Our Political Blogosphere

11 comments on “Our Political Blogosphere ”

• Comments are now closed

1. gingercrush 1

   25 August 2009 at 4:10 pm

   Indeed just listened to that radio item just before. I do wonder how reliable Tumeke Blog ratings are.
   • lprent 1.1

     25 August 2009 at 7:13 pm

     The short answer to that is “more reliable than the alternatives”.

     They are

     1. scubone’s “christians to the top” one over at something should go here, maybe later. I tend to ignore it as I’ve kind of proved that a single person can manipulate its rather narrow main measurements.

     2. Open Parachute is running an interesting set of measurements using public available info across a number of sources. It isn’t of much interest for us to get involved in as we’d have to send that data either from the server of the clients. The pages loads are sometimes slow enough as it is when we get mobbed by search engines after a post.

     Each bean counter bit of javascript seems to slow the page load by a few percent, so we pretty much rely on google analytics, wordpress stats, and the backend awstats (which I pore over sometimes when I want to figure out where all of the bandwidth is going)..

     Tumeke’s system is subject to criticism, but seems pretty accurate about rough readerships. That is why the major blogs feed it data periodically to increase the accuracy
2. Bright Red 2

   25 August 2009 at 4:23 pm

   There are a number of issues with the rankings. It’s heavily reliant on the Alexa rankings, which are pretty rubbish and open to abuse. It gives this strange weighting to the number of posts and the average number of comments in the top commented post each week – why is one post worth one comment and worth one visit? And there have got to be a hell of a lot of people who aren’t caught by stats packages because they read only through rss and igoogle – that’s how I started when i became a regular reader.

   But that’s probably just being too technical, the basic order seems right – the big two at the top, then a following group including whaleoil and public address
   • lprent 2.1

     25 August 2009 at 7:22 pm

     That, amongst other reasons, is why I’m uninterested in rankings. However the rss feeds are a relatively small component of the people read. About 4-5% of traffic. At present they’re being challenged by 2-3% and rising on smartphones.

     Human lurkers are far more common than people who comment. by at least an order of magnitude. Some of them read as deeply down the posts as the people who comment.

     The biggest single readership group of our site are the search engines and other bots. Especially every time that a new generation of bots (spam, virus or search) get launched they read the entire site again . Googlebots from various places are the biggest single bot by any measure.

     We typically massively under-report our traffic because we only count humans, ignore rss, and even then exclude the moderators and writers.
3. the sprout 3

   25 August 2009 at 4:43 pm

   true, they’re far from perfect. lprent commented on this a while ago here:

   http://www.thestandard.org.nz/playing-with-rankings/
4. rocky 4

   25 August 2009 at 6:12 pm

   And there have got to be a hell of a lot of people who aren’t caught by stats packages because they read only through rss and igoogle

   Actually both rss and i-google should count in any stats package as they still have to access the site to get the content. Not sure if the WordPress stats will capture them (they may specifically exclude rss), but I know if you use the raw access logs on the server (or any program like AwStats that reads the raw access logs), it will include them.
   • Eddie 4.1

     26 August 2009 at 8:33 am

     I’m pretty sure the google analytics we use says that it doesn’t capture rss etc.
5. Russell Brown 5

   26 August 2009 at 8:11 am

   But that’s probably just being too technical, the basic order seems right the big two at the top, then a following group including whaleoil and public address

   I actually think Public Address does a bit better than the rankings indicate.

   Analytics is showing us with 272,523 page impressions this month from 84,928 visits by 23,098 unique visitors.

   Nielsen’s count from yesterday is 246,000 PIs vs Kiwiblog’s 353,000. We’ve had better months this year, but I’m guessing that’s a few more than NotPC and Whaleoil.

   So they make a good talking point but …
   • lprent 5.1

     26 August 2009 at 8:31 am

     I’d agree with that. Public Address is almost certainly under-rated. Whale and Not PC are probably over-rated. But then I’m interested in reading information rather than fantasies.

     Because I’m mainly production/tech-orientated and paying for the server, I tend to look mainly at how to reduce page impressions. That is why there are so many posts on the front page, and so few of them have read-more. It reduces the page impressions dramatically, and also the number of gigabytes that the CPU’s have to crunch and send. The majority of visitors don’t go past the front page.

     If and when we start advertising, there will probably be some restructuring to optimize the income streams. But for the moment it costs $90 per month to run.
6. Russell Brown 6

   26 August 2009 at 9:03 am

   If and when we start advertising, there will probably be some restructuring to optimize the income streams.

   Of course. But I doubt you’ll ever bring yourself to ever build the site quite the way an ad sales guy would have it …
   • the sprout 6.1

     26 August 2009 at 11:43 am

     “I doubt you’ll ever bring yourself to ever build the site quite the way an ad sales guy would have it”

     god i seriously hope you’re right there Russell

• Comments are now closed

• Log in
• Donate
• Contribute post

• Articles RSS