This article in the Ryan X. Charles Times identifies the problem that bitcoin miner compensation is based on economics not a fixed price, so getting a bitcoin transaction processed may cost more than the transaction is worth regardless of workarounds such as the lightning network (Mercator identified this more than a year ago):
“The lightning network is a network of payment channels between economic nodes on bitcoin or similar blockchains. The purpose of the lightning network is primarily two-fold: 1) To increase the number of peer-to-peer transactions that can occur for the same block size, and equivalently to decrease the average fee for a transaction, and 2) to allow for instant transactions.
Yours is a social content site like Medium but with a pay wall. In order to support small payments for content worth as little as one cent or less, we have developed a payment channel network similar to the lightning network. For the sake of our business, we estimated the cost of transactions on our network before launch. We demonstrate that payment channels on bitcoin funded with $44 or less are not economically viable if they are used predominantly for payments in one direction. In order to be competitive with Stripe and PayPal, channels on bitcoin need to be funded with more than $150. Payment channels that are used for many back-and-forth payments can be funded with only $44, but we predict channels of that sort will not be common soon. Our computations apply for our payment channel network, not the lightning network, but the differences are subtle and do not significantly alter the on-chain transaction fees, so the conclusions apply to both networks.
Because on-chain transaction fees for bitcoin are still rising, transactions on our payment channel network would probably be at least as expensive as traditional payment processors like Stripe and PayPal, although unlike them we have no lower bound to the size of the transaction value. It is remarkable that even with a micropayments layer on top of bitcoin, on-chain settlement fees on bitcoin are high enough today to make fees uncompetitive with traditional payments processors in typical cases. For this reason, we recently switched to litecoin. Litecoin is technically very similar to bitcoin, but currently has on-chain fees than are 100 times lower than bitcoin.”
The article finishes with the recognition that not everyone involved with bitcoin want it to be competitive with payment networks and other virtual currencies:
“In this article we first consider the simplified case of a single payment channel, then the case of a payment channel hub, and finally the general lightning network. We conclude with a market of analysis of bitcoin in the context of the one megabyte block size limit, altcoins, and Yours, and recommend that if the bitcoin community wishes to be competitive on a transaction fee basis with traditional payment processors and altcoins that the maximum block size should be increased at least 100-fold, and ideally 1000-fold. It’s worth noting that many bitcoin community members do not wish for bitcoin to be competitive on a transaction fee basis because bitcoin has other properties which are desirable to users, particularly the low cost of running a node.”
Overview by Tim Sloane, VP, Payments Innovation at Mercator Advisory Group
Read the full story here