A clean slate for commenting on today’s session.
Discover more from File 770
Subscribe to get the latest posts to your email.
File 770
Mike Glyer's news of science fiction fandom
A clean slate for commenting on today’s session.
Subscribe to get the latest posts to your email.
If 43 out of 85 of a voter’s nominations are from Beale’s slate, they aren’t voting their own tastes. They’re just here to burn us down.
Thus cutting their power in half, making it easier for EPH, EPH+, 3SV and whatever else we do to minimize their mayhem.
I’ve avoided sleep long enough out here on the East Coast. Pearson Specter Litt is in trouble on my DVR. The firm must endure. Thank you for helping me give this hypothetical a spin.
@rcade: I worry that the following will sound insulting. I’ll be darned if I can find phrasing less likely to convey that message. I’m sorry in advance for the degree to which it sounds that way anyway.
I think you’ve fallen prey to two of the weirdnesses in how human minds work. I don’t say this to sneer, because I struggle with them and others all the time; I’m more self-consciously aware of them because my auto-immune troubles include cognitive impairments and I have to monitor for them. Even so, I get led astray all the time. So the thought here isn’t “ha ha you suck, you fool” but “oh, I recognize you making a pair of mistakes I recognize from doing them all the time myself”.
#1. Memory consists of a bit of recall and a lot of invention. Our brains store conceptual triggers, around which we re-build, drawing on related triggers and a bunch of other sources. You can see the process at work in someone who didn’t just fail to try to correct for it but actively fed the process in which ever Puppy it was who reported a good Worldcon experience at the outset and then worse and worse as time went by and his buddies’ negativity got wrapped around true data. But absolutely everybody goes through it – all experiences become stories.
#2. Both in sleeping dreamtime and in related waking states of musing and such, the role of invention goes even deeper. We get a confidence that all the details are there even though when we stop to analyze, they just aren’t.
You strike me as a careful, honest person, and that’s not idle flattery. I’d trust you to look after my cats or my medical data in a crisis. I’d expect you to execute any responsibility you picked up well. This is not in any sense an attack on your competence, integrity, or honesty. I believe, fully, that you have in mind some clear examples of the kinds of rules that strike you as doing the job of removing actual slates with minimal collateral damage.
But I think that your conscious, judging mind has been misled by other parts, and that what feels clear in your mind isn’t actually that clear or complete.
What we need at this point are some written-out examples of such rules, put down for all to see and analyze. It doesn’t really matter what we imagine is possible for this stuff; we collectively need to see what we can or can’t document.
I’m unconvinced, given the scenario offered.
I’d rather add items to the ballot so if there’s one slate work, the next nominee gets added, two slate works, a second gets added and so on (I know, I know, people didn’t like that one, but I do) than throw ballots out. Throwing ballots out seems to me to have a high chance of false positives unless the correlation is set extremely high, at which point the work-arounds are obvious. Adding works has downsides (doesn’t actually remove terrible works from being able to mark themselves as Hugo nominees; could lead to a maximum 10-12 works on the ballot which is a legitimate concern for judging burden) but none that would be as painful as throwing out ballots from people nominating in good faith who just happen to have Puppyish (or unwilling shields) tastes.
I get the urge to just chuck out ballots perceived as slate ballots, but I don’t like the weight that would put on the administrators shoulders and I’m not keen on collateral damage of that kind, either.
I’m with @Kurt Busiek. I was surprised when VD posted his “final, final RP2 slate” the number of things in common with my nominations or those of others on File770 until I remembered he trolls us. He read the various recommendation list put up to help each other find works we might not come across on our own.
If I’m being insulted, you did it in such an amazing way I can only appreciate it. Thank you for saying I’d be trusted to care for your cats. That’s high praise.
As a programmer I’d love to be in a position where I had access to hard data on how our nominators behave so I could test some of the premises we toss around as we’re spitballing solutions. I’d like to see Schneier’s clustering analysis idea used to identify bloc voting on 2015 and 2016 voting data. Would it work, or would it have an unacceptable number of false positives?
Worldcon 75 is doing some ambitious open source coding and I’ve offered to help. Maybe in the future I’ll be in a position to conduct more concrete analysis (if that was permitted, of course).
While I’m so far outside the loop, I feel less compulsion to be rigorous. The hypothetical I posed was more of a “tell me why this is crazy” and less of a “let’s do this.” I’m just grasping for ideas. The business meeting mindset we seem to have, of chipping away at the problem while believing there’s sanctity in every nominating vote even if it follows an entire slate, worries me. Look at Philip Sandifer’s Twitter feed for how he feels about our pace of addressing the problem. I think we run the risk of losing a lot of Hugo fans to such sentiments.
There is one simple test for the effectiveness of 3SV. If the name Vox Day appears on the final ballot under Best Editor or anything else, I think we can conclude that 3SV has failed. Because if anything should motivate people to take the time to say “no!”, it would be the threat of seeing VD on the ballot again.
Of course, it’s possible that VD simply won’t try to get onto the ballot once 3SV goes into effect, simply so that people won’t be as motivated to participate in the new stage of voting. But hey, that’s a win too! 🙂
@rcade: Actually, I’d encourage constructing some deliberately irrelevant rules, to get a sense for what a possibly viable rule looks like entirely removed from the current mess. We may hope, at least, that someday the current mess will be irrelevant, after all.
@P J Evans & David Goldfarb: Thanks for the info!
@jonesnori: Thanks for the run-down on the suspension clauses! Much clearer now. 😀
@Kendall – you’re welcome! Writing up my notes helped me, too. A lot happened over the four day meeting.
Tasha:
If a semi-organized group runs around during a con and harasses people, certainly the whole group risks expulsion and not just whoever organized them? If we take it as a starting point that “pissing off the voters and bringing the awards into disrepute” is a CoC violation, I see no reason to limit that to the slate pusher and not include the slate followers too.
Given VD’s current streak, can the next WorldCon not give him a “Liftetime Achievement No Award” and be done with it?
@Johan P
I’m uncomfortable apply the CoC to ballots rather than public actions. How much would the ballots need to match the published slate? 60%? 70%? 80%? 90%? False positives here is a humongous deal where your talking vote disenfranchisement and violation of CoC.
rcade: If it doesn’t get fixed until 2018, that means we lost the nominating process to a malicious clown from 2014-2017. 2014 was trivially affected (one work).
Fraud does not mean “I didn’t do what you want”; it’s misrepresentation. Where on the nominating ballot is the nominator asked to sign a statement ~”These are my personal choices unaffected by any outside influences.”? You’re abusing the word the way many people abuse “censorship”
And Hypothetical: Would it be fair for a Hugo administrator to announce, before the nomination period has begun, that any ballot with more than 50% of its nominees from the Rabid Puppies slate will be thrown out? No, it would not be fair, as answered by many. Note also that with the current distribution it probably wouldn’t even work; the figures I saw for last year indicate that Puppy nominees in most categories got twice as many nominations as non-Puppy, so a Puppy ballot could be 49% Puppy and 51% random. (This came up a few weeks ago, in the discussion of whether 3SV is sufficient without EPH; the evidence was that the Puppies could make up 3 different nominating ballots, divide among them, and still swamp categories.)
And The business meeting mindset we seem to have, of chipping away at the problem while believing there’s sanctity in every nominating vote even if it follows an entire slate Were you at the meeting? I wasn’t, but from various conversations I doubt that was the sentiment; a lot of the people involved are old enough to remember “We had to burn down the village in order to save it!” and/or have other reasons to distrust sweeping “solutions”.
@snowcrash: I haven’t read “Penric’s Demon”, but I’m surprised you find it a disqualifier; did Bujold suddenly shift towards Puppy politics? AFAICT she’s still a popular author among fandom-at-large.
@Tasha: Find ways which keeps power in hands of membership (3,000-20,000+) I agree with your (omitted) conclusions, but the biggest Worldcon didn’t break 5 figures.
@Chip Hitchcock
I was adding 3 conventions together for nominating – I might be overestimating 2015 and/or 2016 possible eligible to nominate numbers – this would include supporting & attending membership for 3 years
ETA: Penric’s Demon – I nominated and voted for it. I’m not a puppy. LMB asked to be removed from the SP4 list which is when the RP2 added her.
I was counting by the years the awards covered, not years they were held. So 2014 was the year Beale first filled the entire ballot with junk, 2015 the second and so on.
I think of this in terms of the opportunity lost for the authors and other pros who could have been nominated and won, if we weren’t treating votes cast in bad faith like every other vote.
No, I followed it via liveblogs and YouTube. I had hotel reservations and plans to attend, but had to cancel for work reasons.
Lets remember one thing:
This is not about Beale.
This is about adjusting the voting process to survive into the internet age. To create a troll defense that is likely to work for 30-40 years, regardless of which bad actors may appear and how they will try to slate works onto the ballots.
Beale is no more important than being the current troll.
Oh, and I nominated and voted for Penric’s Demon.
@rcade
In that case, 2014 and 2015 are likely to be the only such years. Although EPH is imperfect, it should prevent any slate from sweeping any but the weakest categories, making it impossible for VD to “burn the Hugos down.” An indirect effect of that could be that his support drops from ~400 to ~200, as suggested by the WorldCon stats this year. To continue to support his crusade, his non-fan minions would have to shell out $50 apiece, even after it has been demonstrated that he cannot win. His core support within fandom seems to be 165, so it’s not unreasonable to assume that 200 is the max he’ll have for WorldCon75.
That would keep him off the ballot entirely in Best Novel, Novella, Dramatic Presentation (Long Form), and Editor (Long Form).
It would limit him to one in Best Novelette, Short Story, Related Work, Graphic Story, Dramatic Presentation (Short Form), Editor (Short Form), Professional Artist, Fan Writer, and the Campbell Award.
He’d get two in Best Semiprozine, Fanzine, and Fancast.
He’d get four of six in Best Fan Artist.
That assumes the fans manage to nominate as much as we did this year, so it’s the rosiest projection, but I don’t think it’s too far off. There are no guarantees, of course, but I’m cautiously optimistic that Vox Day will not control the Hugo Ballot again.
It has struck me that there is a, very rough and ready, test for the effectiveness of 3SV: since 3SV is meant to bring forward the operation of No Award, we can predict that at most those nominees which came below No Award would fall to 3SV. (‘At most’ because in some cases we were only in a position to No Award them after reading/watching and thinking about them.)
This confirms my view that this year’s Novella ballot would not be affected by 3SV.
Thanks to all for their reports from the Business Meeting, and subsequent disussion on what it all means.
Not that I actually understand the all of the latter… a lot of those disagreements are very confusing.
Anyway, the basic outcome sounds good!
Maybe it will all work better, or I will understand better, next year…
…perhaps see you in Helsinki?
Love and Peace/
rcade: Look at Philip Sandifer’s Twitter feed for how he feels about our pace of addressing the problem.
It is unclear to me why you think his opinion on this should have any more import than that of any other fan. Sandifer has done some great work for the genre, but I frequently disagree with his opinions — and after he advocated nominating and voting for Pratchett’s book without actually reading it, I lost a lot of respect for him.
That’s helpful, but having only 1 or 2 legitimate candidates in a category is a lot of fire damage. As an EPH sponsor, I thought it would do far more to reduce bloc voting than it has in actual testing.
We didn’t need troll defense for 60 years. No competing slates have emerged since last year because Hugo voters of good faith hate the tactic. Sad Puppies didn’t slate.
It’s possible the bloc voting problem begins and ends with Beale.
But I do applaud the notion we can anticipate the future well enough to protect the Hugos of 2056. How SF is that?
I didn’t say that it should. I used his reaction as an example of an opinion I am concerned will grow with each arson-damaged ballot.
Here’s how I see the overall situation.
a. The current rule changes are not, as everyone admits, a panacea. If they are applied and have the best possible effect, slaters still have an unacceptable degree of power (witness this year’s Novella ballot, which neither EPH – apparently – nor 3SV would have changed).
b. It is possible (likely, even) that the situation will solve itself through declining enthusiasm from slaters. But it is also possible that it will not.
c. If it doesn’t, it is a bad thing. We will be able to give Hugos to worthwhile works, and will probably be free of outright abuse, but that’s not good enough; slaters will still be able to skew the ballot in a particular direction, and prevent it being the actual choice of members of Worldcon. (‘But they are members of Worldcon!’ Yes, but their votes do not represent real preferences.)
d. Increased participation has only limited power, as this year’s results show. We can increase overall participation further, but it’s not clear we can increase participation in every category. The expectation that everyone will nominate in everything puts people under unreasonable strain, and limits what categories we can have. (I’m also afraid that the pressure to vote in everything, when added to 5/6 and the YA award and the series award, will weaken the expectation that we should read the stuff, and make the Hugos into more of a ‘vote your favourite’ thing like the Dragons.)
e. We can go on making changes to deal with specific problems. EPH deals with last year’s problem (attempted sweeps); 3SV deals with one of this year’s problems (abuse); we may be able to find a way of dealing with next year’s problem. But the system remains vulnerable. (What Hampus says is relevant here.)
f. So the possibility of a method which would cut slating off at source remains attractive. Obviously we don’t yet have an effective description of such a method, which could be applied mechanically enough to avoid giving administrators unacceptable power. But these things take time to work out. The first attempts to approach the matter as a problem in voting systems theory produced ‘surely you could not….’ objections; so did the first proposals to use a longlist. The fact that no one has yet described a workable method shouldn’t stop us investigating the field.
rcade:
Sad Puppies didn’t slate.
I have a lot of sympathy with your approach, but I should point out that this was an accident. They started their list with a recommendation to vote the top five, which made it at least possible to use it as a slate. However, the list ended up in such a form that it would not make sense to use it as a slate. I agree it’s unlikely they will slate again.
@ Hampus
This is not about Beale.
This is about adjusting the voting process to survive into the internet age. To create a troll defense that is likely to work for 30-40 years, regardless of which bad actors may appear and how they will try to slate works onto the ballots.
This sums up my opinion pretty well. Even if Beale finds some excuse to slink off the stage right now, someone, somewhere, somewhen will try to attack the weaknesses in the nomination process now that they’re known just because they exist and, you know, humans are human. We have to do the systematic fixes to protect the process as much as possible.
Tasha:
So how do you feel about applying the CoC to something that happened in a room with only two people in it? Or a quiet corner of a public room, or any other situation without witnesses? A ballot is arguably better evidence than what one will often have in harassment complaints – it’s basically a paper trail of the accused acting a specific way.
Looking at this year’s nomination stats, I am confident that it would be easy to identify at least 300 voters who based their nominations almost exclusively on the rabid slate. Depending on exactly how the ballots are spread, and how much we will allow people the benefit of doubt, it might be possible to raise that number to 400 who show clear influence. That would not eliminate rabid influence completely, but even the lower number would have given us much more real choice in many categories.
***
Re Penric’s Demon: Noone have suggested that only rabids nominated it. What Snowcrash implied was that nominating that book along with certain other works is a certain sign of rabidness. It’s the combination, not that book by itself.
It would be good to have a troll defence that would last 40 years, but given that it is apparently mathematically proved that no voting system is completely invulnerable to bad actors, it’s hard to see how this might be done. We may find ourselves restricted to coping with actual threats.
A general ban on slates would be more powerful than the various current plans, but can only be applied if a workable way of doing it is found, which hasn’t happened yet. (And even that would only apply to slates. Black Genesis is generally agreed to have been a case of bad action, but it wasn’t a slate in a voting systems sense – you obviously can’t have a bloc vote for one work.)
@Johan – yup, I was referring to any ballot that contained all of those items (Safe Space; If You were an Award; MGS V; Castalia House Blog; Penric’s Demon; Vox Day; Kukuruyo) was clearly being driven by a slate, not a ballot that contained any of them.
@Tasha: two points if you’re adding:
1. 3000 is way too low. (That’s why I thought you weren’t adding); it’s less than any single recent Worldcon, even the one in Australia.
2. 20,000 is still too high, as there’s significant overlap. One person’s UUSWAG was a core Worldcon attendance approaching 3000; I have my doubts, but there’s certainly \some/ overlap even when one of the 3 cons being merged is in a different hemisphere from the others.
If I remember correctly, the initial voting block we had for Sasquan (Loncon + Sasquan + MAC II) was roughly 13,000 individuals (non-duplicates) in the beginning of 2015. We added several hundred more by the time we reached the must-be-joined-by deadline of January 31st, 2016.
It took us about a full month of work (multiple people working on it) to winnow out the duplicates, because people showed a real tendency to either move or change their email address or change the form of their name that they used when they registered (or all three)–so it was by no means a straightforward task.
@Greg: I hope you’re correct, but it’s possible that the Puppies simply figured out that they can effectively troll the nominating stage, and have no impact on the actual voting. If over a hundred Rabids bought new memberships just to vote on the final ballot, then that might be an indication that a whole bunch more will buy memberships once nominating rights are also at stake. (Alternatively, it could be a sign that they’re dumber than a box of rocks. Hard to tell with this group.)
@rcade: Remember, we don’t just have EPH in effect next year, we’ve also added a sixth finalist in each category. In combination, used on this year’s data, these measures would have produced three categories (semiprozine, fanzine, and fancast) with one non-Puppy nominee, and two categories (novella, fancast) with two. (That doesn’t count withdrawals this year by Puppy nominees in fanzine and fan artist, which would bump those categories up two two non-canines.) Plus, a couple of those categories had Puppy “shield” nominees, including File 770, which would’ve made the ballot whether or not the Puppies nominated them. So a pretty substantial supermajority of categories would have three or more legitimate nominees.
I agree with Hampus and junego that we need to look at this as “long-term anti-trolling measures” rather than “short-term shut-Beale-up measures”. And whatever system we arrive at isn’t going to be perfect and invulnerable – I don’t see how it can be.
But, as they say, the perfect is the enemy of the good… and what we need to do, I think, is come up with something that is good enough to make trolls look for easier targets. EPH, EPH+, 5/6, 3SV – these are all steps in the right direction. It will take time and detail analysis to see how well they work out in practice. But they’re real solutions, whereas saying “oh, just ban slates” is more a pious wish than anything else.
I don’t see Beale giving up trolling any time soon – he’s been posturing on the fringes of SF for over a decade now, hasn’t he? If he was ever going to get a more productive pastime, he’d have done it by now. My own unscientific guess is that we will see a slow decline in Puppy numbers from now on, as even his hardcore supporters (by now, he must be pretty much down to the hard core) become disillusioned and drift away. “Peak Puppy” was last year, when they could still pretend they had some sort of legitimate case to make. I’m guessing “persistent but steadily lessening annoyance” for the next few years.
And for most of these 60 years, there was no internet from which trolls could coordinate campaigns.
Not SF at all. It is basic security, that when you find a vulnerability, you try to remove it.
Andrew M on August 23, 2016 at 7:11 am said:
I don’t agree we can figure this out based on the data we have now.
Three Stage Voting requires:
3,130 valid votes were cast in Best Novella. 60% of that is 1,878.
There were ~6715 preregistered members. I think that includes supporting members. 20% of that is 1,343.
I don’t see how we can know (even given 3,130 people bothering to vote in the second stage) that 1,878 of them would have voted to reject slate entries.
The biggest drawback of 3SV is getting the membership to vote in BOTH final stages. Lots of people will refuse to vote in the second stage on the basis of not having time to read *all 15* entries. Yes, this is meant to be a quick “throw out the bad actors” phase. But a lot of people feel they can’t vote for or against a work without reading it first. Can we be sure of getting 20%+ of ALL members involved in that second stage? I hope so. We will have to see how it pans out.
I also just realized that the 20% figure is a moving target. 20% of all members *on what date*? The date the second stage ends? We could have many join between the nomination cutoff date of 12/31 (assuming that date change is ratified next year) and the end of the second stage. And they may not all bother to participate. (Hmmm. I wonder if it would be a lesser change to add a cutoff date to that section? Or at least make explicit that the cut off date is the last day of being able to vote in the second stage.)
@Johan P So how do you feel about applying the CoC to something that happened in a room with only two people in it? Or a quiet corner of a public room, or any other situation without witnesses? A ballot is arguably better evidence than what one will often have in harassment complaints – it’s basically a paper trail of the accused acting a specific way.
I keep staring at these questions in shock. As a woman whose been raped and sexually abused seeing someone try to compare sexual and racial harassment to voting for an award is unbelievably offensive.
@Chip Hitchcock
I admit my estimate was bad. I apologize for not doing my research. It was bad because it derailed this conversation. I’ll try to do better in the future by using Google or appropriate websites in the future. Please forgive me?
@John Lorentz
13,000 – thank you. I was only off by 7,000 in the upper limit. 🙁
Sorry for being one of the difficult people with multiple name & email changes. Different people registered me in different years and my attempts for 2014, 2015, and 2016 to get my name and email fixed were unsuccessful until this year.
@Tasha – Johan P’s hypothetical bugged me as being disingenuous but I couldn’t figure out exactly why. And then you put your finger on it so devastatingly. I can only suppose that I missed naming what he was describing because I’ve been unusually fortunate in my life experience as a woman not to have had very much in the way of “something that happened in a room with only two people in it, or a quiet corner of a public room, or any other situation without witnesses” happen to me thus far.
Hugs if you want them.
@Stephen desJardins:
I’d say not just possible, but clearly so. Nominations got targeted because of low participation, less vote-choice convergence among organic voters, low cost of participation, and the system’s reliance on a vulnerable algorithm.
@Steve Wright:
It was also the last occasion Mr. Beale could claim the RPs were pursuing a righteous cause to merely redress ideological bias, as their 2016 nominations were just a combination of Castalia self-promotion and attempted vandalism.
During the 2015 voting cycle, my wife Deirdre and I spent some time talking with Theo Beale in e-mail. (We’ll basically talk to just about anyone who’s willing to seek common ground.) One of my points to Theo is that WSFS are just not his enemy: We’re a profoundly democratic big-tent assembly of fandom with no particular ideological tilt, and collectively have no particular concern with his lot in the first place, other than to curb screwing their around in our public functions. I think it’s clear that he knows this already, but seeks the cheap publicity, i.e., WSFS are just a drive-by target in his larger culture war and promotion of his micropress.
Can I ask, please, for more consistent clarity about the Sads no longer being a problem, and no longer blame them by implication? As I pointed out to Xopher at the Business Meeting, I participate annually in the BASFA Hugo Recommends list process (here in Silicon Valley), which has its opposite number on the opposite coast of the USA in the NESFA recommendations list (in New England). The point being: SP4 was essentially indistinguishable from those two well-accepted efforts, except (IMO) for a severe case of bad attitude and a persecution complex. I am thus obliged, on pain of being a hypocrite otherwise, to plead their case: They’re a totally legitimate part of the WSFS community, in my view, and I’d appreciate more-consistent recognition of that fact. And their taste in SFF doesn’t even suck. ;->
(We saw that, IMO, with Ms. Paulk at the Business Meeting. While parts of what she said to the assembly about voting left me scratching my head in bewilderment, she did participate and respect the process, which is more than my excitable friend Jo Rhett did.)
rcade – We didn’t need troll defense for 60 years. No competing slates have emerged since last year because Hugo voters of good faith hate the tactic. Sad Puppies didn’t slate.
It’s possible the bloc voting problem begins and ends with Beale.
I think you’re being needlessly optimistic. Once a weakness in a system has been exploited (as Correia’s Get Larry a Hugo did), the chances of it happening again go up. In this case VD, who is not an original thinker but is a fairly effective adapter of the ideas of others, combined a now known weakness with his ongoing animus, and presto! another opportunistic infection was born.
As Hampus points out, VD is just a flavor of the year internet troll and while it feels important to stop his tactics, that’s actually not the real point of the exercise. Instead, if everyone looks systematically at the nominating and voting process, subjects it to every sort of imagined griefing strategy and then comes to some sort of consensus about best practices, that makes for a robust response.
It’s not actually a good idea to publicly discuss ways to break the Hugos (it would be like leaving crumbs for cockroaches and then complaining the cockroaches ate the crumbs, plus I hate the idea of the smart kids doing VD’s homework for him), but the slow process of adding amendments at the various Business Meetings, seeing how they work and where their weaknesses are and then adjusting strategies really will allow Worldcon to develop that robust process (eta – without completely changing the nature of Worldcon and the Hugos).
@ULTRAGOTHA:
I’m more optimistic, especially given the Internet jungle telegraph, that many, many hundreds or (I hope) thousands of current-year Worldcon members will be willing to apply the intended ‘Are you serious?’ quick test in the semi-final round. Reminder: The WSFS community has until April 2018 to get used to the idea (assuming the Helsinki Business Meeting grants 3SV ratification).
Unlike EPH/EPH+, 3SV actually helps Hugo Administrators, by crowd-sourcing the contacting of prospective Long List additions during the semi-final round, vetting their eligibility and their willingness to accept nomination. This is an important serendipitous advantage to 3SV, and it’s also quite transparent and obviously democratic (no matter what Ms. Paulk thinks).
But they’re real solutions, whereas saying “oh, just ban slates” is more a pious wish than anything else.
Saying ‘Oh, just ban slates’ is equivalent to saying ‘Oh, just change the voting system so that everyone is represented proportionately’. The pious wish comes first, and then we can put our minds to how it would work. Just because we don’t yet have a description of how it would work, that’s not a reason for abandoning the search (or else EPH would never have been developed).
But, as they say, the perfect is the enemy of the good… and what we need to do, I think, is come up with something that is good enough to make trolls look for easier targets.
Well, one problem here is that this is VD. Actually, come to think of it, defeating an abstract hypothetical troll may be easier than defeating VD, because an abstract hypothetical troll is probably trying to achieve something by their trolling, so if we fixed it so that they couldn’t achieve anything, they would stop (as the SP group seem, more or less, to have stopped). But VD is just being disruptive, so he can say ‘Look, I’m disruptive! See how much I can disrupt!’. So we may be able to come up with a system which is proof against almost any imaginable threat – except the actual one.
But in any case, ‘the perfect is the enemy of the good’ is a fair response to those who say we shouldn’t adopt EPH or 3SV because they aren’t perfect; but I don’t think anyone here is saying that. What people are saying is that as they leave a problem in place, we must go on looking for ways to solve that problem. ‘Ban slates’ is at least a proposal for a direction we could look in. There may be others, but I haven’t heard any proposed yet.
@Nichole J. LeBoeuf-Little
I’m always happy to meet someone who hasn’t had to experience the darker side of predators and those who help excuse them. Reminds me the world isn’t all bad. 😀
Hugs gratefully accepted. Thanks.
ULTRAGOTHA: My estimate was meant to give an upper bound; people won’t vote down anything they didn’t No-Award. That’s enough to show 3SV is not a comprehensive anti-slate measure. My actual guess is that they would have voted down ‘If You Were an Award’, ‘Safe Space as Rape Room’, ‘SJW’s Always Lie’, possibly ‘The Story of Moira Greyland’ (but I’m less certain about that), and VD himself as editor. But you may be right that they won’t vote down anything. (It’s quite possible that nothing which is a suitable candidate for voting down will appear after this year anyway, in which case this stage will just be a puzzling ritual.)
I’m still puzzling over the ‘two months’ thing. Someone said in the meeting that we would have two months to do it; was this known in advance? I certainly hadn’t heard of it; I was thinking in terms of two weeks (which would make it a lot clearer that it wasn’t meant for general assessments of quality, only for dealing with abuse). I think it would be immensely disruptive to people’s schedules.
@Andrew M, I heard one month elsewhere for 3SV, which seems more plausible. I think the provision leaves it up to the adminstrators.
@Rick Moen
Can I ask, please, for more consistent clarity about the Sads no longer being a problem, and no longer blame them by implication?
To the extent that the SP are no a longer problem, this is a fair request. I’d agree that the final analysis seems to show that there weren’t many nominating – it looks like the identifiable SP-but-not-RP markers are “Sad Puppies Bite Back” in BRW with 95 noms and Jason Rennie in BESF with 85 noms, so perhaps 90ish SP paying attention to “The List” and nominating at least one rather unpleasant griefing work in “Sad Puppies Bite Back”, but not voting anything resembling a full slate (otherwise we’d be seeing other SP works at the end of some longlists). They may well be partially supporting some of VD’s efforts as well, but that is harder to discern. Is that no longer a problem? Compared to VD, not really.
I think it’s a good idea now to try to discern between SP and RP when talking about them, and credit each accordingly.
@Andrew M
I think 3SV may turn out to be interesting as people who don’t nominate because it’s too overwhelming may get involved at this stage. I think your underestimating what might have been downvoted when people were looking at the longlist and comparing it to RP2.
I remember when 3SV was modified to be 2 months instead of 2 weeks. If you do a search on Google for “File 770 3SV” I believe you’ll find discussions here.
THIS. A thousand times, this. Thanks, @Hampus. 🙂
It’s crucial to understand how tenuous our position here is. Categories like “Best Related Work” are low-hanging fruit for abuse, with a very low portion of fans investing enough time to vote meaningfully. Categories like “Best Short Story” are so fragmented we just went and canceled the 5% rule, because with (e.g.) 2,000 nominating ballots, we just can’t count on getting enough stories that a “mere” 100 people agree on.
With a small nominating base and a LARGE pool of eligible works, any coordinated group beats the uncoordinated natural results BADLY.
This isn’t bad on the level of “OK, we need more people nominating”; for every 100 “good” nominators, they need maybe another 5-10.
This isn’t bad on the level of “OK, we need a nomination-tallying scheme that’s more representational”; a “winner” with 5% or less does not have a very strong claim on the basis of proportional representation.
And, no, this isn’t really bad on the level of “Well, with the right mechanism the nominations will work out fine”; trolls adapt, and we are very, very vulnerable.
Ultimately, the Hugo’s answer is going to have to be what the Internet at large has already figured out as the only real solution to trolls and other disruptions: moderation. You need somebody with the ability to look at whatever the current situation is, make a judgement call, and be able to say, “nope. This doesn’t pass.”
3SV is a solid, noble and well-considered attempt at crowdsourcing that moderation. It’s not perfect, but it’s as close as we’re likely to get, plus we can continue to tweak it if need be. Giving the moderation responsibility to the WorldCon crowd is excellent, and I really hope that it works well.
Now, it might not work well. We wouldn’t be the first to be attempting community moderation, and we wouldn’t be the first to implement it poorly. But if it doesn’t work well, we will still need moderation. Not because of Beale specifically. Because our system can’t withstand the unfiltered power of internet collusion – no more than newspaper comments or Twitter timelines can.
I hope this particular iteration calms down quickly. But these aren’t temporary measures; they’re long-term improvements that adapt the Hugos to changing times.
“Ban slates” is easy to say, but it’s hard to do in practice. A lot of people have similar nominating ballots, but they aren’t following slates.
You’d also have to change the rules to define slates and to allow admins to toss ballots that appear to be voting on a slate.