Log in

No account? Create an account

Previous Entry | Next Entry

All the King's Horses...

Last year at Sasquan the worldcon business meeting passed two proposals to change the voting procedures for the Hugo Awards, to deal with the problems posed by slating. WSFS rules require that a change be passed by two successive worldcons before it takes effect, however, so both 4/6 and EPH will be up again for vote at MidAmericon II in August.

The 4/6 proposal is pretty straightforward. At present there are five finalists in each category, and each voter gets to nominate five choices for those five slots. 4/6 would increase the number of finalists to six, and simultaneously decrease the number of nominations allowed each voter to four. The theory being that a slate voting lockstep might take four slots, but not the whole category.

EPH, which stands for E Pluribus Hugo, is considerably more complicated, and I will not attempt to describe it here. It was designed by mathematicians and voting theorists, and will supposedly prevent a small disciplined minority from taking all the slots on the ballot. There's been plenty of discussion and debate about EPH all over the internet.

Most recently, the designers of EPH have done a test run to see what impact the system would have had on the latest ballot. The results, and a spirited discussion of same, can be found over on Mike Glyer's FILE 770, here: http://file770.com/?p=28946#comments

((For those of an academic and mathematical bent, the hard crunchy bits are here: https://www.schneier.com/academic/paperfiles/Proportional_Voting_System.pdf ))

From where I sit, the EPH results are not very encouraging.

Over the past few months, I've read countless variations of the statement that goes, "well, this is the last year we will have a problem, come summer we'll pass EPH and all will be fine." I had my doubts about that every time I heard it, and this new report just confirms them. We may indeed pass EPH, and it may help... a little... but all will not be fine.

We may pass 4/6 too, and that could also help... slightly... but it's easily thwarted, if you have hundreds of followers who will do exactly as you tell them, and the Rabids seem to have just that.

If EPH and 4/6, or both, are passed at MidAmericon II, and work more-or-less as advertised, the slates will no longer be able to completely dominate entire categories by taking all five slots. The reforms should ensure that there are at least one or two legitimate nominees in every category. Which is better, certainly, than what has happened to Best Related Work the past two ballots, say. But it is still far from ideal. Future ballots will instead look more like last year's Best Novelette, Best Professional Artist, and Best Fan Writer shortlists, or this year's Best Fan Artist, all of which featured one legit choice and four slate candidates. Maybe we'd see some improvement in some categories, and have two finalists to choose between.

Better than what we have now? Sure. But comparable to being able to choose among five strong candidates to decide which one was the very best of the year? Not even close.

I can hear the proponents of EPH and 4/6 saying their reforms were never meant to be a cure all. Yes, I know that, I never believed otherwise, and I applaud your efforts to help. I just wish these reforms helped more. Neither EPH nor 4/6 is going to prevent us from having VD on the Best Editor shortlist from now until the heat death of the universe.

And I also know that there are now other proposals out there, proposals that call for three-stage voting, for negative votes and blackballing, for juries. Some of these cures, I fear, might be even worse than the disease. We have plenty of juried awards; we don't need another. Three-stage voting, with fifteen semi-finalists that get boiled down to five finalists and one winner? Maybe, but that considerably increases the workload of the Hugo administrators, whose job is hard enough already... and I fear it would actually ratchet up campaigning, as friends and fans of those on the List of Fifteen rallied around their favorites to get them on the List of Five. And a blackball round, voting things off the ballot? Is that really a can of worms we want to open, in this present climate? That would dial the ugliness up to eleven, I fear... or higher.

Sadly, I don't think there is an answer here. No magic bullet is going to fix this. And I fear that the people saying, "pretty soon the assholes will get bored and go away," are being hopelessly naive. The assholes are having far too much fun.

A year ago April, when Sasquan announced the ballot, I wrote the Hugo Awards had been broken, and might never be fixed. A lot has happened since that time, and from time to time I've allowed myself to think that there was a light at the end of the tunnel, that this too would pass. Now I am starting to fear that my first reaction was the correct one.

The Hugo Awards have always been an occasion for joy, for celebrating excellence and recognizing the best among us. That's what we need to get back to. But I don't see how.


May. 17th, 2016 04:18 am (UTC)
Actually, I've seen estimates that we would need 40,000 voters in order to counteract a group of 200-400 dedicated slate voters. That's because of the "long tail" of nominations. Just look at Best Novel last year: 1,827 people cast nominating votes. They nominated 587 individual works. It took only 212 votes (11%) to make the ballot. The 15th-place work was named by 4.8% of the voters. That means that the remaining 572 works were spread out in a long, flat distribution, with the average work being named on only three ballots. (I don't know how many works got one vote each, but I suspect it's a very large proportion of the total.)

Increasing the turnout doesn't actually help that much, as long as most of them vote "honestly," instead of joining one political party or another in order to concentrate their votes to some effect the way the Rabids do.
May. 17th, 2016 05:55 am (UTC)
I question those numbers. If it really took 40,000 voters to counter 400, we would not have any real nominations in Best Novel or Best Dramatic Long either.

Yes, 587 novels might have been nominated, but many by only one or two people. Our field has grown, but it is not that big. Even now, there cannot be more than thirty novels in any given year worthy of serious consideration. It's really no different than the Oscars. There were 684 films released to theatres in 2014, 68 of which grossed more than $50,000,000... but how many were serious Oscar contenders? Maybe thirty.

If those 1900 voters who nominated for Best Novel had also nominated for Best Related Work, would they have nominated hundreds of different related works? Maybe... if they could find that many. But enough of them would have found things like the Felicia Day book, THE WHEEL OF TIME COMPEDIUM, and Adam Whitehead's history of epic fantasy to get those on the ballot, I'd wager.
May. 17th, 2016 02:00 pm (UTC)
But my point is that with the current open nomination process you will never get the voters to concentrate their votes among the top 30, unless someone were to go through and pre-screen the field. (The Locus Awards do this with their Recommended Reading Lists, which seed the Awards ballot, and while write-ins are permitted, they rarely if ever make the cut.) When you flatten out the distribution like that, even the most-popular works are unlikely to get more than around 20% of the nominations. That means slates will always be able to prevail.

The only other counter-move to a procedural change that discourages Griefers to go away is the formation of rival slates, which would IMO be just about as bad as the current problem.
May. 17th, 2016 04:27 pm (UTC)
I don't think the facts support your contention "you will never get the voters to concentrate their votes among the top 30." That is EXACTLY what they do in Best Novel. This year and last enough voters concentrated on a few books to prevent the slates from sweeping that category.

If the voters can do that in novel, they can do it in all the categories, if they would only bother to nominate. The number of novels being published each year is not significantly smaller than the number of short stories... and it is in fact much LARGER than the number of novellas or related books.
May. 17th, 2016 04:33 pm (UTC)
I don't think we're going to be able to agree here. Unless you're advocating for the formation of rival political parties to push slates of Good Stuff, which almost certainly you are not actually suggesting.

I think what we're seeing in practice is that increasing the turnout simply flattens out the long tail, and may even make it easier for factions to sweep the ballot in some categories, given that they'll not actually need more than about 10% of the nominations to show up in the top five positions.

Edited at 2016-05-17 05:03 pm (UTC)
May. 17th, 2016 07:25 pm (UTC)
Again, the numbers do not bear out that assertion.

3900 votes for Best Novel may have produced a long tail, but it also produced several legitimate nominees, both last year and this year.

2000 votes for Best Related Work produced a Rabid sweep. Bring in those other 1900 members, and you might get a longer tail, but you would also get several legit nominees on the ballot.

I am baffled that you cannot see that.

Last year at Sasquan, twenty-one votes was all Patrick Rothfuss would have needed to make the novella ballot. Only 1083 nominated in novella, as opposed to 1823 in novel. Thirteen addition votes would have gotten John Joseph Adams on the Best Editor ballot in place of VD. Twenty-one would have placed John Picacio on the Best Pro Artist shortlist. And that was LAST year, when there were half as many nominators as this year.

You don't need 40,000 additional votes; you just need the fans who nominated for Novel to fill out the rest of their ballot, the way the Rabids do.

Fandom is a community. We TALK to each other; at cons, by email, by blog. You don't need to organize a political party to know that SEVENEVES and AURORA and UPROOTED and THE FIFTH SEASON were important books last year. Love 'em or hate 'em, those were the books people were talking about, wherever people talked about books (those and two dozen others, to be sure). It does not require any organization. It's a natural process.

May. 17th, 2016 09:13 pm (UTC)
Two problems there; 1) When you say "X only needed 21 votes to make the ballot", it reads like "If only 21 other people had nominated". Alas, what it really means is "If only 21 * (1 / percentage of ballots to list it)" had nominated in accordance with the rest of the nominations. In other words, if X was only showing up on 10% of the ballots, assuming new nominations would also have only 10% putting it on the ballot, you'd need 210 more general novella nominations to get the 21 specific nominations to get it on the ballot.

2) People nominate works they feel were the best, and that they want to win. I'm not willing to do a nominating ballot for, say novel, of say 2 things I really liked and thought worthy of a Hugo and three things I didn't like/didn't read, but I've seen a lot of buzz about. File 770's had a fair amount of book discussion in its comments for the last year, and it's not infrequent to see something like "I tried to read , but I just bounced off it/had these problems with it". There really are different tastes with respect to specific authors and subgenres.
Gregory Hullender
May. 18th, 2016 01:51 pm (UTC)
Preventing Vandalism is Hard
Kevin is correct (and I suspect he's using my numbers), but the 40,000 number is to prevent any slate candidates from being on the ballot in ANY category. That's what you need if you want to prevent vandalism (e.g. Space Raptor). If all you want to do is prevent a slate sweep, then EPH is enough as it stands.

And the weakest category is Best Short Story.
May. 18th, 2016 04:20 pm (UTC)
Re: Preventing Vandalism is Hard
Weakest in what sense? You mean this year, or historically?
Gregory Hullender
May. 18th, 2016 05:12 pm (UTC)
Re: Preventing Vandalism is Hard
"Weakest" means it has the most "spread". Doubling the number of voters has the least impact on short story (of the literary categories) because the voters tend to get diluted across the enormous number of possible candidates. Best novel has much less spread, probably because even though there are a lot of novels, there are relatively few of them that get a lot of buzz.

So if you want to eliminate "slate vandalism" purely by increasing the number of voters, it's a lot easier to do that for Best Novel.
May. 18th, 2016 05:29 pm (UTC)
Re: Preventing Vandalism is Hard
Okay, I can see that.

This too is an artifact of the internet. Not so long ago, you could find almost all the significant short fiction (short stories, but novelettes and novellas as well) concentrated in the surviving print magazines: ASIMOV'S, ANALOG, and F&SF. Most nominees came from those sources, with an occasional outlier from an original anthology or e-magazine. And the magazines were still widely read by fandom, so that "buzz" you speak of was a real factor.

These days? Not so much. There's been a huge upswing in the number of e-magazines and online sources, and a corresponding decline in the print magazines. They are still there, but they seem to have less and less influence every year. Oldtimers like me still hate reading on screen, but younger fans are not troubled by that... and some only seem interested in reading the short fiction they can download for free from the internet. The very notion of going to a newsstand or bookstore and paying for the latest issue of F&SF or ASIMOV'S seems to be anathema to some of them.

All of which helps create the spread you speak of.

Edited at 2016-05-18 05:45 pm (UTC)
Gregory Hullender
May. 18th, 2016 05:51 pm (UTC)
Re: Preventing Vandalism is Hard
Definitely. Here's a graph showing the effect dramatically:


Scroll down to the next graph, and you'll see that the Nebulas have done the same thing.

By the way, you probably don't remember me, but when you spoke at Amazon a few years ago, I was the guy who asked you about your short story "A Peripheral Affair," which I read in F&SF back in 1973.
May. 21st, 2016 11:43 am (UTC)
Re: Preventing Vandalism is Hard
This is why sites like your Rocket Stack Rank, the Hugo Art Tumblr, the Hugo recommendation wiki and Google doc spreadsheet, and even Sad Puppies IV are important tools. They help people concentrate their search on things that others already recommend.


George R.R. Martin
George R. R. Martin

Latest Month

February 2018


Powered by LiveJournal.com
Designed by Lilia Ahner