Hi, I'm the main developer of tlog.
First of all, thank you for taking this on, Will! Next, I'll be glad to answer any and all questions about tlog you might have. To address what already was mentioned: The main difference from other recorders is that the primary use for tlog is logging what's recorded to allow streaming and centralization using logging infrastructure. Primary suggested storage for now would be ElasticSearch, which would allow (limited) searching capability over the recordings. However, there shouldn't be a problem of using something else. The JSON schema is documented. BTW, tlog can playback from ElasticSearch directly. Yes, we have our own format. I looked through existing recorders and couldn't find any formats suitable for logging and streaming. We already support logging to Journal directly, while also adding a few extra Journal fields to simplify extracting particular recordings. That is used in the Cockpit UI we're working on (demo: https://youtu.be/YqNLJhNyqvY?t=6m38s). We need to document it though. I'll try to release the next version (v4) in January, before my talks at DevConf.cz and FOSDEM. I'll also try to do releases more often. Issues for the v4 release: https://github.com/Scribery/tlog/milestone/7 Nick P.S. I'm a Debian user since "Potato". You're doing a great job, guys!
