. It was fun dabbling around when it was much smaller, simpler and could be overseen much easier, but right now, with all the hidden traps and pits that are in the game, I wouldn’t want to mod it anymroe. The time has passed, and I’m not sure if even a proper documentation could make up for it by now.
It’s a difficult choice regarding modders vs. users. I would argue that you can expect from a modder that things are a bit more rough, as they ought to be a bit more resilient to bumps in the road. Users on the other hand - and this isn’t meant as an insult - are mostly the last tech savy people you could imagine, so catering to them is more necessary. From an UX experience, Steam is probably as easy as you can get - which also means that some features (such as branches) are not available. Either because they wouldn’t be used, or would be too hard to understand - it’s certainly not a storage issue, because I’m pretty sure Steam keeps copies of everything you’ve uploaded nicely archived.
Ideally, it would be nice for both parts. From my experience with other games though, I can say that pure download portals survive a lot longer than communities catering to developers - probably because it’s such a volatile environment to be in. Just the jump from SVN to git, for example, probably killed quite a few sites.
What I can tell you here is that out of all platforms I’ve used, Steam - probably because it is a walled garden to some extent - is absolutely amazing for developers. I have stats and graphs that I absolutely adore. Thanks to these, I know that over 231k people are currently subscribed to one of my items, and 507k have been in total, as example. It’s quite a weird feeling to know that something I made has been downloaded by more accounts than there are people living in my vicinity.
At the same time though, various games that were available over multiple distribution channels have either stopped supporting non-Steam versions, or are not developing new content for it. It’s heading in both directions, I guess and the next few years will be very interesting in that regard - especially since VR is opening up and everyone and their mother is starting a new store for that.
I’ve never said that it should be one or the other. I think that it’s important to support manual installation of mods (which then kind of implies 3rd party applications as well). However, in terms of integration - unless Riot provides some sort of API for modders to use, 3rd party tools will - in terms of convenience for your average Joe - always be worse.
I’ve actually proposed an API, or rather requested a few methods, that would make such an integration possible in the past, but that was denied. The reasons back then weren’t satisfying at all, but basically I could imagine attaching any kind of mod manager if the SH API would offer the following:
- A HTTP API to make HTTP calls, including setting verbs and headers as well as content (text/json and binary blobs)
- A configurable whitelist of domains that HTTP calls are allowed to. This would avoid building weird stuff with it. A user would have to configure, for example, that calls to
https://stonehearth.mysupernexus.io
are allowed. - A method to get the binary content of .smods, and one to write .smods. Writing can have a validation (e.g. “it’s a valid zip file”, “it contains a manifest.json”, “it does not contain any executables”) to avoid writing arbitrary files
- Something to restart the game/reload the mods/activate or deactivate mods. I think this is already handled with the built-in modding manager though.
Given these APIs, I could very well imagine writing an auto-mod-updater that works with a proprietary service and is not tied to Steam. Plus, everyone could write one (or an adapter to an existing one) if they pleased. HTTP calls would allow for quite amazing experiences too, especially if we could get Websockets to run - but now I’m being greedy. This is a pipe dream from the get-go.