October Research Update

Amici!  It is now November!

October has seemed both too long and too short.  We're now well into both job application season for the 'first wave' of the 2020/1 academic year job market (which I talked about in the August Research Update if you want a better sense of what that means), so along with blogging and teaching, I've been doing that.  It has left terribly little time for writing, which is unfortunate, but somewhat typical of academic work, which naturally structures around the school year.

On the job front, the academic job market looks to be conforming to most of the worst fears this year.  I'd hazard new postings are about a quarter or so of what they would be in a normal year.  My current tracking list for positions I think I fit has 6 entries; at this time last year, that number was 20.  The situation is actually better in Classics (the study of the ancient world broadly construed) than in some fields.  As of Monday, there was one tenure-track Medieval History position in the United States (looking, that number has exploded now to 'two' but only if you are willing to work at a school which has a "oath of fidelity to the Magisterium of the Church" - I have nothing against religious institutions, but can also understand folks who would not want to work there).

Thankfully, due to the support here and my fortunate living situation (read: I have a fully employed spouse whose job is not vulnerable to COVID-19) I am in no danger of falling off of a cliff should there be some interruption in my teaching due to COVID-19.   That in turn means I can focus on longer term plans - Article 2 still needs to be edited before submission, I have a concept and basic plan for Article 3, and looming over all of this is the grand project of the Bookification of the Dissertation.

Views, Hits, Links, Oh My!

Given that it has been a mostly slow month loaded with teaching, grading and job applications with not as much research progress (though I do have another traditional media article filtering through the editorial process, essentially trapped behind Election!! backlog, so I have done some writing), I figured I'd talk a bit about the blog back-end and where my views come from.

October was almost as busy as September on the blog, driven mostly by the popularity of the "How Did They Make It" posts (which also drove a lot of traffic in September.  I'm now getting between 150,000 and 220,000 page views in any given month (an absolutely massive increase compared to September and October of last year - 9,836 and 50,313 page views respectively).  Judging by the view counts, it looks like around 10,000 people worked their way through the entire Iron series (minus the Wootz addendum, which I promise is coming).  Watching retention through a long series is always interesting - arranging the older series by views (so as to limit the impact of the initial day-1 reader-flood that every post gets), I find that each post in a series typically has around 25% fewer readers than the one before it (which strikes me as not at all a bad retention rate, to be honest).

In practice those numbers aren't 'internet famous' massive, of course, but my impression is that for long-form blog essays, they are pretty decent.  They're certainly quite a lot higher than I expected when I started!

So where does all of that traffic come from?  Well, I don't have fancy analytics, but the basic WordPress analytics does list referrers.  I've attached a screenshot of what that list looks like for the month of October.  While the number for October are on the high end, the basic order and scale is pretty representative of a normal month.

For whatever reason Hacker News (formerly YCombinator) is pretty much always my biggest driver of traffic; the folks over there seem to enjoy my writing a fair bit and most of my posts pop into the top-50 links for at least a little while.  That used to be an intermittent thing, but really starting with the Helm's Deep Series back in May, it is now pretty regular.

Next are the search engines; a little more than 90% of that is Google.  The sad thing there is I can't see the search terms used when people find me via google.  Amazingly, I still get a few hits every month from Baidu, which means my blog isn't yet banned in China.  So still work to do on that front.  For the search engines where I do have search terms, they're mostly my blog's title or post titles, so my working assumption is that a lot of that traffic represents offline word of mouth, or people trying to re-find a post they looked at already when they don't have a link.

Then, of course, Reddit.  I am always a bit surprised that Reddit drives less traffic than Hacker News, but it makes sense when you think about it: while Reddit is much larger it is also much more fragmented, while Hacker News provides a single feed to a smaller number of users.

After that is Twitter and then Facebook.  Through this project so far I have been consistently surprised by how relatively small the traffic twitter generates is, given that I have a public-facing presence on twitter (whereas I keep my Facebook private, since I mostly use it to talk to family and old friends).  Twitter has been invaluable in other ways (it is a really valuable networking tool, since academics and writers are on twitter at wildly disproportionate rates), but it doesn't usually generate huge traffic surges, though very early on the blog when I was getting sub-10k views per month, twitter was a significant part of how I seem to have built an initial following.

And as an aside, that even goes for August, where I had a tweet go somewhat viral (3.8k retweets, 9.7k likes) - twitter did not produce a noticeable surge in traffic for that month.

The order of those top five has been fairly consistent in the past few months.  Of course after that comes the deluge of smaller blogs and forums where my writing is being discussed.  As a former EVE player who has blown up many Goons, seeing the Something Awful forums always makes me roll my eyes (if that makes no sense to you, consider yourself fortunate and move on).

In terms of what all these viewers are viewing, I've also attached the very top of my page views for October.  One thing that I find very satisfying about blogging is that the things I write have long tails.  Going by the view counts, probably about a hundred people binge the entire archive in any given month (reasoning from the number of people who end up looking at the very earliest posts).  So even the less popular things continue to putter along in the background, picking up a few dozen views here, and a few dozen views there; none of the writing really feels wasted as a result.

At the same time, the top-spots in any given month are always the newest things (naturally) followed, with surprisingly regularity, by the three biggest series: Siege of Gondor, Sparta and Helm's Deep (generally in that order, though Sparta got a surge this past month for reasons unknown to me) with the Fremen Mirage trailing after them.  It looks fairly likely that the How Did They Make It series are also going to be crowd-pleasers in the long run - the reaction to those is very positive.

I don't have any snappy conclusion to all of this - I am a historian rather than an analytics specialist, so if there is some pattern deep in my data, I certainly don't know about it.  But given that, when I started this blog, I expected and was prepared to be extremely well-pleased with a readership of perhaps a couple hundred regulars, the scale of readership and traffic implies by the 205,000 views and 60,000 visitors in October is stunning.

Thanks for reading and I'll see you all next month!