Walt at Random

Here’s an honest question, where I’m actually looking for advice–although, admittedly, factors beyond email and comment responses could influence my decision.

The question:

Would a really big look at liblogs, including lots of year-to-year change data, be a good idea, a waste of time, or a positively bad idea?

Definition: “Liblogs” = what Steven Cohen calls “libr* blogs”–that is, blogs by “library people” as opposed to official library blogs, but not limited to blogs by MLS-holding librarians (as if there was any way to know!).

Now, if you already have an answer without reading further, great: send me email or comment below. If you actually want a little clarification…read on below the fold.

---

Really big look: The population for the new study would consist of:

• All the blogs in my 2005 “60 interesting blogs” survey that are still active. (See this essay or this issue.)
• All of the 213 blogs in my 2006 study of “the great middle” that are still active. (See this issue–since the essay is essentially the entire issue, it’s a better bet than the HTML version.)
• A bunch of others–including those mentioned in Meredith Farkas’ “favorite blogs” study, those in LISWiki’s blog list that weren’t included in 2005-2006, those in the LISZen source list, those in Dave Pattern’s “library blog cloud” source list, and those I just discovered on my own–that meet the base criteria.

Base criteria for those that weren’t in one of the other studies:

• In English
• Not clearly defined as an official library blog
• Somehow at least vaguely related to libraries or library people
• Reachable
• Established before January 2008
• At least one post between August 31, 2007 and March 1, 2008
• “Visible”: The sum of Bloglines subscriptions and Technorati “authority” in the first two weeks of March 2008 is at least nine.

If I do the full study, there would be one more criterion, for blogs that weren’t in earlier studies: “Semi-active”–having at least one post in two of the three months March, April, and May 2008.

That population–not including the final criterion–is now 542 blogs, including 48 added from Farkas’ “Favorites” report, 81 added from LISZen, 37 added from LISWiki, 9 added from the cloud, and 29 others (items were added in that order–if something was added from LISZen, it wouldn’t also be added from LISWiki).

Lots of year-to-year change data: If I do this, I’d have the following:

• March-May 2007 data for all blogs for which it’s available, noting that data would be limited to what’s reasonably available. (E.g.: If the archives for a blog hide most of each post, I’ll include post count and comment count, but not length of posts–I’m not going to take a sample and extrapolate, and I’m sure not going to retrieve each post individually!)
• March-May 2008 data for all blogs.
• Comparisons between 2007 and 2005 for 43 blogs that were in the 2005 report and not the 2006 report.
• Comparisons between 2006 and 2007 for surviving blogs that were in the 2006 report.
• Comparisons between 2007 and 2008 for all blogs available in both periods.

If I do this, I’d establish norms and quintiles based on real populations: Thus, overall length and length per post would only include blogs with easily-retrievable full-text archives; comments overall and comments per post would exclude blogs that clearly don’t allow comments (or that have comment counts hidden in archives).

---

An honest question, this. Last weekend, I did enough experimenting to conclude that it may be feasible to do this megastudy this summer/fall–and I’m planning to do the 2007 metrics for 2005 and 2006 inclusions (they’re about 1/3 done already) for my TxLA appearance. A lot of work for five minutes out of a 50-minute presentation, but it should be interesting.

So the question is: Do I do the other 2007 metrics and do I plan for the big project?

If I don’t, I’ll turn the current project into one or more blog posts or C&I articles.

If I do, I’ll produce a book. It might even have one-sentence summaries of what I believe to be each blog’s focus and strengths–but only when I have something nice to say and am capable of reading the blog. I wouldn’t include a full sample post for each blog; I might include a paragraph. I would have a little writeup on each one.

So: What’s your opinion? (I’m not asking “Would you buy the book?” Different question.)
Note: If someone offers me another part-time gig, this whole discussion might be moot.

Another cheat title, I’m afraid–and again, the primary purpose is to point to a post elsewhere, commend it to you as worth reading, and maybe argue with a little of it.

This time, the post is at a blog that seems to have more than one name: The pagetitle, which is also what Bloglines tells me it’s called, and the banner title, too long to reproduce here, a shorter version of which is what Google Reader tells me it’s called. I do know the blog is by Rochelle Mazar–and, for those of you who remember a certain contretemps a couple of years ago, this is clearly not a blogger who always agrees with me (not by a long shot!).

I’m not using the post title itself because that would indirectly feed more publicity to the post Mazar is discussionng. I’d encountered the post previously (a sure-fire list of questions to make you a better blogger), skimmed it, said “geez, another list posited on the basis that all blogs are essentially marketing blogs,” and let it be. (I’ve said elsewhere that I get touchy if people call me a “para” or “sub” anything. I’ll get more than touchy if you talk about my need to “promote the Walt Crawford brand.”)

I probably dismissed the other post so quickly that I didn’t even notice the recommendation that bloggers poll readers to find out how they should be blogging/what they should be blogging about. Excuse me? Let me think about the libloggers who I believe have the broadest reach. Let me think about the libloggers whose work I value most (it’s a classic Venn diagram–two overlapping circles). Let me think about the circle of libloggers who really would revise their blogging style or coverage based on reader polls.

Hmm. If such a circle exists, I don’t believe there’s any overlap with the other two circles–or at least I hope there isn’t.

Yes, I asked for reader feedback on coverage within Cites & Insights once or twice, a few years back. I even paid attention to the results–for a little while, until I realized that it made no sense for this particular ejournal. Even if it did, an ejournal is a very different animal than a blog.

I have three blogs (oddly enough). Only one is a personal blog–this one. Blogs aren’t one medium; they’re a particular style of lightweight epublishing tool that can be used for quite a few media, whose only commonality is that written items normally appear in reverse-chronological order. The other two blogs are, in a sense, marketing blogs–one to let people (who don’t want to read W.a.r.) know when new issues of C&I are out, the other to keep PLN users informed about new items and remind them periodically to check on PLN.

This isn’t a marketing blog, at least not most of the time. When it is (to get me a new job, to get people to buy C&I books), it seems to be fairly defective. But that’s not the primary goal. Nor is it, I think, for most libloggers.

Where do I disagree with Mazar (in this case)? Primarily this point:

4. Do I need to blog under an assumed name? This is especially important for anyone under the age of 25. You never know when you’re going to change careers and have something you wrote online when you were 15 come back to haunt you. Unless you really trust that you know what you’re doing, the answer to this question is probably yes.

I agree that it’s important to know that what you write may come back to haunt you. I wonder whether blogging under a pseudonym is a reasonable response–unless you’re determined to make sure there’s never any link between the pseudonym and you.

That’s not easy. I’ve seen any number of cases where someone starts out under a pseudonym and then wants to brag about something, or writes something that’s so local and so specific that colleagues and coworkers can readily identify them, or just lets slip something clearly identifiable. (Worst case: the blog is identifiable through domain ownership or other means…)
If you want to blog under a pseudonym, I think you have to assume you’ll drop the blog after a while. You’ll find that the limits of pseodnymity hamper your thinking and your writing, or you really will want to say something from your heart. Not that there’s anything wrong with dropping a blog, of course… until you start another one, signing it, and somewhere down the road make a reference to the old blog that lets the blogger out of the bag.

Doesn’t always happen, to be sure. I don’t believe anyone will ever know with certainty who the team or person responsible (or irresponsible?) for the Annoyed Librarian actually is. But that’s a fairly rare case.

Anyway, niggles aside, Mazar’s response is a good one.

Yes, I know, this post is two days early: On any normal day, I’m the last one who would be trying to answer that question, for a variety of reasons (most of which I’ve mentioned).

But that’s the title of Dorothea Salo’s post (which in turn links to some related posts), and I think it’s an interesting, challenging read.

Which is just about all I’m inclined to say about that.

Oh, except for one correction clarification note: Dorothea sez:

Speaking of Walt, who’s a systems analyst by training and trade,

Well..I was never trained as a systems analyst (or as a programmer, for that matter), unless you consider the extent to which the Rhetoric program at UC Berkeley (technically, Speech most of the time I was there) included the study of logic.

By training, if anything, I’m a writer and editor–although, there again, it’s mostly self-taught (thus letting a bunch of teachers off the hook). And it looks as though that’s my trade at this point, by design or happenstance. Since I started doing that (that is, writing for publication and editing other people’s writing) years before I started doing library systems work, you could say that I’m a writer and editor who had a really worthwhile day job as a library systems analyst for a few decades.

I’ll probably always be an analyst (and synthesist, which I regard as more significant if only because it’s more unusual and less teachable); it’s in my nature.

That’s a sidebar, to be sure. Do I agree with everything in Salo’s essay? Of course not. Does she raise a lot of important points and state them well? Of course.

(Would I take an honorary doctorate? Certainly, especially if it included an interesting trip/speaking combination. I’ve spoken at four library schools in the past and enjoyed it each time. But, well, I’m not going to hold my breath.)

---

The use of crossed-out text in blogs doesn’t always mean you edited it post-publishing. It’s also a cute way to indicate you’re not quite sure what term you want to use, and are ducking the issue by using more than one. But you knew that already, right?

1. I’m 62. I don’t consider myself a “senior citizen.” I doubt that I’ll consider myself a “senior citizen” at 65, or 66 for that matter. Nor do I plan to go away and hide when I become a “senior citizen.” But I promise not to go take a librarian job away from some young person scolding people for not retiring when they should.

2. I’m not a professional librarian–both because I lack the degree and because I don’t work as a professional librarian. (I haven’t worked in a library since 1979, and even then I was in a systems office functioning as a programmer/analyst.) On the other hand, call me a “paralibrarian” or “paraprofessional” or “support staff” or “sublibrarian” and I might get snarky about it… Oh, and suggest that awards for service to the library field should be limited to those with the proper degree (which, I suppose, means I should turn mine in–not M&S, which I’ll probably never have, but some others), and I might sneer a little.

3. On the other hand, I’m a little astonished to find non-librarians scolding libraries for failing to run out and buy books that don’t have ISBNs, that apparently haven’t been reviewed in print media, that aren’t available through distributors, and that have titles that more than a hundred other books have. Oh, and that were free downloads before they became print books, and are still free downloads… Maybe I underestimate the omniscience that good librarians should have, and maybe I underestimate the extent to which libraries are funded for and expected to handle universal digital preservation.

I think I’ll leave the links out of this post. I’ll probably get in enough trouble as is…

Some of you (OK, 500+ of you who get posts via aggregators, maybe) will have seen the post that originally graced this space. It had to do with a third-rate PR firm, a “bestselling” author who writes like a…well, never mind…but who could sink several hundred thousand into self-publishing, and the vagaries of Amazon “#1″ ratings.

And, the more I thought about it, the more I thought it was pointless. So I replaced it with this.

I’ve always treated Cites & Insights as “published”–that is, once an issue appears, it doesn’t change. I don’t correct typos or meaningful mistakes. (When the publication moved to the current domain, I revisited each PDF to change the domain name in the masthead, but made no other corrections.)

I’ll stick with that standard for actual errors–cases where I’ve left out a word or said something incorrect. Naturally, I try to do followups when needed, but it’s good to keep the published record intact. And I don’t plan to go back and fix dumb typos in past issues…

But I just replaced the PDF and two of the HTML essays for the current Cites & Insights, and it’s likely that if a similar situation arises in the first week after a new issue’s published I might do the same.

What changed? Three cases where the string “egan” within a word appeared as “Elgan” instead–one “bElgan” instead of “began,” one “elElgant” instead of “elegant,” and one other (I’ve forgotten the string). In no case could an incorrect meaning have been assumed; it just looked stupid. A reader in Australia alerted me to the problem this morning.

Clever people can probably guess what happened…and I really should know better. Here’s the whole silly story:

In an attempt to minimize typos and other errors introduced in the copyfitting process, and to give the material one last read, I now consistently print out an issue after I’ve gotten it to the desired length, let it sit for at least a day, then read the hardcopy as carefully as possible, marking any changes.

In this case, one section of the Kindle & ebooks essay included notes on a Mike Elgan column–and somehow I’d managed to alternate “Elgan” and “Egan” roughly equally throughout the notes. I wasn’t sure which it was, and did the search to verify that it’s Elgan.

Then (ahem) I did a “replace all”–and, duh, forgot to check the “Match case” box.

Actually, I think I had the section of the text highlighted–but I’d also forgotten that one step backward in Word 2007 (from Word 2000, and this may have changed earlier) is that “replace all” no longer limits itself to a highlighted region, asking before going any further. (I can’t find any way to restore that limitation. Anyone out there know of one?)

So there it is: My extra step to minimize errors worked great…except for introducing a few new ones.

It’s good to be perfct. It’s also unsusual.

Cites & Insights 8:4, April 2008, is now available for downloading.

The 28-page issue is PDF as usual (or not as usual–I’m now using Word 2007 and Microsoft’s free PDF-output download), but HTML separates are available from the C&I homepage

The issue includes:

• Bibs & Blather – a very short introduction to the issue and what got left out.
• Library Access to Scholarship – Harvard and Institutional Repositories
• Old Media/New Media Perspective: Thinking About Kindle and Ebooks
• Making it Work Perspective: TechNos and TechMusts

By the way, if you know anyone who’s been getting issue alerts via email, let them know they need to sign up for C&I Updates or Walt at Random; Topica no longer accepts my posts (and entirely lacks help/contact info).

Let’s wrap this up. (I’m delighted to see three sales of Academic Library Blogs: 231 Examples since I started these posts–but they’re really not sales pitches. I think it’s all the way up to twenty copies now. Woohoo!)

In the case of illustrations, the blogs in the survey have a fairly freakish pattern: To wit, of 3,662 illustrations used in all 231 blogs over the 92-day study period, more than half (1,975) were in one blog, leaving 1,687 or roughly seven per blog for all the others. The truly meaningless average (mean) is 15.9 illustrations per blog, but the median is all of one illustration.

Quintiles:

• Q1: Most illustrations: From 11 to 1,975 illustrations per blog.
  Average: 71.2 illustrations
  Median: 23.5 illustrations.
• Q2: More illustrations: From four to 11 illustrations.
  Average: 6.6 illustrations
  Median: 6.0 illustrations.
• Q3: Average number of illustrations: From zero to four.
  Average: 1.7 illustrations
  Median: 1.0.
• Q4 and Q5: No illustrations.

And, since this is particularly uninteresting data, let’s finish off the set (well, also because I have a more substantive post coming up later this afternoon, if all goes well): Illustrations per post.

Overall, the average (an average of averages) is 0.39 illustrations per post, with a median of 0.10.

First three quintiles:

• Q1: Most illustrations per post: 0.79 to 8.23
  Average: 1.36 illustrations per post.
  Median: 1.01.
• Q2: More illustrations per post: 0.24 to 0.78
  Average: 0.48 illustrations per post.
  Median: 0.45.
• Q3: Average number of illustrations per post: zero to 0.24
  Average: 0.11
  Median: 0.10

And that’s it.

Here’s the equivalent public library blogs post, for whatever commentary I provided then.

Even more so than total comments per blog, the blogs in Academic Library Blogs: 231 Examples lack extreme cases of high interactivity on a per-post basis: The highest is 2.2 comments per post. The overall average (an average of averages) is all of 0.12 comments per post–basically one comment for every eight posts. The median, of course, is zero.

Raw quintiles Q1 and Q2 (since Q3-Q5 are entirely zero):

• Q1: Most comments per post: 0.17 to 2.20 comments per post
  Average: 0.51 comments per post.
  Median: 0.33 comments per post.
• Q2: More comments per post: Zero to 0.17 comments per post
  Average: 0.08 comments per post
  Median: 0.07 comments per post.

And using the 86 blogs with at least one comment as the universe:

• Q1: Most comments per post: 0.50 to 2.20 comments per post
  Average: 0.91 comments per post
  Median: 0.67 comments per post.
• Q2: More comments per post: 0.25 to 0.45 comments per post
  Average: 0.33 comments per post
  Median: 0.31 comments per post
• Q3: Average number of comments per post (18 blogs): 0.14 to 0.25 comments per post.
  Average: 0.19 comments per post
  Median: 0.18 comments per post
• Q4: Fewer comments per post: 0.08 to 0.14 comments per post.
  Average: 0.12 comments per post
  Median: 0.13 comments per post.
• Q5: Fewest comments per post: 0.01 to 0.07 comments per post
  Average and median: 0.03 comments per post.

Two more (illustrations) and I’m done…

How many comments appeared on each academic library blog in Academic Library Blogs: 231 Examples during the 92-day study period (March 1-May 31, 2007)?

Once again, I’ll refer you to the equivalent public library blog post for commentary–noting that, once again, lots of the blogs don’t allow comments, quite often for entirely sensible reasons (e.g., some blogs are just posts of library schedules or new acquisition, using the blog form as an easy way to publish information with no attempt at community involvement).

That said… Where there are a handful of public library blogs that had lots of comments (three with more than 100), there are only two academic blogs with more than 40 comments during the period, and those two were in the sixties (61 and 66 respectively). Overall, there were a total of 575 comments (just less than a third as many as for public library blogs). That’s an average (mean) of 2.5 comments per blog. On the other hand, the median number of comments per blog is precisely the same as for public library blogs: Zero. While 118 of the 252 public library blogs had no comments, 145 of the 231 academic library blogs–nearly 63%–lacked comments entirely.

Here are the quintiles:

• Q1: Most comments: From three to 66 comments per blog.
  Average (mean): 11.2 comments per blog
  Median: Seven comments per blog.
• Q2: More comments: From zero to three comments per blog.
  Average: 1.3 comments per blog.
  Median: One comment.
• Q3 through Q5: No comments.

What happens if we restrict the quintiles to the 86 blogs that had at least one comment?

• Q1: Most comments: From nine to 66 comments per blog.
  Average: 21.2 comments.
  Median: 13 comments.
• Q2: More comments: From four to 9 comments per blog.
  Average: 6.4 comments.
  Median: 7 comments.
• Q3: Average number of comments: From two to four comments.
  Average: 3.2 comments.
  Median: three comments.
• Q4: Fewer comments: From one to two comments.
  Average:1.8 comments
  Median: Two comments.
• Q5: Fewest comments: One comment per blog.

Blue Steel, 1934, b&w. Robert N. Bradbury (dir.), John Wayne, Eleanor Hunt, George ‘Gabby’ Hayes, Edward Peil Sr., Yakima Canutt. 0:54.

As one-hour Westerns go, this is better than most. Sure, some elements of the plot are standard. The leader of the bad guys is the most prominent person in town: Check. The cute young woman winds up with the hero—even though, in this case, he really hasn’t talked to her except to rescue her once: Check. Despite the quick draw and sure aim of the hero, most fights are fistfights—and they’re incredibly phony: Check.

On the other hand, the plot makes more sense than most. A beleaguered town, Yucca City, is in trouble because shipments of supplies (and money) keep getting stolen, and the ranchers are about to give up and move out. At one key plot point, the Big Man offers to buy their homesteads for $100 each—and, of course, there’s a sinister reason. Naturally, John Wayne saves the day, with the help of a crusty old—not sidekick this time, but sheriff. Wayne is young, handsome, and quite effective. The long final chase sequence is effectively done; the long, largely silent opening sequence (a hotel in a really noisy rainstorm) is also surprisingly effective. Most of the acting is good. The sleeve description almost gets the plot right, but messes up one point big time: It has Wayne as “Sheriff Jake” hot on the trail of the man who appeared to rob a payroll. Actually, Wayne is the man who appeared to do the robbing (he’s a Marshal). The Sheriff is the crusty old coot (Gabby Hayes), “Old-timer” as Wayne consistently calls him. I’ll give it $1.00.

Santa Fe Trail, 1940, b&w. Michael Curtiz (dir.), Errol Flynn, Olivia de Havilland, Raymond Massey, Ronald Reagan, Alan Hale, William Lundigan, Van Heflin. 1:50.

Errol Flynn, Olivia de Havilland, a young (29), devilishly handsome Ronald Reagan. Costars like Van Heflin (in a key role). Historic names including George Custer (Reagan), J.E.B. Stuart (Flynn), John Brown (Massey) and many more. This is a big movie—big stars, big historical names, good production values, a major motion picture.

Ostensibly, it’s about the Santa Fe trail, bloody Kansas and building the railroad through to Santa Fe. Really, it’s about John Brown and the prelude to the Civil War—where West Point graduates who would later fight each other fought together to bring down Brown’s uprising. As a historical film, it’s a mess—pro-Southern/slavery, riddled with wild inaccuracies, etc., etc. You may find it unwatchable for that reason.

It’s dramatic, generally well acted and well filmed, including the long battle sequence near the end at Harper’s Ferry. The print’s OK—but the sound is sometimes distorted, bringing this down to $1.25.

McLintock!, 1963, color. Andrew V. McLaglen (dir.), John Wayne, Maureen O’Hara, Patrick Wayne, Stefanie Powers, Jack Kruschen, Chill Wills, Yvonne De Carlo, Jerry Van Dyke, Edgar Buchanan, Bruce Cabot, Strother Martin. 2:07.

The older John Wayne at his most entertaining in a big, well-made movie that’s mostly a hoot. If you don’t already know the movie (I didn’t), I’m not sure how to describe it. G.W. McLintock is a cattle baron(and miner) in the Mesa Verde of turn-of-the-century Arizona, a territory hoping to become a state. He owns most of the nearby town (named McLintock), treats his employees fairly, drinks a lot, plays chess and has a good time. He’s friends with the local tribes (despite an old battle wound) and mostly dislikes the territorial government people he considers incompetent—and, to be sure, homesteaders he thinks are being sold a bill of goods, asked to make a living on 160 acres of 6,000-foot-high land not fit for farming.

That’s just the setup. His estranged wife (O’Hara) shows up, asking for a divorce but mostly wanting to take her daughter (Powers)—just coming back from college Back East—away with her. McLintock’s having none of that. Lots of action ensues, including a rodeo, various romances, and much, much more. Big fight scenes, more slapstick than anything else—I don’t believe there’s a single injury or death in the movie. A combination of comedy, light drama and a little romance, the movie has fine performances by Wayne, O’Hara, Powers, Van Dyke (as an up-to-the-minute college boy with a Letter—in Glee Club), and most everyone involved, all of whom seemed to be having a ball.

I can’t figure out how this wound up on a set with mostly public-domain movies, unless the studio figured DVD buyers would want the wide-screen version so they could give the pan-and-scan away. The print’s OK—if there’s damage, it never gets in the way of the movie. The colors are a little faded, but that may be the way it was shot. Great fun, and at the end of more than two hours I wanted more. I’m sure it would be better in widescreen and with richer colors—but even so, I can’t give this one less than $2.25.

Sagebrush Trail, 1933, b&w. Armand Schaefer (dir.), John Wayne, Nancy Shubert, Lane Chandler, Yakima Canutt. 0:54.

The plot’s a little different, although as usual shootings only happen from a distance—up close, it’s all badly-staged fistfights. A young John Wayne is a convicted killer who’s escaped and is on the run (hopping a freight train bound west from Baltimore). He’s innocent, of course. He winds up with a good-sized gang of outlaws, hoping to find the real killer, which he does…but decides the real killer’s not such a bad Joe. Meanwhile, he’s trying to be part of the gang while foiling their big robberies, in one case by pre-robbing the stagecoach. All turns out fairly well in the end.

The print’s not great. The acting’s not great, but no worse than the run of these things. Some excellent stunt work. John Wayne underwater breathing through a reed. What the heck: $1.00

Another in a series of detailed metric summaries (oxymorons r us) on the 231 blogs in Academic Library Blogs: 231 Examples, which as far as I know is the only broad objective survey of academic library blogs.

I blathered on about the significance of average post length in the public library blog equivalent to this post. I won’t repeat that.

Once again, the average and median in each case is an “average average” and “median average” (or “average median?”)–that is, an average or median on a set of average lengths. You could determine an overall average post length by dividing all the words in all blogs by all the posts in all blogs, but that’s an unusually pointless exercise. (Right around 137 words per post, if you care.)

Overall, the average (mean) average length per post is 178 words. The median is 144 words.

Quintiles:

• Q1: Longest posts (“essays”): 235 to 897 words per post.
  Average (mean): 370 words per post.
  Median: 323 words per post.
• Q2: Longer posts: 164 to 235 words per post.
  Average: 194 words per post.
  Median: 190 words per post.
• Q3: Average-length posts: 125 to 162 words per post.
  Average and median: 144 words per post.
• Q4: Shorter posts: 94 to 125 words per post.
  Average and median: 109 words per post.
• Q5: Shortest posts (“terse”): 11 to 93 words per post.
  Average: 73 words per post.
  Median: 80 words per post.

Compared to public library blogs? Q1 and Q5 are quite similar; for Q2-Q4, academic library posts tend to be a little shorter (e.g., the median point for Q3 is around 94% of the median point for the public-library Q3).

Once again looking at the 231 academic library blogs included in Academic Library Blogs: 231 Examples, this time looking at total words during the three-month/92-day study period.

The complete set of posts total 852,930 words. The average blog had 3,692 words. The median was 2,244. Comparing that to public library blogs, the average academic blog was about 10% shorter–but the median academic blog was about 17% longer.

The quintiles:

• Q1: Longest blogs: 5,656 words to 39,000 words.
  Average (mean): 10,205 words.
  Median: 8,408 words.
  This group includes 55% of the words in all the blogs.
• Q2: Longer blogs: 2,978 to 5,652 words.
  Average: 4,181 words.
  Median: 4,268 words.
  This group includes 22.6% of the words in all the blogs.
• Q3: Average-length blogs: 1,733 words to 2,969 words.
  Average: 2,278 words.
  Median: 2,244 words.
  This group includes 12.6% of the words in all the blogs.
• Q4: Shorter blogs: 888 words to 1,716 words.
  Average: 1,300 words.
  Median: 1,338 words.
  This group includes 7% of the words in all the blogs.
• Q5: Shortest blogs: 69 words to 886 words.
  Average: 529 words.
  Median: 524 words.
  This group includes 2.9% of the words in all the blogs.

You’d need to take the hundred longest blogs–43% of the total–to include 80% of the words.

Comparing these to public library blogs by quintile, it’s a matter of gentler extremes: the longest academic blogs are shorter (the Q1 average and median are both lower), while the other academic blogs are slightly longer (that is, Q2-Q5 average and median are higher for academic than public library blogs).

No long-winded introduction this time. Here’s the comparable post for public library blogs. I used the same sample period and rules for Academic Library Blogs: 231 Examples: March-May 2007, blogs had to have started before 2007, blogs had to have at least one post in two of the three months.

In all, the 232 blogs included 6,229 posts, for an average (mean) of 27 posts per blog–about two per week. The median is 14 posts, just over one per week.

The quintiles:

• Q1: Most frequent posts: 34 to 762 posts.
  Average (mean): 83.4 posts.
  Median: 52.5 posts.
  This quintile includes 61.6% of all posts.
• Q2: More frequent posts: 17 to 34 posts.
  Average: 24 posts.
  Median: 23 posts.
  This quintile includes 17.8% of all posts.
• Q3: Average posting frequency: 11 to 17 posts. (The “extra” blog is here.)
  Average: 14.5 posts.
  Median: 14 posts.
  This quintile includes 11% of all posts
• Q4: Fewer posts: 7 to 11 posts.
  Average: 8.9 posts.
  Median: 9 posts.
  This quintile includes 6.6% of all posts.
• Q5: Fewest posts: 2 to 7 posts.
  Average: 4.2 posts.
  Median: 4 posts.
  This quintile includes 3.1% of all posts.

What percentage of blogs do I need to include for 80% of all posts (the Pareto number)? Quite a few–95 in all, or nearly 41%. That’s not surprising: After four blogs with 762, 468, 240 and 114 posts respectively, none of the blogs averages one post a day, and the number of posts declines fairly slowly.

Here we are at the end of the metrics for public library blogs (I’m not going to discuss “visibility” or how long blogs have been around): Illustrations per post.

Overall, the average blog in Public Library Blogs: 252 Examples had 0.72 illustrations per post; the median was 0.50 illustrations per post. The quintiles:

• Q1: Most illustrations per post: 1.0 to 12.8 illustrations per post.
  Average (mean): 2.19 illustrations per post.
  Median: 1.44 illustrations per post.
• Q2: More illustrations per post: 0.67 to 1.0 illustrations per post.
  Average: 0.87 illustrations per post.
  Median: 0.90 illustrations per post.
• Q3: Average number of illustrations per post: 0.25 to 0.67
  Average: 0.47 illustrations per post
  Median: 0.50 illustrations per post.
• Q4: Fewer illustrations per post: Zero to 0.25.
  Average: 0.09 illustrations per post.
  Median: 0.08 illustrations per post
• Q5: Fewest illustrations per post: No illustrations.

And that’s it. You can identify any blog in the book as to its proper quintile. If your library has a blog and isn’t in the book, you can play along.

Which libraries fit where? Well, for that you’ll have to buy the book–and, since I said I wasn’t pushing it by doing these posts, I suppose I should be gratified that there haven’t been any new sales of either library blog book. Or not.

Soon: The quintiles for the academic library blogs. Same metrics, different results.