# I am the Watcher. I am your guide through this vast new twtiverse.
# 
# Usage:
#     https://watcher.sour.is/api/plain/users              View list of users and latest twt date.
#     https://watcher.sour.is/api/plain/twt                View all twts.
#     https://watcher.sour.is/api/plain/mentions?uri=:uri  View all mentions for uri.
#     https://watcher.sour.is/api/plain/conv/:hash         View all twts for a conversation subject.
# 
# Options:
#     uri     Filter to show a specific users twts.
#     offset  Start index for quey.
#     limit   Count of items to return (going back in time).
# 
# twt range = 1 6
# self = https://watcher.sour.is/conv/z2wiqfq
@prologic For all the twtxt archeologists out there, that is the historic moment on how it all began. Unfortunately, yarns doesn't seem to have [all (?) of prologic's archive feed](https://twtxt.net/user/prologic/twtxt.txt/2), but somehow mine, hence only my replies show up when I search for that conversation. Quite weird. Maybe user error, no idea. But it must have been after the twt hash change, so the hashes should match, as seen by the first link in this twt. Anyways, I reconstructed that conversation by hand using the archive feeds: https://lyse.isobeef.org/tmp/avatar-story.txt
Btw, @prologic, I reckon the yarnd generated archive feed URLs are not stable and thus kind of contradict the Archive Feeds Extension, which states:

> Once they are made public, [archive feeds] are supposed to be left alone and won’t receive further updates. Deletion or editing is still allowed, but feed authors should not expect clients to retrieve archived feeds on a regular basis (or at all).

Looking at your main feed URL https://twtxt.net/user/prologic/twtxt.txt, the prev points to https://twtxt.net/user/prologic/twtxt.txt/1 and that in turn links to https://twtxt.net/user/prologic/twtxt.txt/2. Here the chain ends. That means whenever a new archive feed is generated, the URLs change. Or viewed differently, the URL suddenly provides a differnt content. That probably causes clients to not update their caches properly, because they always see the same …/twtxt.txt/1 in the main feed's prev and think that nothing changed, which contradicts the promise of the spec.
@lyse Yeah as we discussed in the call, I _think_ I might switch this to use a UNIX timestamp instead with seconds precision. Because the feeds' metadata are dynamically generated I have to do handle feed rotation and linking with the prev field more programatically I guess πŸ˜…
@lyse Yeah as we discussed in the call, I _think_ I might switch this to use a UNIX timestamp instead with seconds precision. Because the feeds' metadata are dynamically generated I have to do handle feed rotation and linking with the prev field more programatically I guess πŸ˜…
@lyse Yeah as we discussed in the call, I _think_ I might switch this to use a UNIX timestamp instead with seconds precision. Because the feeds' metadata are dynamically generated I have to do handle feed rotation and linking with the prev field more programatically I guess πŸ˜…
@lyse Yeah as we discussed in the call, I _think_ I might switch this to use a UNIX timestamp instead with seconds precision. Because the feeds' metadata are dynamically generated I have to do handle feed rotation and linking with the prev field more programatically I guess πŸ˜