On the scalability issues of onboarding millions of LN mobile clients

The discussion revolves around the challenges of opportunistic incentivized local network peer for blockchain information when a node does not have direct internet connectivity. The concern is that an off-grid node's direct LN peers could collude to not forward the payment, which will make the money valueless. An alternative HTLC construct has been suggested to pay only for valid headers that hash to 32 B values committed by the HTLC. However, it requires all forwarding nodes, including the direct peer(s) of the light client, to know about it. It seems impractical for on-chain payments. Public watchtowers or HTTP proxy data cache similar to watchtower make sense as they would be expected to be economically motivated and LN payment-aware. Full nodes could potentially be incentivized to exchange new data with other nodes in a tit-for-tat way. There are still privacy concerns when serving generic/un-personalized headers/filters/blocks to light clients. Therefore, using Tor to access public-service nodes or issuing blind tokens can help to make the light client harder to locate. Personal, altruistic, or friends and family watchtower is also possible, but that may lead to the genesis of governments and states.