Now that the software is running with (at least for me) a low level of jank, it seems worth considering what we do with the years of accumulated sneer-strata over at the old place. Just speaking for myself, I think it would be nice if we had a static-site backup of the whole shindig. Unfortunately, since I’m a physicist by trade, anything I do with webstuff tends to involve starting from scratch with compass, straightedge and wget. There’s got to be a better method of archiving.

The other, not-mutually-exclusive option I can think of is to manually rerun “SneerClub classics”, the posts that one way or another helped define what sneering is all about.

N.B. Some of the test posts made today involved writing more on the serious-discussion side and have accordingly been marked NSFW.

  • @blakestaceyOPMA
    link
    English
    21 year ago

    Anyone know how to manipulate compressed JSON (.zst) files? I was able to snarf the SneerClub data from a torrent there that goes up to December 2022.

    • @blakestaceyOPMA
      link
      English
      31 year ago

      Wake up babe, .zst files just dropped: comments and submissions.

      To get the data from this year, I tried the straightforward thing of just doing save-full-webpage in Firefox for each post. This was tedious, but I didn’t feel like figuring out how to get any automated downloading tool to work with my login details so that it could grab the NSFW posts. The result is ~2 gigs, most of which is probably redundant infrastructure. An oddity: trying to save a thread always failed on the first attempt but worked when I clicked “retry download”.

          • @selfMA
            link
            English
            21 year ago

            some work in progress on this is available here. the SneerClub directory is the output of the bulk downloader for all 1000 (deduplicated) posts it could grab from each of SneerClub’s hot, top, new, rising, and controversial tabs, and the jsonl files are just the ones you posted decompressed for convenience. so far I’m just using jq to process the data sets

            SneerClub has 1940 posts with nested comments and attached media where the downloader could parse it; the archive team files have 3851 posts and 100149 comments in a (much less convenient) flattened format without media. both sets have a few posts from 2015, so I’ll need to do more looking to see how much we’ve salvaged overall

    • @selfMA
      link
      English
      21 year ago

      oh yeah I think that’s just zstandard! it’s fairly easy to decompress if you’ve got access to a Linux machine or similar, where it’s just unzstd if you’ve got the zstd package for your distro installed

      any chance they’ve got the script they used available? we could use it to grab everything from this year and complete the archive

      • @blakestaceyOPMA
        link
        English
        31 year ago

        I don’t think the script is available (and it may be nonfunctional now, going by the terse notes at the above-linked wiki page).