devlog

I had a lovely chat today with miegawa[sic?], a core developer of the plagger feed aggregation / re-interfacing project, and found the quick overview fascinating. Right now, plagger is a kind of feed multiplexing system, which talks to different well known services like gmail, delicious etc, but has generic interfaces to stmp, irc and other protocols. Two weeks after development started, it’s already racked up 18 committers and 67 different plugins. It’s perl-based, POE-based, and a package release is expected in a couple of weeks.

Why do i like it so much? Because it connects, with resonance, to what i’d been trying to do with the nodel microframework, particularly the sets of interfaces to different services i’d started to build into it. But more than that, it looked like the interface of my dreams for pouring hyperlogging data into. For the relatively brief time i was hyperlogging everything, i definitely noticed bursts of serendipity in my own output stream, that tightened the loop of focus on what i needed to do, and reminded me of connections i’d forgotten i had made. At the time, I was generating RSS feeds of all the information i emitted through the keyboard via different communications protocols and clients, and using drupal’s RSS aggregator to provide a very simple temporal overview of the streams intersecting.

One aspect of plagger that i really enjoyed, was its use in pushing conversations through from one in put stream to another. A lot “drops out” when pushing different logs through gmail; using gmail’s fil tering, tagging and search capacity but separating out commit logs, irc chatter, selected list archiv es, etc from the main inbox stream. Here there was a goodish chunk of what i’d been able to see through ghug, at least in the first implementation; use the communicat ions network information gleaned from shared emails, as a key to augmenting that information with web resources, the contents of FOAF files, etc.

I’ve been thinking about picking back up with nodel/bbox more and more recently. After four months of paralysing code burnout, i think the circuits are starting to grow back a little bit. At least, i’m rebuilding a simple ghug application, using the mail aggregation part for a very low-key workshop/demofest organising project. My ideal is a conversation which documents itself, can be participated in through anyone’s interface-of-choice, and which helps participants to correlate their interests and nudge them into planning activities together.

This is part of a bootstrapping effort for me. This year i want to start something new that forms an offshoot from World Summit on Free Information Infrastructures. I have a fairly clear vision of what i think can happen, and who i’d like to be involved. A lot of these people haven’t really talked much before - that’s partly why i’m so interested in getting them together.

I don’t want to impose tools on people; i find this problematic myself, having to move out of channels, creation mechanisms, in which i feel comfortable, into different forums. Resistance to starting mailing lists, starting wikis, starting yet more input and output channels which have to be monitored and maintained. I want to bring more of the information i create into one place, where i have a chance to reflect on it.

This is a very personal process, yet i think it is a key to answering the questions that the many people looking for the heart of ‘Web 2.0′ and not finding it, are asking. Before other peoples data and habits become much more useful to us, it makes sense to make better sense of our own data and habits. I have heard said, the good is the enemy of the best; partially useful, partially obstructing tools undermine this process of constructive self-reflection. It can be easy to become immersed in other peoples’ terms for things, as it is to become accustomed to other peoples’ tools.

I think that one can’t extrapolate a collaborative conversation completely, by looking only at its traces in the world; one needs the means to participate in it. But one needs to find one’s own terms to express; if i write truth, in words or in software, then it will mean the most, initially, to myself. If i can explain myself clearly, i can live on my own terms. If those terms become transformed through learning from others, in a hyperlogged information archive, hyperconnected there is a tool to compare past and present; a means to reflect on one’s own tone, tendencies, ethics even, in writing code to do different things, in making concept sets to convey different things.
When i made tools that were useful to me, i often stalled at the classic dogfood stage, where i could consume what i produced, see the shape of the problem that i’d solved, and then defocused, moved onto something new. The semantic web approach changed my world, because everything i made could learn from everything else. An application could be complete in itself, until something came along to learn from it and potentially teach it in a new way.

Having to Stop Coding for a few months has given me a chance to examine my own drives towards technological solutions to social problems. I’m trying to pragmatically reassess those drives that i still feel. What drove me towards the semantic web in the first place was the desire to build tools that would allow people to collaborate better by communicating more clearly. Networking together information resources and reconnecting them underneath in new, nonobvious ways - particularly through simple temporal and spatial associations that conversations make, just as a byproduct of where they take place.

When i was writing bbox/nodel, i hoped and planned for it to disappear; for the software that i made to work down into increasingly thin slivers of glue, until the glue dissolved and the pieces that it held together would either stick together by their own momentum, or fall apart naturally. So i’m alw ays really pleased to see “other things like it”. I see a fair bit of similar potential, though less outre a future vision, in reBlog, too, were that extended with more kinds of simple metadata like GeoRSS annotations.

I have an awful lot backed up, and could probably ramble on all night, but i’ll stop here for awhile, and see what catches my attention over the next few days, while i dream up entertaining things to say about hacker ethics, free information infrastructures and awareness of the world around us.