Author popularity shown

I've added a "popularity" number to author pages. As with works, #1 is the top author (Rowling). An author with the popularity of 23,100 has 23,100 that are more popular than him or her. If #11 and #12 are tied, they are both called #11 and there is no #12 at all. (So, at present, everyone with one book has a popularity of 1,419,744.)





I am, however, currently, only showing the field for authors with more than 10 members/copies. This is because the data is of necessity cached, so it won't update instantly if someone adds three books, taking an author from 1 copy to 4.

I was hesitant to put this out there. Authors who use LibraryThing may find it competitive. But it's interesting enough and the data was just sitting there.

Data is now semi-cached. Basically, it won't seem cached.

By induction, if 'an author with the popularity of 23,100 has 23,100 that are more popular than him or her.", then Rowling, author #1, has one that is more popular than her

Hmmm. Anonymous, maybe? :)

Authors who use LibraryThing may find it competitive.

Well, yes, there is that of course. And sure I'm somewhat disappointed - but not surprised - that I'm the only one on LT having any of my publications.

In other words: go ahead. It's informative - I'll get over ranking very low ;-)

I love your author picture! :)

Have some bubbly. ;-)

Who are the top 10 authors? I'm guessing Stephen King is #2.

You'd better have some more too Matt, to cheer up mate!
I'd love to buy your book(s) to add membership numbers but they are, you must admit, rather specialized.

(You can still smoke in restaurants there?! Must visit again!)

I've actually got a handy page for it now:

http://www.librarything.com/zeitgeist/authors_list

The higher the number (so, low ranking) the better for me - I prefer the hidden types, not the ones that everybody's talking about, hehe

9> Neither Amazon(uk) nor BD knows about them.

> 9: You can still smoke in restaurants there?!

Ah no, that picture is a couple of years back. Being a heavy smoker I have some difficulties with the new situation, but having some friends who are asthmatics I fully support it all the same. But, anyway, no more...

The top ten is interesting. C S Lewis really does NOT deserve to be at No.5, and, curiously, if you look at Dan Brown (1)'s page, his popularity is said to be 366,546.

I've never read any Neil Gaiman (and that's OK) but should I try Nora Roberts?

>14 abbottthomas: " C S Lewis really does NOT deserve to be at No.5"

You're wrong.

>15 cpg: Well, of course I'm wrong because that's what the stats say, but I am still of the opinion that his Narnia Chronicles (which presumably get him to No.5) are a lesser body of work than, say, Roald Dahl's books. In that, I am not 'wrong', merely have a different opinion from yours.

ETA I am a fan of A Grief Observed, but his Perelandra books are not up to much in my view.

14> She writes fairly solid romances and/or mysteries. Not really thrillers, as I can still read most of them.

16> I would tend to agree that Dahl is a better writter, both for children and for adults (if somewhat gruesome at times). On the other hand I am a lot more likely to reread Lewis. Apparently I have 10 books by Lewis catalogued, and 4 by Dahl.

>16 abbottthomas: Thanks, MJ, maybe I'll give Roberts a try.

>14 abbottthomas:
C S Lewis really does NOT deserve to be at No.5

Nor indeed should he be at no. 5, if popularity is measured in members, which I guess means copies of books:

3. Terry Pratchett (314,712 members)
4. J. R. R. Tolkien (245,743)
5. C. S. Lewis (224,288)
6. Neil Gaiman (260,978)
7. William Shakespeare (199,751)
8. Nora Roberts (184,720)

Surely Gaiman should be no. 3, with Tolkein and Lewis each moved down one place?

Is it still rolling out? I see the stat on my author page, but it isn't on the @ibonewits author page.

>16 abbottthomas:

I've no desire to defend the Narnia books, but the rules are what the rules are. If you don't think the list should be based on popularity, then don't focus on his most popular works. Did Dahl write anything better than Till We Have Faces or A Grief Observed or The Great Divorce? I don't think so. And Cambridge University Press is reissuing a truckload of Lewis's scholarly works next month in its Canto Classics series.

You're wrong.

Wrong or right, it's true that popularity is based on the number of books not—say—the number of members. So it favors prolific authors. Each Narnia book counts. This advantages Lewis or say Daniel Steel, and disadvantages Homer.

In that, I am not 'wrong', merely have a different opinion from yours.

We all agree with that merely by participating in a site like this. Still, you're wrong :)

Nor indeed should he be at no. 5, if popularity is measured in members, which I guess means copies of books:

Some oddness there. I'll look at it.

While you're looking, the popularity number of one of my favorites (David Weber) seems to be very different from my other favorites. I would have expected them to be much closer in popularity.

Who the heck is Nora Roberts? Honestly - never even heard of her (of course I do live under a rock) ;-)

From Wikipedia:

Nora Roberts is an American bestselling author of more than 209 romance novels. She writes as J.D. Robb for the "In Death" series, and has also written under the pseudonyms Jill March and for publications in the UK as Sarah Hardesty.

Nora Roberts was the first author to be inducted into the Romance Writers of America Hall of Fame. As of 2011, her novels had spent a combined 861 weeks on the New York Times Bestseller List, including 176 weeks in the number-one spot.

I think she's one of the top-selling romance novelists or something? She's extremely prolific and something of a bad-ass, according to what I've seen of her blog.

Nora Roberts is a very popular and prolific romance and detective writer, at least here in the USA.

>22 timspalding: Still, you're wrong :)
Do you mean that you think the Narnia books are better than Dahl's stuff? Really?

>21 cpg: The authors are rated by 'popularity' meaning, so we are told, the number of books - not necessarily the most appreciated or whatever. I was surprised to see that there are more copies of the Lewis works you list on site than any of Dahl's books bar his top 5 or 6. There's no accounting for taste ;-)

>24 Nicole_VanK: My reaction, exactly.

#27 by @abbottthomas> Considering cpg listed three books non-Narnia Lewis books, I think folks may be considering all the authors works rather than a subset.

So if it's semi-cached, when will I notice it for James Branch Cabell?
http://www.librarything.com/author/cabelljamesbranch

>28 brightcopy: My point, really. This hasn't got much to do with what folks 'consider' but with the books that they buy and list. Even Lewis's less 'popular' works out-score most of Dahl's. I can live with being in a minority ;-)

#24: Every one of her new books can be found everywhere that still has a rack or two for new books (Wal-Mart, CVS) and she's put out 38 In Death books in 15 years, plus lead author on collections of 4 novellas, plus she's doing non In Death books. A very large reader base combined with an extremely prolific author puts above any author that doesn't have either one, and almost any author that only has one or other. Same way Stephen King, Terry Pratchett and Agatha Christie got there; lots of books with lots of readers. J. K. Rowling and J.R.R. Tolkien are impressive for getting on there on the basis of a handful of books.

Given my compulsive need to numerify everything...

Authors in popularity order, by number of works over a threshold


Author 15k 5k 1k
Rowling 7 11 14
King 0 35 67
Pratchet 1 35 55
Tolkien 6 8 23
Lewis 2 12 28
Gaiman 3 11 37
Shakespeare 2 12 30
Roberts 0 0 76
Christie 0 2 75
Austen 3 6 8


So Rowling has the most works over 15k (a little past Tolkien), King and Pratchett tie for 5k, and Roberts narrowly edges out Christie for most works over 1k.

What I find most interesting is that Pratchett is number 3 with Good Omens and Gaiman is number 6 with American Gods.

So either Pratchett has more overall books than Gaiman, but Gaiman's most popular work outdistances Pratchett's by a huge margin, or an awful lot of people list Pratchett as the primary author for Good Omens, as opposed to Gaiman being the primary author. Considering they co-wrote it could go either way, but I'd expect a much closer split to 50/50 than what it looks like.

This is for all their works; Good Omens just happens to be Pratchett's biggest work, and American Gods Gaiman's. All the copies of Good Omens are combined, so it doesn't matter how many are listing Pratchett as primary author as opposed to Gaiman. I assume they're both getting counted for Good Omens, as Gaiman is listed as a main author for all copies. (American Gods has more copies in the system then Good Omens does, which is why it gets listed before Good Omens for him.)

Pratchett has 35 5k+ works in the system, and 55 1k+ works, compared to Gaiman's 11 and 37. Put another way, Gaiman has written several very popular books, but Pratchett has written a 40 book series where almost all of the books of that series have more then 5k copies in the system.

22: it's true that popularity is based on the number of books not—say—the number of members.
Oh, I thought that it was based on how many members had their books on their collections.
Then if it's based on number of books, it's not necessarily linked to LT, is it? Could be done by any database that has all books written by all authors...

#35: It's by the number of copies of books, not the number of books in the system. Rowling is on top with relatively few works in LT, but 60,000 copies held of her big ones.

Ah ok, I misunderstood Tim, then.
Or rather he wasn't clear - it does depend on how many members have got each author's books.

I'm seeing this on some but not all author pages, seemingly on the more "popular" ones. Is it going to roll out for the less "popular" ones eventually?

38 > It was on my author page almost immediately, but not on my husband's author page. He was far more prolific and popular than I am, so apparently the roll-out is based on something besides sheer popularity.

Either that, or Tim likes me. Or wanted to humble me. Either one.

Do you mean that you think the Narnia books are better than Dahl's stuff? Really?

Yeah, we've been on a Dahl kick recently with my 7 year-old—Charlie and the Chocolate Factory, Matilda, James and the Giant Peach, The Twits, Fantastic Mr. Fox, and (my favorite) Danny the Champion of the World. My wife loves him more than I do, although I'm very fond of the last two (Fox and Danny). While my son has enjoyed them, he's also protested at the general level of inhumanity in some of them, and I worry it's coarsened him. Dahl takes a lot of joy in bad things happening to bad people. That's a staple of children's literature, but he takes it to an extreme.

Put another way, Dahl is fun, but he's also a hater, and I think that shows. A kid reading The Twits learns that it's funny when bad people suffer repeatedly and then die. In that and others, my son kept looking for the good in people--for example in Matilda's parents. It wasn't there. He creates bad characters in order to have them torture the good until such time as the wheel turns and they can be tortured. That's not a vision of the world I want my kid to have.

Ultimately, I find Lewis better on many levels. Lewis wasn't a hater. Some bad characters suffer, but there are also a lot of redemptions--Edmund and Eustace, for example, act like jerks and come to realize it. I don't think it's restricted to Christians by any means, but that's apiece with Lewis' essentially Christian worldview.

#19, #22: I think #33 obliquely pointed out at least part of what's wrong there. I expect that Neil Gaiman is not counted for Good Omens in the popularity because he's a main author, not the primary author. That's only 19,000, so that doesn't seem to explain everything.

Or as a "better cover" put it: "It's okay if giant fruit kills your aunts as long as they were bitches."

http://janeausteninvermont.files.wordpress.com/2011/08/book-cover-james-peach-be...

40> But Edmond and Eustace don't act like jerks. This is a discussion for elsewhere, I'm sure, but with Eustace, at least, the only thing he's done wrong is be brought up by vegetarian parents that taught him to use their first names to address them, to sleep with the window open, and to prefer non-fiction books to fiction. He does treat the Pevensies poorly at times, but it's not really a jerk thing to want to be included in secrets when your cousins are staying at your house, and then to be annoyed when all they do is talk about their imaginary country, and don't include you. And then he gets sucked off to this strange, foreign place without any warning and everyone expects him to just go along with it, even though he's not really given any sympathy or kindness for being a nine-year-old boy who is seasick and homesick and doesn't know anything. He's treated as "bad" because he doesn't believe in a talking, magical lion, despite having never been given any reason to do so, other than Lucy and Edmund's imaginarion games (and why shouldn't he believe that it's all imaginary?).

I've always felt badly that Eustace was given short shrift, the poor kid. I loved Dawn Treader because I liked the adventurous Odyssey stuff, but I hated it because everyone was so unfair to Eustace.

Two books, one LibraryThing: parallel Narnia/Dahl reads. Easy to get copies, short enough that participating wouldn't be a huge time commitment, and apparently lots to talk about.

44>

More like ten or twelve books, one LT, what with seven Narnia books and a handful of Dahl's best-known.

I wouldn't want to start off that feature with a giant holy war about the proper reading order for a series where the publication and internal chronological orders don't coincide, though.

(Link for those who don't know what _Zoe_ and I are talking about.)

Another anomaly: Douglas Adams is listed as having 112,248 members and has a 564,770 popularity. The 20,000 copies of HHGTTG alone should put him higher then that.

> 40: Okay, get your point - fair enough. And admittedly: I haven't read all of either author. But, the way I see it: Lewis gets boring (Magician's Nephew, Last Battle - come on...) while Dahl doesn't.

To each his/her own though.

>45 lorax: I still think multiple very short books (say, 4) would invite more participation than one very long book, because it would be easier to take part in at least some of the discussion. But I'm not actually wedded to the idea of these particular books :)

>43 keristars:

Kudos for sticking up for him. Dahl would have tortured and killed him. But Eustace certainly does behave poorly to many in the story, as well as selfishly and cruelly. (He misbehaves in a way children misbehave—they're not deeply bad, they're just trying it on; if they don't learn about it, they'll eventually settle into it.) Does he have some cause? Sure. That's how we behave badly--it's rarely when everything is going swimmingly. Lewis understands what makes people treat others poorly in a way Dahl does not; and he understands what makes us reverse course and NOT treat others' badly.

Okay, I'm moving the Lewis/Dahl discussion—if people will permit me—to here:

http://www.librarything.com/topic/160018

I was babysitting for a friend once, years ago, and saw a book of Dahl's adult short stories lying around ... I didn't sleep much that night, what with the nightmares, I would never again read more than one of his stories at a time.

46
Douglas adams (1) seems to have been given the same popularity as Douglas adams (2). In the page linked by Tim in #10 above he ranks 22nd

#1: If #11 and #12 are tied, they are both called #11 and there is no #12 at all.

Right now, http://www.librarything.com/zeitgeist/authors_list says that Agatha Christie and Jane Austen are tied, and that they are each at #10, following #8, whereas you said they should be #9.

The very fact they're tied is probably a bug. The member numbers are pretty different, and the odds two authors with 100,000+ books have exactly the same number of books is pretty unlikely.

So, there's an interesting issue here:

1. Author pages count up the members (which should be labelled "copies") based on all the authors works, including ones they're a secondary author for, or are only the author for "some editions." That how Gaiman gets to 260,696—60k more than it would be without collaborations, introductions, etc.

2. The author list ( http://athena.librarything.com/zeitgeist/authors_list ) only counts primary authors.

How should this be dealt with. It seems wrong to proclaim that someone has 20,000 copies just because he introduced some popular title, or even introduced one edition on a popular title. At the same, it would be odd if someone who did an introduction had a member number of 0.

I'm hesitant to give Gaiman a popularity of 200,000 / 260,000 or whatever. Any ideas?

At the same, it would be odd if someone who did an introduction had a member number of 0.

Why? They didn't author any works, they just introduced them. You're not measuring "introducer popularity."

Also, for "some editions" author roles the number of copies is meaningless. I'd prefer no number to a meaningless number.

If they're a main author on all copies, it should count. Gaiman did those collaborations and is listed equally with Pratchett; why shouldn't they count?

Why? They didn't author any works, they just introduced them. You're not measuring "introducer popularity."

Well, you'd have no good measure of that.

If they're a main author on all copies, it should count. Gaiman did those collaborations and is listed equally with Pratchett; why shouldn't they count?

Yeah, I agree with that.

Okay, how about if the main-author number is the main number. If the secondaries and partials amount to much, it'll list them in parens, or something?

So, for example, well-known audiobook readers would all be members 0?

#60 by @timspalding> I just don't get how that's "author popularity"...

If you want to come up with a general purpose role-specific popularity feature, more power to you. You could list popularity for "Narrator", "Introduction", etc. But you're making an "author popularity" and I just don't see how it's germane to count someone who read someone else's book.

As I mentioned in another thread, I once bought a copy of The Wind in the willows just for the cartoonist who illustrated that particular edition. Should Harry Hargreaves be in any way associated with that popular book? Yes, as an illustator only. But his contribution shouldn't count in any way to the books standing/popularity.

Primary and main all. Those are the authors.

Unless you want to get fancy and say each copy only gets counted once, so works with a primary and two main alls get divided and a third counted to each of those.

>60 timspalding:
So, for example, well-known audiobook readers would all be members 0?
I would prefer a blank, a NA (not applicable) or some other short message to a 0, but yes.

The problem is that the audiobooks with that specific reader are combined with all other editions of their respective works. The available numbers have no meaning with respect to "some editions" authors. Unfortunately, there are no edition specific numbers to work with.

Also:
I agree with those above who say that main author, all editions should definitely be counted.

I'm ambivalent about secondary author, all editions. One the one hand, it seems that if counted at all these should count less than fully authored works. On the other hand, it is interesting if some author has been included in several popular anthologies and I suspect that the counts on these works are generally smaller so they wouldn't distort the overall counts too much. I would be interested to see cases of authors for which counting the secondaries would produce unexpected and undesirable results.

I see from Terry Pratchett's author page that his popularity is rated at 3: a new cell in the one-row table at the top of the page.

James Branch Cabell still has no such cell in his table. I understood (vaguely) from Tim's post 2 that the change should be available to all authors (semi-cached, whatever that means). Why don't I see it for Cabell?

ETA Same for Howard Pyle, seems I could go on and list many authors. Both have >10 timspalding: members for a copy, so that's not it. I fear I just mis-understand how and when this rating applies.

I agree with the others that only "Main Authors" for "All editions" should be counted. That's occasionally a tiny bit unfair on some translators (i.e. Seamus Heaney's Beowulf), but overall it's the best decision, I think.

Looks like the first figure marks the number of copies as main / primary author, and the second includes secondary. See: http://www.librarything.com/author/dematonscharlotte for instance.

>69 Collectorator: It is starting to look like a stock marker ticker on some author pages.

It's now counting in two ways.

The main way is primary author of all editions of a work, including joint authorship, so Gaiman gets points for Good Omens too.
The secondary way is for that and secondary authorship and authorship of some editions, giving Gaiman points for introductions to books.

If they two differ, it shows the latter in parentheses:


The Zeitgeist: Author popularity goes by the former alone:
http://www.librarything.com/zeitgeist/authors_list

>69 Collectorator: It is starting to look like a stock marker ticker on some author pages.

Yeah, I know.

There have to be definitions.

#73 by @timspalding> giving Gaiman points for introductions to books

Was there anyone that this made sense to other than you? I'm baffled.

I don't understand why the information about introductions is important enough to show in the top bar.

Because it could be anything. It could be illustrations, for example. We have primary and secondary. We do not have 100 possible roles. If we're going to give credit to people who write stories or illustrate books, we have to give it to people who write introductions.

>78 timspalding: So don't give credit to people who write stories or illustrate books. At least not there.

78>

What _Zoe_ said.

Yeah, I just don't see how it has anything to do with popularity. I mean, if you did it just for illustrators, maybe. But I'm not buying a book because it has an introduction by Neil Gaiman. I'm buying it because it was written by the person who wrote it.

To me it's just noise added to the signal. It measures something, sure. But what it measures is meaningless.

But I'm not buying a book because it has an introduction by Neil Gaiman. I'm buying it because it was written by the person who wrote it.

Might you buy a book because it has a story by Gaiman?

>84 _Zoe_:

Flawed. It counts both, X (Y).

>85 timspalding: What's the flaw? Some of us think that the current situation is not ideal and want the extraneous information to be eliminated/moved elsewhere.

This is like the stupid OPD issue all over again. The most likely resolution is probably that you'll eliminate both numbers in order to maximize unhappiness.

So, the "Members" field is a bit of a misnomer, right? It's really "Total books cataloged", isn't it? Or am I missing something?

>88 saltmanz:

Yeah. I think we rename it "Copies." At one point it was members per se--and that's also interesting, but it's a lot harder to calculate on the fly.

The main way is primary author of all editions of a work, including joint authorship...

I'm confused about the meaning of primary, main, and secondary author. In the Edit Other Authors dialog it has Primary? as the label for the line where you can choose either Main Author or Secondary Author. So is that saying that if you think an author is Primary you should check the Main Author radio button? So Primary and Main mean the same thing?

Yeah, sorry. Terminology. I should tweak that.

#83 by @timspalding> Might you buy a book because it has a story by Gaiman?

But that's not what you're measuring! You're counting both books with stories by him, introductions by him, forwards by him, narration by him, afterwords by him, edited by him, etc. etc. I've said this over and over again in this thread. If you actually broke it out into popularity by role for secondaries/some editions, it would have meaning. But instead you've agglomerated it all into one single number which no longer has any real significance.

Long argument short: you've just provided us with the average of three apples and two oranges. The answer? Chocolate.

(Because chocolate is always the answer.)

85>

Yes, we all understand that. That's why the poll says "should show ONLY" X, rather than "should show" X.

This is a vote about whether we want only pepperoni on our pizza after you've added anchovies, not about whether we want pepperoni there at all.

I'm not so troubled by apples and oranges counting. But as I said above, counting "some editions" authors makes no sense. The count for all copies of the work has no necessary relationship to the popularity of the editions of the work that the "some editions" author has contributed to.

94. I'm with eromsted. If you're going to count in copies, then the data counted should apply to all copies.

I also think that counting all names that apply to all copies makes the most sense. If you only do primary author, it only takes an edit on the work page from a disgruntled Gaiman fan to take all the Good Omens copies away from Pratchett and give them to Gaiman ....

95>

And a Pratchett fan or a combiner will quickly add his name back. I don't think the risk of that is nearly enough to warrant the random and bizarre inclusion of people who wrote an introduction or an afterward as top-level authors.

Particularly if they are listed as only some copies.

Could you move the bracketed number to the author's ALSO BY section?

>98 casaloma: That sounds good. I like the idea of having the information somewhere, I just don't think it's top-of-the-page material.

No, I don't think it's just Narnia. It is the entire body of his work. I read all 3 genres in which he wrote, fiction, literary works and Christian apologetics. All are topflight. "The Discarded Image," his study of medieval literature is still considered one of the top works in that field more than 40 years after its publication.

Please excuse this hasty post. I have to have office hours tonight and cannot stop to verify dates and subject matter. But Lewis certainly deserves #5 for many reasons.

We have several zero popularity authors topping the author list now; http://www.librarything.com/zeitgeist/authors_list now has Charles Timmerman at the top, not J. K. Rowling.

Nice. Looking.

Got it…

The earlier situation was that the number was for "everything this author was ever involved in - including things where it applies only to some editions". So giving a separate number for primary/main authorship is an improvement. (Personally I don't have any problem with the numbers showing where they do. I wouldn't oppose having the larger number moved elsewhere on the author page, I would mind having it removed altogether).

Something just jumped out at me from the popularity; the six most popular authors are all fantasy. I'll give you quibbles about Stephen King (who's fantastic horror) and C.S. Lewis (who's on the list because of fantasy, but would drop below Shakespeare if you took out Mere Christianity and the Screwtable Letters, etc.) The pattern seems to break up after that; going out to 25, there's three more fantastic horror (/romance) and one fantasy author. Still, quite a heavy load. Comparing it to book popularity is interesting, as that's more balanced once you drop the Harry Potter.

Any chance we could have an option for selecting the zeitgeist author popularity list by author nationality (CK "Nationality" on author page)?

As in the "Trends by language" feature?

I don't know if this is easily doable, but it would be really interesting to see just the authors we have in our collections. I know I can do it author by author to see popularity, but it would be great to be able to filter the Zeitgeist list by our authors.

I'd just like a check mark next to the top book of the author that is in my library.

> 29, 67

The metric appears to have propagated down, it's now appearing for Cabell, Walt Kelly, and others.

My wish list: to see an "median / mean author popularity" by collection, including for my complete library. Similar to the median / mean book obscurity metric currently provided under stats / memes. And I'd be curious to learn whether that metric is correlated with (a) average size of collection entered on LT, (b) average activity / participation on talk, or (c) contribution to specific aspects of LT (such as entering / editing Local events and venues, posting reviews).