Kudos sort on AO3

[petronia]

Good idea in the short run, bad idea in the long run -- not for social reasons, but because it'll probably make incumbent advantage more advantageous to the point of reducing the value of the peer review itself. I mean, there are forces that work against that too, and everyone has their compensation heuristics, but it's not necessary.

EDIT -- working it through... before this you had to sort by hits, then scan down the page for kudos/hits ratio, which is more indicative than either metric by itself (baseline also differs by fandom size, fandom maturity, and fandom who-the-heck-knows, but all this is heuristic). Geez I have to get off the computer though

Comments

[ayalesca]

HAHA YES THIS I had all this worked out somewhere in a text file and held a long discussion with J but IIRC it comes down to this sort of thing is far more "accurate" if you have to fork over one bitcoin or whatever per kudos.

I've found that the hits/kudos ratio *of the first 100 hits* (I had too much free time that week) is a decent gauge, actually, but small sample size and where the fuck am I going to obtain that statistic in the future.

Also multi-parters, unless they were completely loaded within the span of one day, have inflated hits relative to one-shots for obvious reasons. XD I have not worked out whether you could just divide by the # of chapters for a better approximation. (Though I suppose that could be changed by treating hits the same way they treat kudos, ie, tied to IP or MAC address. But that's probably not a true hit count anymore, so, feature or bug?)

And yes the entire thing goes out the window if the fandom is still heavily LJ based, Star Trek I'm looking at you.

[ayalesca]

Forgot to add -- I think sorting by comments is even more wildly variable than kudos and/or hits. IMO # kudos and/or # hits -- compared against # comments -- track slightly better with, uh, some nebulous measure of quality popularity.

(Once again the only thing I can say for sure is that a multi-parter is more likely to have comments than a one-shot, given two fics of roughly equal appeal to the same demographic, if only because people are more likely to ask "what comes next" than to say "thanks, good job", since leaving kudos has replaced that act for a large % of people. But you can't leave multiple kudos, so if you've been following a 20-chapter fic, people who want to leave SOMETHING at the end start using the comments for that. In my model, anyway XD)

[petronia]

Yeah, as far as I'm concerned sort by comments is useless, except as an extremely fuzzy measure of "controversialness." Like, ppl are more likely to leave a comment if there's a trick ending, or memorably kinky rarepair sex**, something like that. XD Not to mention, the comment count could just be conversation threads between the author and her readers. And your point about why multi-chapter fics attract more comments is exactly right (also, the longer the fic is, the more complex the plot/characterization -- hopefully -- to the point where you can have a conversation about what's going on in the fic, rather than in the canon. Cf. something like Captive Prince where you have 400 ppl leaving walls of text on each chapter).

** OTOH if the story is pr0ny, especially if it's trashily pr0ny, ppl will kudo but they won't bookmark publically (private bookmarks don't show up even to the author, I guess?), because that makes a statement on your own profile. XD Case in point, those two Avengers stories I wrote. Two chapters each, similar wordcounts. The pr0ny one has twice as many hits and more than twice as many kudos, but it has about the same number of public bookmarks.

[ayalesca]

I generally feel bookmarks are only useful as a measure of how well the fic serves as comfort reading XD There's lots of fantastic works that people read/watch once and never again because why get punched in the feels twice? OTOH stuff that indulges one's kinks, narrative or otherwise, is the kind of thing people pull out at 3 AM when they're drunk or whatever. Not that, um, I would know. Because I privatize my embarrassing bookmarks, too. XD

[lacewood]

Actually the existence of private bookmarks WILL show to the author but you can only tell by comparing the numbers between your statistics and the public record, so to speak, so this is probably more trouble than it's worth. XD (I only realised this because of the discrepancy on one of my recent stories)

[petronia]

Like, I always look at all these numbers together (not to mention the "real" metadata), because they do tell me in a holistic way whether the story is out there, if it's id fic, if it's arty/tasteful, if it does its own complex thing etc.

[ayalesca]

Agreed!

A hypothesis: I posit that all these things together work the best in more self-aware* fandoms because they understand both that they are describing themselves and how to best advertise themselves -- for example in Avengers and Trek, I've generally had a pretty good feeling from the summary alone what is coming my way. (I was realizing the other day that "median age of fannishly active people" tracks pretty well with "how rage inducing will this fandom be for me" XD )

* ... ... ... nm maybe? Look at that again, maybe I just mean "in fandoms that are more likely to be plotty" XD so you can tell the quality of the plotting from the summary because the authors are more predisposed to GAF about it. But in fandoms like Sherlock, which is 99% feels without plot, the summary is fairly uninformative to me unless it tells me there will be sexual porn. XD