I’m trying to set up the lbry API daemon for some development, as suggested by the documentation at https://lbry.tech/api/sdk. To start with, the majority of docker images published by lbry are missing any kind of documentation (for example lbry/lbrynet-tv, the most downloaded image). I then found the lbry/wallet-server image, the description for which shows it’s the lbry SDK, however running this only results in a connection problem - is your daemon running error, suggesting it’s documentation is wrong and it’s not the SDK daemon.
After a large amount of trial and error, I found the lbry/lbrynet image. This is lacking much useful documentation for running the image, but the sample command confirms it’s for the daemon. Running this results in a large number of on-demand fetching height log messages, however the image then exits.
As a new developer to the lbry platform, this all contributes to a feeling that lbry is not yet mature enough for any reasonable amount of work, and that a huge amount of time would be wasted attempting to get even the very basics up and running. Am I missing something obvious here?
I can tell you a couple things. 1) This forum is totally useless for getting any useful assistance. I have posted a few times here and the posts never get answered. So go to discord server instead. 2) While I love the idea of lbry I am starting to have a feeling of discontent. Documentation is terrible and even for things that I’d think would be interesting and fairly easy to discuss.
ie: If this is a true peer to peer network then how exactly does it work? How are videos consumed and from where? At what point does the P2P pickup from the central servers that we know are there?
I could go on for an hour on all the unanswered questions and am blown away at how difficult it is to get any answers at all. I love the idea and the platform but seriously, if new people have to hunt this hard for answers they are all going to leave and look elsewhere.
I tried the Discord server, all I could see was something to do with card numbers for a validation, and a discussion channel for a podcast.
While from a usability point of view lbry seems reasonable, from technical and usability standpoint it’s awful at best. The unfortunate thing is, that doesn’t seem to matter any more. The shittiest solution will win if it’s shiny enough for people to use. See also, Facebook data sales, or the fact that this page (containing 2 pieces of text) is 4.6MB and counting.
In 2 sentences you’ve described my biggest problem with LBRY and why I continue to use and support it. Given the amount of youtubers Odysee got to jump ship from Youtube or at least to set up a backup channel, I think it’s currently the best bet for making a dent in the current social media monopoly.
Regarding the documentation, I’m still in the “be the change you want to see” phase, and I set up an Odysee blog where I aim to address some of the shortcommings in the official docs (I’ve been on a break the last couple weeks, but I’ll be updating it soon). It sounds like this article in particular would have been useful to you before you jumped in, but maybe you’ll still get something out of it: