In Seneca Doane’s diary this morning, he noted that there have been fewer mega-comment diaries (more than 800) since making the switch to dKos4. An interesting discussion ensued, where many opined that since the switch it seemed that diaries in general were less lively.
As it turns out, this is true. But not, however, directly because of the new platform. Rather, because the volume of diaries has nearly doubled (which is arguably an indirect effect of dKos4).
Here’s the percent of diaries with fewer than 10 and more than 100 comments over the past 2 months, by week, excluding weekend diaries (to avoid the transition time).
There’s an obvious change when we made the jump to the new system. But, the problem is the number of diaries changed as well. Prior to the jump, there were 720-839 diaries per workweek; afterwards, 1364 and 1312!
So I went back to November 1-4 to find another high volume time period to compare. During this time there were 1492 diaries. Plopping the new week onto the chart we get this:
Well, that explains (most of) it.
But not everything. Let’s look closer – here’s how often diaries had a given number of comments during Nov 1-4 and February 21-25:
The first thing that stands out is that in dKos4, a lot more diaries get no comments other than the tip jar, or just one more. In fact, what we’re seeing is the filtering function of the new ‘recently recommended’ list, in action. In November, only about 10% of diaries with zero recommends had just a tip jar; in dKos4, about 40% of these diaries have just a tip jar. Without a recommend, it is less visible, and less likely to get even one comment.
But back to the long tail – let’s blow up that portion and take a look:
Sure enough, there are fewer diaries now with a large number of comments. The increase in diaries doesn’t explain what Seneca Doane noticed: the change in the frequency of mega-comment diaries. Why? Who knows. But we do know from the graph above that aside from the diaries with very many or very few comments, comment numbers for a typical diary are about the same in November and dKos4. About a tenth had more than 100 comments; about a third had fewer than 10. In the range of 50-200 comments, we had 15% in November and a little more, 19%, in dKos4, so dKos4 may actually do better at promoting moderately large discussions compared to dKos3 at the same diary publishing rate.
One last bit: Over the past eight weeks, the percent of diaries with fewer than 10 recommends has been constant, at about 40%; this number was not affected by the transition. However, in November, with a similar number of diaries being posted as last week, this number was nearly 60%. So dKos4 is able to distribute recommenders to a larger number of diarists than dKos3 was.
This ends today’s session of catnip for metageeks. If you were interested in this, you might also like yesterday’s diary.