"TL;DR: Don’t depend on Delicious; host your own, pay for it elsewhere, or hope for the best. Use real-time feeds to stitch the bookmarking diaspora back together into topical aggregate indexes." When both Les Orchard and Jeremy Keith are saying something similar, I listen.

Event aggregation, done prettily, but (I contend) wrongly. By smushing photos and blog posts down to the level of posts on Twitter, it drowns good content in mundanity. It also tries to work around Twitter's lack of decent metadata (hashtags? you're making me snort) by using times to pull out session commentary, which only sort of works. Still, add some of the slick design of this to the versatility of metalatest and you'd have a winner.

The latest from Alf Eaton: a page that pulls together posts on Twitter, Flickr, blogs (via Google) and delicious for a given search term (in this case, OpenTech, the conference today in London). A friend of mine was asking for this a week or two ago, and it's really nice to see it pop up. (I'd call it 'aggregation' but I've always meant that to be personal. Event aggregation?)

"You keep personal data in all sorts of places on the internets. Jellyroll brings them together onto your own site." Aha. I love it when the internet turns out to have built something before I need it. (Oddly, I found this while looking for Django + CouchDB hints.)

A reasonable rant on not trusting data to websites. I like Les Orchard's comment, though: "I’ve also been thinking that this is the year that I write self-hosted replacements or archiving proxies between me and all the cloud services I use." I don't care for replacements, but a local proxy (what I've been calling 'deep aggregation'): hell yes. One day...

"We will launch with the ability to backup Flickr and a blog (via RSS feed). Google Docs, Twitter feeds, Youtube and Facebook backup are all in the works" Interesting. I think I'd prefer self-hosted deep aggregation to an offline backup, though. Now I just have to, you know, build it...

A consensus on Facebook Connect becoming de-facto single sign on everywhere (albeit with reservations, and one vote for Google instead), with a few people talking about lifestreaming (and more importantly dealing with the firehose, so filtering/recommendation). Worth a scan through, although it does peter out towards the end.

More on aggregation from Phil, complete with pretty pictures.

'Some people love this ["river of"] aggregation. Good for them. I, however, am human and my eyes glaze over when trying to comprehend a chronological stream of equally-weighted events, a format only robots could love.' Phil on integrating comments he's made into a deeply aggregated site.

An activity stream of people talking about / posting photos of / etc Picnic. There's also a live stream, but I prefer using VLC: mms://streams.hosting.nob.nl/live01

"one of my design goals with whoisi was to deal entirely with public data. There’s no way to add password-protected feeds or hidden feeds." Well, that's something.

I'm not entirely sure the world needs another FriendFeed, especially when it has no privacy options. Also, call me old fashioned, but I like controlling my own representation.

Phil distinguishes between what he (sensibly) calls data-centric and people-centric aggregation, and goes on to note that duplication is a problem. As I note in the comments, there's beginning to be a consensus here.

"Fluidity, and smartness, and the understanding that friendship ebbs and flows." "We each need a secretary who can compile a weekly or monthly summary of all the interesting bits of our friends' activities"

"We need routing not aggregation"

Of course I've been tangentially wittering about this as I talk about personal aggregation (itself threatened by the likes of Friendfeed), but this is a good "stating the obvious" post about the death of the home page. (Sometimes you need to state the obv

"jaiku is now being run as a '20%' project by the jaiku staff who joined google when they were bought out." I can't read the original post, but somehow this has the ring of truth. Meanwhile, people discuss Friendfeed as a post-Twitter site. Sigh.

"Those of you who post daily dumps of your twitter messages: fucking knock that shit off. It's annoying" I'm not sure I agree with the rest of that, and I wouldn't have phrase it like that, but he's got a point.

"The huge opportunity ahead is a filter to bubble up the things you need to know without missing anything you want to know."

I should probably write a post properly examining this, but it's a good look with the problems when you combine vertical aggregation for lots of people.

Sure, ok, stop doing this. But please do offer different content in different feeds (in the same format) if it makes sense to. I do, anyway, and I wish others would too (cf horizontal aggregation rants passim).

An interesting post and discussion about whether accumulating (hm, a better term than collecting/aggregating?) your stuff makes you more or less likely to write long-form posts, and how to choose what to display.

Further thoughts from Jeremy Keith on lightweight personal aggregation ("streams") and the web app platforms that enable them