My shared feeds

It’s truly gratifying to see the re-launch of the Share Your OPML site. I joined the original SYO site (http://feeds.scripting.com, which now redirects to http://share.opml.org) late in it’s lifetime, after much of its vibrancy had faded, but it was nevertheless a fascinating excercise. At the time I was new to subscriptions and was collecting them like days, so the site was a great source of new fodder for my aggregator.

These days I’m more interested in a less-is-more sort of feed collection, and don’t read all of the feeds you’ll see if you follow that link every day, one of the reasons I like Rojo’s river of news style of aggregation is that any guilt I might have associated with not reading absolutely every item has long since vanished. I’m also not tremendously enamored of Top XX lists of any kind, but I am interested in information like the Subscriptions like mine feature.

I’m also interested to see what kinds of interesting data could be gleaned from the interplay between shared OPML and my attention stream; I don’t know quite yet what it might be, but there’s potential insight to be gleaned from the interaction of those 2 sources of data. Actually, I can think of one thing: I believe an analysis of those two sets of data would reinforce my insistence that OPML Subscriptions Lists ? Attention Data; I suspect relatively little of my clickstream would be related to my OPML .

I spoke briefly on the phone with Dan MacTough, who’s one of the driving forces behind this launch, and we had a good conversation. He assured me that some sort of APIs would be available at some point in the future (I did hear that, didn’t I, Dan?) and he commented that he believed that one of the things that lead to the downfall (if it can be termed that) of the original Share Your OPML site was that it didn’t scale well in Frontier and that he was confident that this implementation would scale a lot better.

We also talked a little about what’s missing and how well the site acheives it’s goals. Much of what I had to say there wasn’t all that useful, from my point of view, since I was a user of the previous iteration and since this site does a remarkable job of re-presenting the functionality of the initial site I was already familiar with the UI and could find my way around easily.

I mentioned disambiguation of feeds as one feature that would be very valuable. As I write this, numbers 15 and 16 in the Top 100 are both Digg, one with a “www” and one without. Not only that, but a deeper level of clarification would be nice; in my mind there’s no need to treat multiple feeds that contain the same content as separate entities just because they have different URLs (the three elliptical… feeds being a good example). We also talked about the value of being able to find the people with subscription lists least like mine.

Since I talked with Dan I’ve also noted a few other features to wish for that I’ll jot down here:

• Being able to sort the tables based on the column of your choice.
• Being able to look at the feeds you either have in common or in disparity with another user (though we may have discussed this)
• Being able to see your similarity to the Top 100.

In short, this is a great effort on Dan’s and Dave’s part, and a good day for OPML and attention overall. Thanks guys!