pnasrat
commented
1 year ago Related I'm wondering if we can get the information of what communities are running through observability (logs, metrics, traces) or real user monitoring to be able to better identify patterns and components we might want to move into support. Obviously this can be paired with qualitative data from community engagement (surveys, interviews, discussion, participant observing, etc)
Our infrastructure is composed of many different moving pieces, layered in fun cake-like ways. However, this sometimes makes it difficult to figure out which parts we actively support and which parts we do not.
We should make a fairly explicit list of things (infrastructure services & software packages at least) that are 'our' (2i2c) responsibility, and things falling outside this are not directly supported by us.