The more accessible spatial data collection becomes, the more important one question gets: what else are we collecting along with the environment? A phone recording a warehouse, shop, or street might capture faces, license plates, or other details that have nothing to do with 3D mapping. If every contributor has to manually clean footage before submitting it, scaling the network becomes harder. That's why the privacy side of @vangrid_io matters to me. Processing that blurs sensitive details on-device, before data leaves the phone, could reduce unnecessary exposure while making participation more practical. But I don't think blurring alone settles the issue. The resulting 3D reconstruction still needs to be useful, and the system has to balance privacy protection with enough geometric detail for downstream applications. Personally, I think that balance is harder than it sounds: remove too little and you expose people; remove too much and you risk degrading the data. For a distributed spatial network to scale, privacy can't be an afterthought. It has to be part of the capture pipeline itself.