News

The Tale

Once upon a time (earlier this year, actually), a young Eskimo lived where you might expect her to live - in some rural, secluded place that is below freezing 24/7/365. Despite her seclusion, she wasn't completely alone. Every day, for several weeks, a small airplane flew overhead and dropped a gift in front of her igloo. Each day, she eagerly opened the present, only to find some sort of refrigeration device within each box. One day, it was a large freezer. Another day, it was a small refrigerator. Yet another day, it was a portable cooler. No matter how often the airplane dropped its gift in front of her door, it was always something the young Eskimo obviously didn't need, and had absolutely no interest in whatsoever.

She didn't know who was sending the gifts to her, but she decided one day to try to influence the sender's gift-buying, in hopes that she might receive something more useful. So she gathered up lots of brightly colored objects lying around her igloo, and used them to form letters in the snow. The next time the airplane flew overhead, the pilot saw her "wish list" spelled out for him. "I like solar-powered gadgets and warm clothes," he saw from overhead. She never again received a refrigeration device. Instead, her gifts always contained exactly what she asked for and was interested in receiving.

You must know about Twitter's whale, the sort of official spokesperson of twitter. As the so-called Fail Whale made more appearances, people moved in droves to Plurk.

Plurk has been holding up better than Twitter, which is no doubt very dependent on the fact that it has much fewer subscribers. But it's not immune to the downtime as today, for the first time, I spotted Plurk's version of Twitter's Fail Whale:

As promised previously, the Reddit data is now shown in the Alerts Summary page. Right now you can see all submissions from the past 24 hours from the domain names you're tracking.

The alerts are not being emailed out at the moment pending some more tests to calibrate when the alert should go out. The rest of the post explains how I'm thinking of structuring the reddit alerting service to give you an idea of what the calibration entails and how the service will work.

A while back I blogged about how to track MyBlogLog profile views. It was a simple trick looking for hits to the associated RSS feed on your site coming from a set of Yahoo!-owned IP addresses. Now there is a minor update.

Until very recently (today even!), the user agent header of those RSS requests was blank. This was annoying me because it's a bot that's not identifying itself properly. I even emailed MyBlogLog support about the issue. As of this morning, I'm seeing a new user agent:

As per the title, Social Alerter is now tracking Reddit. The real-time tracking is for both the "what's hot" list and the "new" list sorted as rising.

Over the weekend, the final tests will be carried out, and all being well, early next week the alerts will show up in the Alerts Summary page and be sent out by email.

So... what other services would you like Social Alerter to track?

A story on Read Write Web asks if Digg is in trouble. The premise is based on a new comScore report that Yahoo! Buzz has now overtaken Digg in terms of traffic volume. The money quote:

Question: How many links does a story that just went popular on Digg have?

Another question: Say a story goes popular on Digg today, how many links will it have in one week?

Let's go for a hat-trick: how many links will a story have in 30 days?

Answers:

Hot off today's launch of the Mixx API, I'm very happy to announce that the Mixx stories are being tracked by Social Alerter.

I came across some really odd behavior at Digg.com involving what appears to be banned users. I'm not sure what's going, but it is very intriguing.

The Blackhole

One story in the Social Alerter database is logged as having 2604 diggs (which is why it caught my attention) and 255 comments. The story is supposed to be at http://digg.com/general_sciences/64_65_4 but browsing there gives a 404. It's a super-Digg and it's not found?

We already talked about the effect of Digg's January algo change. The summary of that analysis is that the chances of a story getting promoted after getting 25 diggs was cut by 38% after the algo change.

Today we visit the algo change again and, let me cut the story short, the algo change had exactly the opposite of its intended effect. Let's dive in.