rename user_id file

Telemetry/analytics are anonymised and can be disabled.

However for unification across Iterative’s stack (MLEM, TPI, CML etc.) it’s a good idea to put the config in one place. Basically move the existing $CONFDIR/dvc/user_id to $CONFDIR/iterative/telemetry. More explicitly:

import json
from shutil import copy
from pathlib import Path
from appdirs import user_config_dir

# DVC backwards-compatibility
old = Path(user_config_dir("dvc/user_id", "iterative"))
# cross-product path
new = Path(user_config_dir("iterative/telemetry", False))

uid = generate_id() # see above
if uid.lower() == "do-not-track":
    return
if new.exists():
    uid = new.read_text().strip()
else:
    if old.exits():
        uid = json.load(old.open())["user_id"]
    new.write_text(uid)

[omesser]

@skshetry To your concerns about transparency and user trust - we should make https://github.com/iterative/iterative-telemetry public (see: https://github.com/iterative/iterative-telemetry/issues/30), I don’t think that it warrants the duplication personally

Today this package doesn’t do much, but the idea is to unify the more important parts of telemetry in the future (not just user_id resolution and config file location). And we see that for every small decision we have 20 opinions, so it makes sense from a process stand point as well. I believe that our current divergence in various projects (MLEM, GTO, TPI, DVC, VSCODE) showed us that we need this centralized as part of a healthier process around telemetry that situates it as a cross-team effort

[efiop]

@mike0sv @casperdcl So what’s the status of telemetry package? Can we publish it and use it at least partically to retrieve the file location, etc? Seems better than having multiple projects do their own thing by hand. I can also contribute backward-compatibility code there to use old user_id as noted above (this will be useful for years, because some people stick to older dvc version for years).

EDIT: as noted by @omesser there is an issue for it already https://github.com/iterative/iterative-telemetry/issues/13