> I suspect that if this were to start gaining traction there would be a way to flag Nexus with an "I downloaded this" flag associated with the users' account. I certainly hope not! Exposing any programmatic way to increase download counts for files sounds like it's ripe for exploitation, no matter how carefully the tracking would be done. That's not an unsolvable problem, but it's a pretty big one given the problem it's solving is, you know, paying authors for their work. > Licenses etc. Your points are all very sound, and make sense for someone who is used to working with open source code. Mod files, however, don't follow the same rules. The reason lots of mods don't include licenses is because lots of mods don't *have* licenses. Mods don't follow the same standardised method of licensing assets that software does so all the conveniences of things like SPDX lists just don't exist. You're right that authors *could* post IPFS identifiers but you're now back to the problem of having to parse some arbitrary text fields (the only API-exposed text fields per-file are the name, description and changelog), which is far from ideal. > code vs non-code mods Yeah, that ones a common misconception that also tripped me up early on. For an example, the most recent game I've been modding was Ace Combat 7. It's skin mods are literally image files, packed in a proprietary format and installed as-is. Mod creators never see a line of code, and their only real option for effective licensing would be the CC family, which is definitely not what most of them would choose. --- As for why I don't allow external uploads: the primary reason is support. Every time I get a user reporting something wrong with one of my tools, I want to know that those files came from a known source that I trust. While hashes can help with this, I don't want to have to ask users to start checksumming files just so I know that no-one has messed with the files or that they're not hilariously out-of-date (honestly the bigger concern of the two). It also lets me more effectively control the impact/reach if I accidentally introduce a critical bug in an update. If someone wants to compile it themselves, they a) have some idea of what they're getting into and b) should understand that I can't necessarily help them.