Sunday, February 17

There's some software I would like to have. It's not, on the face of things, very complicated. It's mostly a set of shortcuts for aggregating and annotating stuff published by other people. By way of illustration, here's a sample set from today of the kind of links and fragments I would like to archive and republish, in a low-key way.



Uncreative Writing: Redefining Language and Authorship in the Digital Age


@vruba and @ayjay are both worth following. They also each have a "things I'm reading" account: @vrubareads and more than 95 theses, respectively.


So I guess there's a Long's Peak webcam:


Wikipedia: Giordano Bruno:

Giordano Bruno (Italian pronunciation: [dʒorˈdano ˈbruno]; 1548 – February 17, 1600), (Latin: Iordanus Brunus Nolanus) born Filippo Bruno, was an Italian Dominican friar, philosopher, mathematician and astronomer. His cosmological theories went beyond the Copernican model in proposing that the Sun was essentially a star, and moreover, that the universe contained an infinite number of inhabited worlds populated by other intelligent beings. After the Roman Inquisition found him guilty of heresy, he was burned at the stake. After his death he gained considerable fame, particularly among 19th- and early 20th-century commentators who, focusing on his astronomical beliefs, regarded him as a martyr for free thought and modern scientific ideas.


It turns out you can get a fountain pen highlighter.


None of this is all that momentous, but there must be half a dozen items a day that I'd like to note in passing: for future reference, for organizing my own tangled thoughts, for a signal boost to anyone who happens to care what I'm reading, to make the web a little more redundant and a little less dependent on the centralized power of big corporate actors with billions of dollars.

It's easy enough to copy and paste these fragments in the context of writing a blog post. What I'd like is a way to make the accumulation of all these fragments as simple as clicking a button. Something that could become an automatic gesture, with no more overhead than upvoting a comment or favoriting something. Google Reader used to scratch some of this itch by making it easy to reshare articles, but then they took that away.

I know there are services like Storify, and that the social networks are doing their level best to consume the entirety of publication behavior. I'm also well aware that for lots of users, syndicating a stream of the things they notice floating past is basically the use case for Twitter, which is actually pretty good as a place to find things and share them. I know people are doing this stuff on tumblr and Pinterest and so forth. I'm just not especially satisfied with any of it.

Ideally, the toolset I want would have the following properties:

  • Open source.
  • Built around standard web tech.
  • Self-hosted, not dependent on a service or managed by any of the usual walled-garden suspects.
  • Browser-integrated via a bookmarklet or similarly low-impact thing.
  • Equally workable from my laptop, phone, and tablet.
  • Smart enough to snag individual tweets, Flickr photos, comments, and other units of publishing.
  • Archival, searchable, and random-access.
  • Syndicated by default in a machine-readable format (Atom would be nice) and maybe optionally to services like Twitter or an e-mail list. Preferably with a distinction between "save just for me" and "drop into public stream".

For serious bonus points:

  • Some kind of Gmail-like label and filter system.
  • Smart enough to integrate with Google Reader or another feedreader at least as good.
  • A git-backed storage mechanism.
  • Export to standard citation formats.

Any thoughts? What do you know about that I don't? Leave a comment in the margins or tweet at me or something.


A few additional notes:

I've tried to do a simple version of this since maybe 2006 or so in the LinkDump page sidebarred on this site and elsewhere on the p1k3 Wala. This has the standard problems: It's almost as much work to copy-and-paste there as to anywhere else, the resulting feed isn't very pretty, and, most irritating after a few years, everything lacks context. It would be nice if the tools could snapshot things for later retrieval and thus provide at least some buffering against the inevitable link rot.

Zotero, which I've looked at before, has a lot of the right ideas, including automatic snapshots of everything it saves and support for lots of formats (it knows, for example, when it's looking at a scholarly publication or a tweet), but it's probably not quite the beast I'm after. Using it with Chrome requires a standalone application, and it's not really geared towards republication.

I just gave Maciej Cegłowski $10.03 for a Pinboard account because he's an interesting voice and the site gives every indication of being web software done right, including a feed version of most everything, simple bookmarklets for saving stuff, and an archiving feature. The sticking points are that I can't work on the code and I can't host it myself, but it's something I might give a serious try anyway.