Attention podcast part 1, Joshua Porter and Alex Barnett, part 1

This is the first of a two part podcast Skype chat with Joshua Porter and I (part 1: .mp3, 43mb, 45mins).

Part 2 tomorrow. (update, here it is).

Joshua is a little faint to hear at first, gets a bit better later though, sorry about this.

Notes and links below.

Tags:  A, , , , , podcast

Comments (8)

  1. Danny says:

    Ok, I listened to my third podcast this month – that’s two to you, one to <a href="">Tim Berners-Lee</a> 😉

    Shame about Joshua being quiet for the first few minutes, but overall it was very listenable.

    In general I think the way you talk about attention is on the nail. Empowerment, data ownership etc, yes indeed, these are very important. The self-reporting issue is interesting.

    On the bit where you mentioned my name – yep, you’re right, I don’t think the specific format used for link lists, attention or whatever matters. For the end user it certainly shouldn’t matter. For the developer too, it shouldn’t matter, as long as the data is usable.

    OPML has a head start for users in its use for feedlists, and being XML based it is potentially (re)usable for developers. But I’ve been bitching about OPML ™ for at least three reasons. Firstly, with its current specification it isn’t actually an easy format to work with, mostly there’s no real way of telling what you’re going to get (e.g. whether the title of a piece of data is in the "title" attribute or the "text" attribute). Secondly, it’s not clear how it could be *consistently* extended to different applications, such as attention (will there be new type attributes, or namespaced attributes or what?). Thirdly, (which is tied very closely to the other two points) I haven’t found Dave Winer’s approach to specifications very helpful at all. It’s all very well saying that users only care about applications, but someone has to write those applications. Anything that makes that harder for the developer has a knock-on effect to the users experience.

    Having said all that, RSS 2.0 is pretty crappy too, but for its faults it has made a lot more data available on the Web, and helped application innovation. Whether the same or better might have occurred had it been better specified we’ll never know. I’m fairly sure Atom (especially the protocol) will lead to another burst of innovation.

    But as with RSS, I expect to be coding around whatever formats people are using for lists and attention, there no doubt will have to bits of hackiness to get around format glitches and the nature of data-in-the-wild whatever is used.

    Incidentally, there’s a little irony in you talking of third parties messing with your attention data, and talking of using OPML for attention data. The only significant breach of "Attention Trust" I know of is the sale of, a receiver of attention data, without clear notification to the producers (owners) of that data.

    But it was interesting how quickly you got to the issues of trust and access control of the individual’s data. These are not easy questions technically or socially, but there’s a lot of ongoing work (mostly around Semantic Web technologies) which should help with the technical side at least.

    Whatever, the discussion around attention is definitely worthwhile, and thanks for the podcast.

  2. Danny says:

    PS. if you’re doing lists of what’s on the podcasts (nice to have), this might be of interest:

  3. Danny says:

    PPS. Why don’t I personally think the specific choice of format matters is that big a deal:

  4. &amp;nbsp;

    Here’s an OPMLish podcast for you, March 10, 2006

    &quot;It’s all about the draft OPML 2.0 spec…