Liblog mortality rate: An interim note

Given my planned (ha!) “liblog investigation” for 2007 (or one of them, anyway), I’m periodically checking the 213 liblogs from the great middle that I discussed in the August C&I.

Just finished the second such check, and plan to check again every two months or so.

The good news: Only a few of the blogs have explicitly ceased, and even fewer have simply disappeared, being replaced by spam ad pages or other web graffiti.

The perhaps less good news: Taking, say, two months without posts as a sign of possible morbidity, then 31 of 213 have either ceased or gone seriously idle since the end of the scanning period (6/30). That’s just under 15%.

“Why, at that rate, nearly half of them will be gone by the end of June 2007. That’s awful”

(Yes, this is a strawman: The voice of the linear progression believer. Nobody said this. People have certainly assumed either linear or, worse, algebraic progressions in other cases that make no more sense. See the NEA and “the end of leisure reading, just as one bad example.)

I find that highly unlikely. Actually, even if the same percentage disappeared in each third of the year (10/25 is close enough to the end of October, isn’t it?), that would mean that another 26 would go silent over the next four months and another 23 after that, leaving 133–62% of the original 213–not the 55% you might extrapolate from the loss of 15% over four months.

But that’s also unlikely. Some of the 31 will come back to life. Chances are, some significant portion of the 31 went quiet because bloggers graduated or otherwise went through end-of-spring life changes.

Here’s a crude guess: I’ll guess that at least 150 (70%) of the 213 will still be active during the April-June 2007 period. I’m hoping for closer to 165 or 170, but that may be too optimistic.

No halo effect expected: If being mentioned in C&I didn’t encourage people to keep posting (and why on earth should it?), this W.a.R. post certainly shouldn’t have that effect.

4 Responses to “Liblog mortality rate: An interim note”

  1. Lazygal says:

    Walt, I tried to read your C&U article, and found the analysis very interesting. However, the listing of the blogs was… “difficult to follow” is the best way I can put it. It probably wasn’t within the purview of the piece to mention the slant of the blog (but it wasn’t easy to tell from some of the titles), and I couldn’t find live links to anything. Any chance of an update (or two) with those things added?

  2. walt says:


    I deliberately avoided attempting to define “the slant of the blog”s after the brouhaha from my 2005 discussion–and, frankly, because it’s hard to define the “slant” of many liblogs. I offered the subtitle/slogan when there was one, the title of the first post of the quarter studied, and (when comments were enabled) the title of the post with the most comments; those were intended to provide hints as to what the blog was about. Otherwise, given feeds and the relatively low frequency with which most libloggers post, I think it makes sense to subscribe to anything potentially interesting and later unsub if it proves not to be interesting.

    As for live links: C&I is primarily intended for print; the HTML sections are dumbed-down versions of the PDF. I don’t include links. This one’s actually an extreme case,and I almost didn’t do the HTML version at all–because the HTML version requires 48 pages to print out, where the PDF requires 32 (including a small essay that’s not part of the HTML version). So it’s wasteful to print out the HTML version, and I think it’s far too long to be read on the screen.

    However, as I say in the article (and as you may have missed if you were reading it on the screen…)

    “An Excel workbook including all 213 blogs (with URL as of early July 2006), start date (in an artificial “ymm” notation, where April 2006 is “604”), post count, comments, and comments per post, with a second spreadsheet including the 198 blogs that have length measures—showing count, length, and length per post—is available at liblogs2006.xls.”

    There’s a separate workbook–also mentioned in the article–containing reach figures, but that one may not include the URLs.

  3. Lazygal says:

    Sorry, I guess “slant” wasn’t the right term. Perhaps “area” might be better (school, academic, special, general, cataloger, etc.). I did miss notice of the spreadsheet… I’ll take a look! Thanks.

  4. walt says:

    I guess I didn’t feel that I had enough information to slot enough liblogs as to area (particularly the non-English ones); quite a few are pseudonymous or don’t really have any author info. Not a bad idea, though, if I was more industrious!