javiervargas
commented
4 years ago Proposal
Research and document about Lightning Network's c-lightning and LND implementations taking into account the focus of the Learning Bitcoin (and Lightning) from the Command Line course and the focus of new developments. The research will delve into topics such as the ease of backups, the use of WatchTowers, API's for developers, the universe of related LAPPS and new ongoing developments such as LSAT in LND.
Investigate and document the best way to run a c-lightning node in terms of cost and storage since c-lightning requires a full node with txindex activated. C-lightning can work using a full remote node accessed by JSON-RPC. LND can work with Neutrino or a pruned node if the blocks are kept until March 2018 minimum. Document about the possibility of having both implementations on the same server using the same bitcoind server.
As this research phase has progressed we could detail more in the tasks.
J.
jodobear
watersnake1
ChristopherA
Fonta1n3
shannona
Several interns were interested in adding Lightning to Bitcoin Standup, and ultimately to FullyNoded 1 & FullyNoded 2. There was also interest by one intern in possibly some limited FullyNoded 1 like functionality to BitcoinStandup for Android, which could be a start toward a Fully Noded 2 for Android as well.
The biggest problem with this project is that it is too big and insufficient scoped to an internship project. In fact, it is so big if it was fully funded by our Patron's that I'd still have reservations about the large scope.
There are some sub-projects that might be good internship projects, but one of the first issues is which version of Lightning?
@shannona and I always planned to add C-Lighting to our Learning Bitcoin from the Command Line course. This version of Lightning we felt was well aligned with the goal of the course to teach how Bitcoin really worked, in ways that they could leverage people getting jobs in the industry. I still feel that C-Lightning is more the "industrial" and "cash register" version of Lightning for use in bigger projects. There are some advantages also that is is written in C++ and uses LibWally which are used by other Blockchain Commons projects.
However, I believe Lighting Labs lnd tends to be more of a leader in new approaches. In particular, there is something called LSATs that is of interest to a couple of Patron's of Blockchain Commons that could lead to real paid work from Blockchain Commons.
So I think the first step is to look at both of these implementations from the perspective of the command-line course, and from the perspective of advances like LSATs (for instance, maybe LSATs are on the C-Lightning roadmap), and post it to our Research repo.
It is also relatively trivial to fork one of the linux standup scripts to also install lightning. A challenge is that this require a full node (and likely tx-index). This is not inexpensive to host. So another project is to discover the cheapest way to host the bitcoin data on a VPS service separate from the VPS itself. For instance, on Linode a single VPS that can hold a full node is $160 a month, but that includes 32GB RAM and SSD which is overkill.
Once that is addressed in Bitcoin Standup linux scripts, then we can try out both Lightning implementations.
Other ideas? Again, we want to scope things such that the intern leading the project can complete something meaningful in 8-12 weeks.
-- Christopher Allen