Sunday September 25, 2005
Site Meter is one of a number of web sites that provides stat counters for web pages, including blogs. This allows the owner of a site to find out how many visits the site receives, what days of the week have the most visitors, and so forth. What's more, it can provide some interesting information about the software and hardware of the site's visitors: their country, time zone, default language, operating system, and so forth. Earlier I was looking at one of these statistics for visitors to this site, namely the monitor resolution, when I noticed something odd.
If you look at this blog's visitor monitor resolution stats, you'll see that the following screen sizes are the top four:
1024 x 768 32.46% 1920 x 1200 25.5% 1280 x 1024 12.96% 800 x 600 6.74%
Now, 1024 x 768, 1280 x 1024, and 800 x 600 are all fairly common screen resolutions, but what about the number two screen resolution? Can 1920 x 1200, which is large and widescreen, really be so common among 'net surfers? It would be nice to compare this blog's stats with those of some higher-traffic blogs to find out. Fortunately, Site Meter stats are readable by anyone (actually, this is optional, but most sites don't make their stats private), so we can get an idea of the average monitor resolution of blog readers.
First, let's look at Daily Kos, whose stats actually appear to be the aggregate stats of every Kos Diary. This is handy for our purposes because it gives a really large number of visitors. (If there's a Site Meter-monitored site with a larger number of visitors, I'd be curious to know what it is.) Here are the top four monitor resolutions of Kos visitors:
1024 x 768 58.59% 800 x 600 16.16% 1280 x 1024 8.08% 1152 x 864 5.05%
1024 x 768 48.96% 800 x 600 16.67% 1280 x 1024 12.5% 1280 x 854 4.17%
There are a few more 1280 x 1024 visitors, but 1920 x 1200 is still nowhere near the top—it's number six with 2.08%.
It's possible that the difference between my blog and the above blogs is a result of having a different readership. Maybe the linguistiblogosphere draws discerning readers with a penchant for fancy monitors. Hey, it could happen! To find out if this is the case, let's look at the stats for Language Log. Its top four monitor resolutions are:
1024 x 768 53.84% 1280 x 1024 11.14% 800 x 600 11.03% 1280 x 800 5.41%
1920 x 1200 is way, way down the list, at number thirteen with 0.65%, so I guess we can't pin it on language blog readers.
Comparing the stats above, it's pretty clear that one resolution (1024 x 768 a.k.a. XGA) is far and away the most common, two other fairly resolutions are perhaps a fifth as common (800 x 600 a.k.a. SVGA, and 1280 x 1024 a.k.a. SXGA), and these are followed by a bunch of other oddball resolutions down in the noise.
So why do more than 25% of the visitors to this blog seem to be using the rare 1920 x 1200 screen resolution? The answer is this single post, which describes various web sites that have desktop wallpapers in that resolution. It remains, as of this writing, the number one Google hit for 1920 x 1200 and 1920 1200 wallpaper even after more than a year, and hits from those and related searches apparently provide around a quarter of the referrals here. This is sort of cool since it means a lot of people's eyeballs pass over at least a part of TstT. However, it's clear that most of them don't become regular readers. I suppose that's my punishment for writing an interesting and informative post that draws a lot of traffic, but which isn't about either of the main subjects of this blog (linguistics and science fiction). It's a Catch-22: if I want more readers, I can try writing about subjects that are interesting to a broader audience to draw them in, but they won't stick around if the blog is mostly about L&SF. Ah, well, I guess blog-fame and blog-fortune aren't just around the corner after all.
TrackBack URL for this entry:
Listed below are links to weblogs that reference Resolution Anomaly:
What always annoys me about (how browsers report) these statistics is that many webmasters interpret them as browser size, not screen size.
I like to have my browser at a comfortable width for reading, instead of taking up the whole screen. (Even with the sidebar, this site, for example, is too wide to read if I maximize my browser window).
Lies, damned lies, and... :)
Posted by: Decklin Foster at Sep 26, 2005 1:56:12 PM
Yeah, I don't keep my browser maximized either, which is one of the reasons I tweaked the standard TypePad column layout so it resizes politely to different widths. I've considered adding a maximum width (as many sites do), but...well, if a user finds that the text on web pages is too wide, that strikes me as a really good reason not to maximize.
I'm sure there's been discussion of the maximized-or-not issue online, but I'm too lazy to look it up.
Posted by: The Tensor at Sep 26, 2005 4:23:21 PM
don't you see why you have the highest number of people with 1920x1200 resolution??? They are all searching for that on Google and then click your link - how many people without that resolution are going to search for that? BTW, thats what I did - I have 1920x1200 on my M170 and amazingly its not bad at all - so I searched for it.
Um, isn't that exactly what I said in the last paragraph of the post? You have to read all the words.
Posted by: The Tensor at Oct 18, 2005 8:32:24 PM