When the itch is inside the boot, scratching outside provides little consolation.
—old Chinese proverb (reportedly)

This is a list of things I want to get working on this site. It is more for my bookkeeping than your reading, but if you have suggestions for things or how to best ‘scratch’ any of these, let me know!

To do

  • automate sending webmentions in some sane way. (Django post_save signal failed.) Look into Django signals
  • parse received webmentions using an asyc queue.
  • move to a file-based storage (consider html+mf2) and use MySQL for meta-cache only. See databases-antipattern (responses to articles moved to YAML files.) Started Pullback-Note to move notes from DB. Started Manifold-Note notes now stored in html+mf2+json files. Add delete and undelete. Change update to write only what is changed.
  • eliminate explicit post types (aricles vs notes) in DB?
  • migrate bookmarks from Feedly
  • use micropub to build a good posting UI alternative to Django's admin.
  • indie reader (far-future; look into microsub)
  • better homepage idea: use homepage for updates, including status (like currently reading), latest notes and articles and also entry points to explore other things on the site. Reserve about page for info about self and site. Why?: homepage is where readers would land first, and presumably they are more interested in "what's new?"
  • better reply-context and embedding.
  • auto-linking (such as @username) in notes and responses
  • chaining conversations.
  • notifications for new activity such as new responses to posts
  • self-host photos/art. move from Google and deviantArt. import old tweets and G+ posts to own site.
  • move draft posts from old blogger site. put up move notice on posts with new URL.
  • composite stream of all post-types
  • accept webmentions from Twitter and G+ through form
  • cleanup sub projects and put on Github for others to possibly use.
  • accept webmentions linking to alternate URLs of posts e.g. https or http
  • publish page listing replies to some URL. Useful to see all marginalia to some particular post/ebook.
  • import ebook notes from Google Play to site as marginalia
  • markup papers data with mf2 to represent all information like DOI and journal-accepted versions.
  • fix hfeed2atom to include response-context
  • delete the responses container if all responses are marginalia. (might not actually be an issue IRL)


  • Improve webmention parsing using mf2py. (Done: as of at least 2014-04-22)
  • enable backfeed using Bridgy. (Done: as of at least 2014-04-22)
  • start posting likes, reposts, links, bookmarks
  • enable https and SSL stuff (Done: as of 2014-05-23)
  • enable micro-level responses aka annotations using fragmention. (Done: on 2014-07-10 see: marginalia-active)
  • receive mentions on Notes. (Hint: use url to identify posts instead of DB objects.) (Done: as of 2014-04-23)
  • cache or lazy load avatars for improved performance Done: cacheing avatars for a year using Google's image cache/resizing (see Google's authentication-less, on-the-fly image resizing service)
  • move itch page to own site (Done on 2014-07-03)
  • paginate notes feed with some sane date grouping. Possibly by year-month (Done on 2014-08-02)
  • separate colophon page listing the plumbing innards of the site. Not necessary
  • fix responsive images height issue: currently my responsive images don’t know their heights and so cause flow when they load. This interferes with fragmention for instance. (Done as of 2017-01-14; responsive images have a placeholder of correct size until they load.)
  • remove the over-zealous fragmention handling in receiving webmentions which assumes the double-hash “##” notation; use whatever URL is received. update marginalia to accept this. (Done as of 2016-12-27)
  • remove marginalia.js from notes (since it is useless there), and put only on article permalink pages. (Done as of 2016-12-27)