I thought Eric had a better mastery of grammar than that. Did he say it themself? I also thought he was better informed than that.

There's plenty of argument for that usage, to read up on the topic check out "singular they" and "third person plural".

But, factually, isn't he correct? Most blogs have no other readers besides the writer of the blog. True or false? Or even if technically false, only a slight exaggeration for humorous effect?

I suspect blog-evangelists want to make this into a straw-man of "There are NO blogs with more than one reader" - which is ridiculously false, but the maker of the straw-man can feel good about knocking it down.

I assume we're talking about active blogs here. The dead probably outnumber the living, and they're not read by anyone.

It's probably not literally true. Most bloggers read several other blogs themselves, and many are quite promiscuous. And not all readers are bloggers, so there are a fair few pairs of eyes to go around.

I suspect that most active blogs have no more than a handful of readers. I don't know how many readers my blog has. A very good post will rack up maybe a couple hundred page views over a week, but I don't know whether they're unique views, or whether they included my co-bloggers. (Wordpress.com says that they don't include "your" visits, but I don't know whether that's a singular or a plural "you".)

Ah, but remember, readers are by no means evenly distributed. But we're only debating whether it's literally true or poetic license.

Someone who really wanted to study the end of the curve would have to be very careful about spammers and search-bots inflating the view count. One spider downloading every page of a blog can generate a huge number of spurious page views. Even unique-IP can be fooled, since some spambots use multiple IP addresses to try to evade detection.

I assume that bots of all kinds are likely to result in an access profile which is flat across all the pages. The actual number of hits to any single page follows an approximate power law distribution. The most popular article gets two or three hundred hits in a week. The tenth most popular article gets maybe twenty.

Plus, I know I have some readers because I get comments. Some of them even pass the Turing test. :-)

I've no doubt you have more than one reader. I meant my statement more generally - someone who really has zero other readers could easily think they have a few, which are in reality spambots. Gathering useful statistics down that low requires being careful of such distortions.

Then I guess we are pretty fortunate to know all of our 10 regular readers by name ;-)
(and I find that the number of readers mostly match with blog-software-proivided numbers – but you are right that you have to interpret the data in a pessimistic way)