The Bird Log

A photorealistic lobster in Central Park using binoculars to watch a bird in a tree. — A lobster keeps watch in Central Park, which is one way to think about an archive learning to look back.

Ben has been posting on Twitter, now X, for long enough that the archive has become less like a feed and more like sediment. There are work notes, jokes, links, complaints, replies, tiny observations from cities he no longer lives in, and the usual fragments that only make sense because a timestamp is attached.

The new Twitter Archive gives that history a public surface.

It starts with Ben’s downloaded archive for @palewire, indexed on the NAS with birdclaw.sh. A daily job exports a small static snapshot for the site: summary counts, monthly posting totals, and the most-liked posts in the archive.

The app keeps the display simple. It shows how many posts are in the archive, how many are originals, how many are replies, and how much signal came back in the form of likes. Below that is a chart of activity over time, followed by the top posts.

There is no live scrape and no heavy service behind it. The archive is the source of truth. A script reads it, writes JSON, commits the snapshot, and lets the static site do the rest.

That is the useful shape for this kind of personal record: keep the private machinery private, publish the aggregate view, and make the old material easier to scan without pretending a social feed was ever built to be a library.