Sifry's Alerts

Very interesting. I'd love to see each and every spike annotated and then maybe some stats geek could come up with some detailed conclusions on what so far does interest the blogosphere and what doesn't and how those interests are changing as blogging opens up.

Also, not sure if technorati tracks geo-data, but for those bloggers who do enter their geo-coordinates, it would be interesting to see what stories generate the most posts based on region. Something along the lines of what Ethan Zuckerman has been doing with Global Attention Profiles - http://h2odev.law.harvard.edu/ezuckerman/ - of the major media.

David,

I would still like to have some more details on technorati's ability to encompass and follow both blogs and events which are beyond US borders.
It would be very interesting to know how many, let's say, spanish language blogs does technorati follow; how many portuguese; how many italian...

I'd like to know how many of these 500,000 posts are actually being read by more than half a dozen people. How many of these posts come from places like xanga and livejournal where most of the bloggers have an audience of only their close friends? How many of them are new content, and how many of them are just links to other sites posted at boingboing and slashdot? There is certainly a large volume of posts being made on blogs, but how many of those posts are worth reading?

sounds like tyler needs to get beaten with the long tail bat.

how many are worth you reading? probably not very many at all, relatively speaking. how many are worth one person reading, even if it is only the person who wrote it? probably all of them.

There is certainly a large volume of posts being made on blogs, but how many of those posts are worth reading?

Surely even if they're being read by just close friends, the value remains high (for writers as well as readers). Blogging, like anything else, has a long tail. I don't think that Kottke or Reynolds are inherently more "worth reading" than anyone else; they're just worth the same, higher-than-average, amount to lots of people.

One of the side-effects of traffic spikes is that while more people may post about an event, that doesn't mean that blog readers abruptly materialize to equally read about it. I would say that the people blogging, and reading blogs, about the US elections were a decidedly incestuous lot. Some criticism of blogs makes sense, and there was certainly more "worthwhile" reading about the elections being posted than any one person could absorb -- for myself, I cut back to practically nothing more than DailyKos, and I'm a pretty politically tuned person. In that sense, I got all the blogging that I needed, for the election run-up, from just one site and a handful of outside links every day. I'm sure this is true of more people.

What I find especially interesting about this chart is that Dave has managed to identify spikes related to non-political things such as the lock controversy and the Superbowl. That's going to be more and more the case. I imagine some of those spikes are related to, say, the hacking of Paris Hilton's phone -- something we'll have to get used to.

I shouldn't have implied that many of these posts are not worth reading. That was poor choice of wording on my part. But what I'm saying is that of these 500,000 posts, how many of them don't fit into this "long tail" that you're talking about? The point of the graphic on this page is to point out the popular topics, not the ones that fit into the long tail. I just wanted to see how many of those 500,000 posts are being read by a significant number of people, so yes, I'm interested in the most popular posts.

If the volume of posts has grown significantly, but the number of posts being read hasn't changed much, does that say something significant? I dunno, but it seems like it's worth looking at.

I should have said "being read" instead of "worth reading".

Given what you said in the previous post --that the number of blogs has nearly doubled, as has the number of new blogs-- I think the post volume is less "congruent" and more anomalous. It would seem likely that the newcomers are either posting much less frequently or are posting for a short while before abandoning the endeavour. Are stale blogs delisted?
Also, you might want to break the stats down by service, as there's been anecdotal statistical evidence to suggest that the blogging tool effects the frequency and length of posts...

Speaking of worth... Dan said, "I imagine some of those spikes are related to, say, the hacking of Paris Hilton's phone -- something we'll have to get used to." I'm guessing you didn't mean it that way, Dad, but that statement comes across as a little bit arrogant, as if only political or hard-news blogs have value, and that entertainment or vanity blogs are, long tail aside, to be scorned, especially when they actually manage to acheive a readership through exceptional writing or sensationalism. I'm curious to know what you actually meant, because I'm sure that isn't it.

Tyler: If a ten thousand people read a post at a popular blog, and a thousand of them like it enough to post the same link to their own blogs, and each of the lesser blogs only has five readers... the link has still been shared with fifteen thousand people. I think that's part of the value of blogs over traditional news services - useful or interesting information tends to spawn like a virus and spread. You can't discount the number of posts just because 95% of them are re-posts that are only read by a handful of people.

regardless of post "value", the point is more and more people are using the medium and are comfortable with the environment. so, if you are a company or someone trying to reach thousands/millions of people, you might want to understand the medium.

who cares if one or one thousand people read a post....these numbers can't be ignored. notice there are fewer spikes. either more people are posting more frequently or we have reached a temporary plateau.

also, i think the november/december lull is really the holiday seasonal stuff that happens in countries that observe holidays in that timeframe. this could infer that most blogging is North America/European centric....hard to say until we get a blogger geography map indicating the relative location of blog origination.

how about it Dave?

David
kudos for the change in language in this post in that when discussing the number of posts you refer to the number Technorati tracks as opposed to reflecting this figure as being the totality of the blogosphere.

In relation to lsantos comments whilst he has a point in relation to the extent of non-english blog tracking, I think you've better defined the issue by referring only to what Technorati tracks.

Overall, great stuff though, keep it up.

Just to address the question of worth-- I think what would help is more self-classifying content according to intent. What are the thoughts on the http://civilities.net/OnlinceCensus ? Or the much simpler http://civilities.net/TheFourQuestions ?

make that
http://civilities.net/OnlineCensus -- questions about how people self-identify.