learnandteachonline.com

Have you been following the blogosphere conversation about Techmeme, its new leaderboard, piling on, and groupthink?

Seems like everybody who's complaining also visits regularly. Me too, but I do agree it incents bloggers to behave like iron filings drawn to a magnet.

Maybe that's the trouble with relying on algorithms. They can seek out items like other items, but math has a harder time detecting something original and new (while also being important or consequential). You need human judgment for that, just like you need people to intervene in the social problem of juvenile and rude behavior in comments, even though the techies keep saying identity systems are the answer.

Human brains and machine brains dividing the labor in a smart way. Calacanis isn't all wrong when it comes to that part.

In strolling through my referrer log I discovered a SocialRank site that aggregates news from elearning blogs: Learning Signal. Nice! I'm subscribed. Here's how the intriguing new SocialRank site works, from the MindValley Labs blog. Not sure if I'm mischaracterizing it, but it seem a little like a cross between Techmeme and Mahalo? No?

MindValley, parent and incubator of SocialRank, looks like an interesting company. It's located in Kuala Lumpur. Staff is on the young side. It's sad how an old fart like me can resonate with a stirring description of the character of a workforce and company mission, then totally know I'd never fit in when I see a group employee photo.

In a different section of my stats report -- keywords -- I see something that wasn't there last time I studied them. Are hackers or spammers making use of MS Live Search for some nefarious pursuit? I see more Live searches than Google searches recently, and the terms they tend to seek are "account" and "username." What's going on there?

Adam Green and Anne Zelenka are talking about big reading lists.

I like Anne's filtering idea. She's fed an OPML file containing feeds on parenting into Megite, the new personalized Memeorandum.

I think Megite so far has used blogrolls to pare the world down to a user's little corner of it. But Anne's topic-based file lets her discover parenting feeds specifically. It also might give her ideas on category feeds to add to her list. I ran into the same problem Anne encountered when I was trying to draw up a list about literary adaptations. There just weren't any feeds devoted to the topic, but there were blogs having adaptation categories, and some of them offered category feeds.

As for the 5-10 feed reading list? I agree with Adam that list compilers shouldn't make decisions for users about how much information they can handle. He suggests posting the number of feeds along with a reading list link as fair warning. I'd suggest also estimating and reporting the number of daily posts the feeds might yield.

Also, to clarify, it's not my aggregator that stumbles when it gets fed a list yielding 1,500 posts; it's my brain.

Maybe there does need to be a couple different kinds of lists -- big ones to discover stuff for special information missions, and the smaller, more manageable lists to add to your daily diet of feeds. Hey, this is just getting started, right? We're all just making it up as we go along.

There are bees in the butter churn.

Buzz about the notion of dynamic reading lists -- OPML files containing RSS feeds that get swapped out periodically.

• Me

• Anne Zelenka (smart blogger)
• Matt Terenzio (who's working on SSE -- 2-way RSS)
• Pito Salas (developer of BlogBridge, which is supporting and evangelizing reading lists)
• TDavid (Make you go hmmmmmm guy)
• Adam Green (who's been all around and currently seems much into Ruby)

Have you seen this? Opinmind is a specialized search engine that combs blogs for opinions about a search term and displays links to the opinions in a split screen, positive on the left, negative on the right.

For instance try putting in George Bush
http://www.opinmind.com/search.jsp?q=George+Bush

It comes out 41% positive, which I think is just about exactly his approval rating right now as reported by the official opinion polls.

- Chicago comes out 86% positive.
- Tivo is 91% in the plus column
- Colin Firth, also 91% positive, no surprise there
- Dell is almost 50-50

But it will not give you accurate results on something like "broadband" which got 35% positive, 65% negative. People don't hate broadband; they just get pissed off at their provider. That makes me wonder if the results tend to skew negative because people tend to emote more about things they don't like? Well, maybe not. Teenage female bloggers and I will gush a lot. Maybe we correct the other tendency.

I'm trying to figure out how it works. It seems to look for words like: like, love, hate, followed by a the noun you searched on. But it's more sophisticated than just that. Here's the About page. No enlightenment. I wonder if they weight hate heavier than like.