To Say Nothing of the Dogs; or, How We Confound the Hugos’ Third Slump (Hugo voting proposal discussion 5)

By Jameson Quinn: After 1, 2, 3, 4 threads here, and countless more that are peripherally related and/or on other blogs, discussing how to make the Hugo nomination process more resistant to slate voting and other “griefer” attacks (in other words, protect them from those intent on using the rules to disrupt and provoke), I believe a logical way forward is becoming evident. Below, I outline a set of proposals that could be made and passed at the Business Meeting.

If you count every separate idea below, there are 5 new proposals. That’s a lot. Personally, I support all of them. But I realize that it’s unlikely that the business meeting will pass all 5. That’s why these proposals are designed to be as modular as possible. Sure, they could all be passed together and would work to reinforce each other; but in most cases each one would also be a sensible step forward if passed alone.

Following that, I will answer some obvious questions that these proposals bring up, in FAQ format.

At the end of this post, I’ll present a rough draft of how these proposals might be formalized. This is for illustration purposes only. In the case of 3SV, there is a group of current and future Worldcon Chairs and Hugo Administrators working on formal language, and I expect they’ll do a much better job than I have. But I needed to present an illustrative version of 3SV to make sense of the extra proposals that would modify it.

I will also have some things to say in the comment thread that shouldn’t be here in the main post.

Overall plan outline:

  1. Pass EPH this year. This should at least keep a slate from sweeping any category. (As for 4/6, it is not as important. I personally would support amending it to 5/6 passing it; but the plan does not hinge on this.) Also, remove the 5% threshold this year.
    1. Present a proposal for EPH+, a technical enhancement to the EPH vote-counting system which is projected to give slates about one fewer finalist per category in circumstances like the current ones.
  2. Create a “base 3SV” proposal, which inserts a new qualification ballot between the current nominating and final ballots.
    1. The Hugo Committee would publish a “longlist” of the top 12 or 15 works in each category from the nominating ballot without specifying their voting totals or the order in which they finished.
    2. A second round of voting would then be conducted in which members of the current Worldcon can “preemptively no-award” any of the works on the longlist.
    3. Longlisted works which are rejected by a majority of the voters AND by an absolute minimum number of voters would be excluded from consideration for the final ballot. (The absolute minimum would be a substantial number – potentially 20% of all eligible members – to ensure that a slate-voting minority cannot hijack this round of voting).
    4. The final ballot would then proceed as at present, with finalists being the highest ranked candidates from the nominating (first round) ballot who were not excluded by the subsequent qualification vote.
  3. Create several additional proposals, intended primarily to modify or go alongside the base 3SV proposal above:
    1. First, a “mercy panel”, established by the prior year’s Worldcon, with the power, by unanimous vote, to “spare” any longlist work(s) from the 3SV voting round. Works so “spared” would be able to become finalists without possibility of second-round rejection, but would of course still be subject to being voted below “no award” in the final round.
    2. Second, a “+2 against trolls”  (Hat tip) round of voting, simultaneous with the 3SV round. This would allow any eligible first-round voter to add nominations for 2 works per category, that they had not already nominated, from the longlist. These nominations would be counted in determining the finalists, just as if they had been made in round 1. (Note: most of the discussion on this point has been made on the “+1” version of it; it was first proposed in combination with 3SV as “3SV+1”. But passing +2 this year would allow us to amend that to +1 next year, while the reverse is not true; so +2 is the best way to keep our options open.)
    3. (Note: this is a new proposal which has not been discussed in the prior threads, and as such, it is far more tentative than the rest) Third, a “extend finalists” option in the 3SV voting, whereby voters may choose for each work between the options “Reject”, “Accept with extension”, “Accept”, and “Abstain”. If one or more works that had gotten more than a certain threshold of combined “Reject” and “Accept with extension” votes became a finalist, the target number of finalists in that category would be increased by 1 for each such work, up to a maximum target of 7 finalists. As with rejection votes, the totals for “Accept with extension” votes would be published after the Hugos were awarded, but the names of specific works would be anonymized, to make it clear that the judgment of “Accept with extension” has to do with voting process and not with the quality of a work itself
  4. (optional) Discuss the possibility of a proposal to go back and fix the no-awarded categories from this year and last year using retro Hugos. I do not believe that a proposal to do this should be made this year; the Business Meeting will be plenty busy enough without it. But I think it’s a good idea to begin this discussion and have a proposal like this ready for 2017 or 2018.

FAQ

Overall structure

Q. You say that a way forward is becoming evident. In whose view?

A. Many of the aspects of the plan I’ve outlined have been arrived at collaboratively and have drawn more-or-less broad consensus in the relevant discussion threads (linked at the top of this post). It’s true that other aspects more clearly come from one person: me, Jameson Quinn. But the proposal drafts above are not the end of the road; please comment below, and we will continue to work on them (including discarding ideas, if appropriate).In this thread, even if you have no new arguments to add, it is useful to just say just “I like this part” and/or “I don’t like that part.” Of course, if you’re going to say one of those, please do read through this FAQ, as your concerns may be addressed already.

Q. We can’t make any actual decisions until the Business Meeting, and many people who will be there aren’t even paying attention yet. So why have this discussion now?

A. There is no question that the Business Meeting has the final word. We’re not trying to take over the decision process here, just to smooth the way. That is: the point of the discussion now is to refine the proposals, clarify their advantages and disadvantages, and get consensus on the points that become obvious through that process.

Q. Is this set of proposals too complicated to even work as designed? (joke youtube link)

A. I have several things to say to that. First off, you’re right, they are somewhat complicated; but I’d argue that well-designed voting systems, while they should avoid senseless intricacy, can reasonably be a bit involved. Secondly, each of the the three main parts of this — EPH/EPH+, 3SV, and +2 against trolls, — could stand on its own, helping solve some aspects of the current situation even if the other parts were not present. This partial redundancy means the whole is more robust, not more fragile, than any of the parts.

Q. Are these proposals worth the extra complexity they add?

A. I’ll address this question for each proposal separately. But more generally, it depends on what problem(s) you’re trying to solve. See the next question for more on that.

Q. Out of the 5 new proposals here (plus EPH), what is the least we could do that would address the problem?

A. That depends what you mean by “the problem”. You might think it’s:

  1. Slate sweeps (that is, the possibility that all the Hugo finalists in one or more categories come from a slate) In this case the minimal solution is to Pass EPH. See my paper with Bruce for more details.
  2. Slates nominating works that are intended to defame, harass, offend, or otherwise discredit the Hugos? In this case the minimal solution is to Pass 3SV.
    1. How could we solve the above without causing undue negativity? In this case the minimal solution is to may be 3SV, if social norms are enough to prevent campaigns to reject good-faith works. If not, adding on the Mercy Panel would help.
  3. Multiple categories without a choice between at least 2 organic (non-slate) works: In this case the minimal solution is to Pass EPH+. If you do not like EPH+, then you could get a similar result with a combination of EPH, 3SV and “+2 (or +1) against trolls”.
    1. If you think the problem is even 1 category without such a choice, then the minimal solution with enough of a safety margin is probably combining EPH+ and 3SV+2 (that is, both of the possible solutions just above).
  4. Slates acting as “kingmakers” (that is, supporting some works which already have organic support, merely in order to deny others the chance of becoming finalists)? In this case the minimal solution is either EPH+, 3SV, and +2 (or +1) against trolls; or, alternatively, EPH, 3SV, and extend finalists. For the greatest possible strength against kingmakers, therefore, you’d combine EPH+, 3SV, +2 (or +1) aganst trolls, and extend finalists.
  5. All of the above: to solve all the problems above, you’d probably need all the proposals.

Q. Even if you consolidated all of this, it would still be at least 2 and probably 3 different proposals. Do you really expect all of that to pass through the BM?

A. Well, it’s not impossible. But frankly, no, I wouldn’t expect all of this to pass. But I still think it’s worth comprehensively laying out all the different problems and what it would take to fix them all, so that we can make this decision with our eyes open.

Q. Isn’t there a simpler way? Have you looked for one?

A. We have looked for one, and I believe the proposals above are the best of what we’ve considered. Others may disagree. For now, this FAQ is just a starting point for more discussion.

Q. OK, enough beating around the bush. What do you actually hope passes?

A. I’d like all of it to pass. But I’d be pretty satisfied with EPH+, 3SV, and +1 against trolls. I’m not attached to the mercy panel or extended finalists.

EPH and EPH+

Q. Is EPH+ worth the extra complexity it adds? Why?

A. I think it clearly is; but then again, I (Jameson) may not be the best person to answer this question. The advantages of EPH+ are clear; it helps ensure more of the finalists are determined by organic voters. As for the disadvantages and/or complexity: since I live and breathe voting theory, it actually doesn’t seem any more complex than EPH to me. Though I realize that is probably not true for other people, it’s hard for me to judge how much of a problem that is.

Q. This is complicated. Where can I read more about why it was proposed?

A. Start with Bruce Schneier’s post. If you’re up for some academic jargon, you can supplement that with our paper. Finally, read this post from the previous thread in which I try to explain the reasons for EPH+ in plain language.

 Q. Is EPH or EPH+ still needed if we have 3SV, or if we have 3SV along with some of the related options?

A. I believe that at least EPH is still necessary to prevent a slate from taking over the longlist. Beyond that, see my answer to the “what is the least we can do” question.

Q. I want to look at lots of graphs!

A. I have a bunch. (more to follow on this, in the discussion.)

3SV base

Q. Is this proposal worth the extra complexity it adds? Why?

A. I believe it definitely is. If we want keep offensive and/or harassing works from becoming finalists, we need to empower somebody to do that. And fandom is too broad to entrust that power to a small group, so we need some proposal similar to 3SV. Obviously, we could write the proposal differently, but I believe the version above is a reasonable starting point and sticks to the essentials.

Q. Why 3SV (voting to eliminate finalists) and not just a second round of positive “semi-final” voting to decide finalists?

A. Several reasons:

  1. A semi-final ballot would give voters just a few weeks to assimilate and vote on an extended list of longlisted items. Many of the most careful nominators would barely vote for any; while the most prolific voters would probably be going mostly by kneejerk reactions. This is somewhat true for 3SV, too, but it is less of an issue as explained below.
  2. A semi-final ballot conflates two questions: “Do I like this work and feel it may deserve a Hugo?” with “Do I feel that this work’s presence on the longlist or in the list of finalists would be a legitimate result of honest fan preference?”. In 3SV, those questions are separate, and votes to disqualify a work are based on the second question alone — one which does not require fully reading/reviewing every longlist work.
  3. Unlike a semi-final ballot, 3SV deals decisively with the issue of “troll finalists”: that is, works promoted by slates explicitly in order that their shocking and/or offensive nature might cast discredit on the awards.
  4. 3SV would be similar in spirit to the “no award” option which is already enshrined in the constitution, except that works thus eliminated would not take up space on the list of finalists, and awkward moments at the awards themselves would be minimized.
  5. A semi-final ballot opens up new kinds of attacks on the list of finalists, such as actually increasing slate voter’s capacity to act as “kingmakers” and/or perform “area defense” against certain kinds of works. All they’d have to do was to have enough voting power to reverse the gap between two works which both have significant organic (non-slate) support. But under 3SV, actually eliminating a work would not be possible without a relatively high “quorum”* of voters, and we hope that community pressure would lead to a low background level of organic rejection votes, so a minority of slate voters would be unable to use rejection as a weapon.

Q. Wouldn’t this lead to constant campaigns to eliminate works people just happen not to like, and thereby to hurt feelings?

A. There are several safeguards against this. First and most important is social pressure; it should be clear from the outset that this is a just safeguard against outright bad faith, not a chance to express differences in taste, and I believe that any Worldcon members who promote disqualifying a work just because they don’t like it will not get much support. Second, there’s eligibility. Various rules, discussed below in “open issues”, have been proposed to prevent a campaign to bring in Worldcon outsiders after the longlist is public. Third, there’s the quorum; if participation in the second-round voting is low, it will not be enough to pass the threshold to eliminate any work. Fourth, there’s the relatively short period of the semifinals, also discussed below. And fifth, there’s the possibility that (under some versions of this proposal) elimination votes for a specific work would never be publicized; only anonymized distributions of votes for each category. (In some cases, of course, the identity of which work got a certain vote total would be easy to guess, but that would still be just a guess.)
See also the Mercy Panel FAQs below.

Mercy panel

Q. Is this proposal worth the extra complexity it adds? Why?

A. That’s up to the Business Meeting to decide.Multiple people have expressed concern with 3SV, calling it too negative, and worrying how possible “rejection campaigns” would feel to the authors in question. This proposal would remove the possibility of such campaigns in clear-cut cases, reserving the power of 3SV for the edge cases which need it.It would also simplify the task of voting in round 2. This is a case where a little extra complexity in the rules actually simplifies life for the majority.

Q. Would this proposal let the panel keep something from becoming a finalist? That sounds like a bad idea!

A. No! The only power this panel would have would be to allow the nominations (from round 1 and possibly the “+2 against trolls” round) to stand. In other words, the panel can be a “good cop” by letting a work with strong support become a finalist without passing through a rejection vote, but could not be a “bad cop” by preventing it from becoming a finalist. The only group who could keep something from being a finalist would be the voters at large; exactly the same group that currently has the power to rank something below “no award”.

Q. Would this be too much power to give to a small group?

A. I think not. The only power this panel would have would be to exempt a work from the 3SV process. Even if they overstepped that power, we’d be no worse off than we are currently. In particular, the voters as a whole would still be able to put an offending work below “no award”.

Q. Wouldn’t this just be painting a target on the individuals on the panel, setting them up for harassment?

A. To a certain degree, the answer is yes; if you can’t handle people on the internet saying mean things about you, you probably shouldn’t be on this panel. But I don’t think this is a good reason not to have such a panel. Here’s my reasoning:

First off, there are manifestly people in fandom who can handle being a target, whether it’s because they bear it as a negative, because they actually enjoy the battle, or because they are such towering figures as to be beyond good and evil. I believe that there is a subset of these people who have earned fandom’s trust and could well serve as “your mercy”.

Secondly, I think that a “mercy panel” is designed to minimize such targeting. The question they’re answering is not the fundamentally controversial one, “does this work deserve to be kept from being a finalist?”, but rather, “are there reasonable questions about whether this work should be kept from being a finalist?”. It is at least possible to believe the answer to the first question is an emphatic “no”, and still accept that the answer to the second may be “yes”.

Finally, the fact is, haters gonna hate. There is really nothing we can do to prevent some people from becoming targets in this mess. With a panel, at least the targets will be people who’ve signed up for the job.

+2 against trolls

Q. Is this proposal worth the extra complexity it adds? Why?

A. I feel that it is. Because there are so many eligible works each year, many people nominate without naming even one thing that has any chance of becoming a finalist. This would focus attention on a list of 15 strong works, and give nominators some extra time to review these and add one or two that deserve it. The resulting finalists will have been reviewed by more fans, will have won more support, and thus will almost certainly be higher-quality overall. The fact that this proposal helps resist slate voting is almost just a side-effect.

Q. Why not let people add as many nominations as they want?

A. The 3SV voting round, when +2 is happening, should go by relatively quickly. Voters have limited attention to devote to this issue. Asking them to look at the longlist, do whatever extra reading/watching is necessary, and pick one or two extra nominations per category is probably the limit of what they can do without cutting corners.
Slate voters, on the other hand, are all about cutting corners. They can easily decide to add as many extra nominations as they’re allowed to.

EPH or EPH+ both make the story a little bit more complex, because it’s no longer optimal to add too many extra nominations. Still, there are some cases where adding more than 2 nominations could be a smart move for a slate. Why give slate voters that freedom, even if the cases where it helps them are rare?

Q. OK, so why not allow just +1 instead of +2?

A. That may well be the best course. But if we pass a +2 proposal this year, then changing it to +1 next year during ratification is a valid “lesser change”. So +2 keeps our reasonable options open. (For instance: what if we decide it should be +2 for Best Short Story, but +1 for all other categories? With all the proposals to look at this year, it’s probably not worth getting into details like that now; but next year, things may be clearer.)

Extend finalists

Q. What’s the idea here?

A. This proposal helps 3SV deal with the issue of “hostage” or “kingmaker” works, which have clear merit of their own but which have evidence of mindless support from slate voters. For instance, this year (2016), Neal Stephenson’s Seveneves was supported by a slate and became a finalist. He’s had several books nominated for Hugos, and several that weren’t; without the slate support, it’s impossible to be perfectly sure whether Seveneves would have made it. This option would allow works in this situation to become finalists, as the slate support was not the author’s fault; but to do so without pushing any other work off the list of finalists, because the list would just expand.

Q. Is this proposal worth the extra complexity it adds? Why?

A. Perhaps; it’s worth at least considering, though for simplicity in the business meeting it may be better to fold it into 3SV rather than taking it up separately. This proposal helps 3SV deal with the issue of “hostage” or “kingmaker” works, where a slate throws support behind things that have clear merit of their own, so that those things gets some votes from people who read them and some from people who didn’t.

Q. Would this make authors feel bad?

A. It shouldn’t. A vote of “accept with extension” is in no way a judgement on the work itself, just a judgement that there is reason to doubt the motive of some significant fracion of the work’s supporters.

Proposal Texts (rough drafts, for discussion purposes only):

A proposal for 3SV is currently being prepared by a group of highly experienced Worldcon runners led by Colin Harris (2005 Worldcon Chair) and also including former Worldcon Chairs Kevin Standlee and Vince Docherty and former NASFiC Chair Warren Buff. They expect to have a draft proposal published within two weeks (by 12 June), and if this were just about 3SV, I would rather have waited for them and not written something myself. But the other proposals following 3SV relate to it closely. So in order to write those other proposals in a form clear enough for further discussion, I needed to first write an illustrative text for a 3SV proposal.

EPH+

Moved, to amend Section 3.A.1 (1) of the E Pluribus Hugo proposal, as follows:

(1) Calculation Phase: First, the total number of nominations (the number of ballots on which each nominee appears) from all eligible ballots shall be tallied for each remaining nominee. Next, a single “point” shall be assigned to each nomination ballot. That point shall be divided equally among all remaining nominees on that ballot. each nomination ballot shall give a point or fraction thereof to each remaining nominee on that ballot, according to the number of such remaining nominees, using the following pattern: 1 point for 1 remaining nominee, 1/3 of a point each for 2 remaining nominees, 1/5 of a point each for 3 remaining nominees, 1/7 of a point each for 4 remaining nominees, and 1/9 of a point each for 5 remaining nominees (extending this pattern as needed if a ballot legally has more remaining nominees).  Finally, all points from all nomination ballots shall be totaled for each nominee in that category. These two numbers, point total and number of nominations, shall be used in the Selection and Elimination Phases.

3SV base (for reference only; to be superseded by the text from Colin Harris’s group)

Moved, to amend Section 3.7.1 (Tallying of Nominations), Section 3.9 (Notification and Acceptance), and Section 3.11.4 (Tallying of Votes) as follows:

Section 3.7: Nominations.

3.7.1: The Worldcon Committee shall conduct a poll to select begin the process of selecting the nominees finalists for the final Award voting. Each member of the administering Worldcon, the immediately preceding Worldcon, or the immediately following Worldcon as of January 31 of the current calendar year shall be allowed to make up to five (5) equally weighted nominations in every category.

Insert new sections 3.B and 3.C after section 3.8 and, if appropriate, 3.A from the E Pluribus Hugo proposal.

Section 3.B: Longlist publication.

3.B.1: In each category, the “longlist” shall consist of the top 15 nominees, as selected by the process detailed in section 3.8 [if EPH has passed] and section 3.A[end conditional]), but changing the number of desired nominees to 15 as appropriate. Any numbers involving limits on individual ballots shall not be changed.

3.B.2: In order to foment quick and accurate publication of each category’s longlist, the Worldcon Committee may exercise reasonable discretion in increasing the number of nominees on the longlist up to a maximum size of 18 nominees. Possible examples of situations that would call for such discretion are given in sections 3.B.3.1 and 3.B.3.2 below:

3.B.2.1: If eligibility cannot be quickly determined for a nominees, but it is thought to be ineligible, both it and an extra nominee may be included in case it is not eligible.

3.B.2.2: If the 15th-place nominee is nearly tied with the 16th-place one, and it is thought that a recount might show that their proper positions had been reversed, both may be included.

3.B.3: The names of the nominees on each category’s longlist, but not their order or vote totals, shall be made public with all due haste after the nominations poll is closed.

3.B.3.1 “Made public” means that the information should be conveyed to all eligible voters through some direct, personal means such as email and/asor postal mail, and also made generally accessible in some medium or media such as a web page.

3.B.4: Nominees on the longlist shall not be referred to as “semifinalists” or otherwise given any honorary status unless and until they have passed the eligibility voting described in section 3.C without elimination.

Section 3.C: Nominee eligibility voting.

3.C.1: After the longlist is published, each member of the administering Worldcon who had been eligible to vote in the nominations poll as described in 3.7.1 may vote on the eligibility of the longlist members.

3.C.2: At the discretion of the Worldcon committee, and with reasonable prior notice, the set of persons eligible to vote on nominee eligibility may be frozen before the close of voting, in order to ease calculation of the vote thresholds in 3.C.7 below.

3.C.3: The ballot for this round of voting shall present the voter with the following options for each longlist nominee: “Accept”, “Reject”, and “Abstain”.

3.C.4: Any voter who does not submit a ballot will be considered to have voted “Abstain” on all nominees.

3.C.5: Any voter who does submit a ballot but does not explicitly choose one of the three options for a given nominee will be considered to have voted “Accept” on that nominee. This default choice should be made clear on the ballot insofar as practical.

3.C.6: Postal mail ballots should be accepted insofar as it is practical given the schedule, but the Worldcon committee may if it chooses schedule this voting round in such a way that some members may not have ample time to submit such physical ballots.

3.C.7: A nominee shall be eliminated and considered ineligible if it meets the following two criteria: the number of “Reject” votes it receives is greater than 20% of the pool of voters eligible under 3.C.1 and 3.C.2; and the number of “Reject” votes it receives is also greater than the number of “Accept” votes it receives.

3.C.6: At the end of this voting round, the 5 finalists shall be chosen as follows:

3.C.6.1 The process described in section 3.8 [if EPH has passed]and section 3.A[end conditional] shall be used to put the entire set of nominees (eligible or not) in as strict an order as possible (that is, if the process is one of elimination, it shall continue until all nominees have been eliminated).

3.C.6.2 The finalists shall be the top 5 eligible nominees on this list.

Section 3.9: Notification and Acceptance.

3.9.1: Worldcon Committees shall use reasonable efforts to attempt to notify all of the nominees on the longlist, or in the case of deceased or incapacitated persons, their heirs, assigns, or legal guardians, in each category prior to simultaneous with the release of such information. Each nominee shall be asked reminded at that time to either accept or decline the nomination. If the nominee declines nomination, that nominee shall not appear on the final ballot.

3.9.A: [after 3.9.1]Since the longlist will be public information, members of the public may also help contact the responsible persons. Any nominee on the longlist which has not explicitly withdrawn by the end of the second round of voting shall be assumed to accept becoming a finalist.

3.9.2: In the Best Professional Artist category, the acceptance should include citations of at least three (3) works first published in the eligible year.

3.9.3: Each nominee in the categories of Best Fanzine and Best Semiprozine shall be required to provide information confirming that they meet the qualifications of their category.

3.11.4: The complete numerical vote totals, including all preliminary tallies for first, second, … places, shall be made public by the Worldcon Committee within ninety (90) days after the Worldcon. During the same period the nomination voting totals shall also be published, including in each category the vote counts [if EPH has passed] and points tallies as of elimination [end conditional] for at least the fifteen highest vote-getters and any other candidate receiving a number of votes equal to at least five percent (5%) of the nomination ballots cast in that category, but not including any candidate receiving fewer than five votes. 3.11.A This information shall also include eligibility and reason (if any) for ineligibility for each member of the longlist.

3.11.B In addition, for each category, the list of numbers of “reject” and “accept” votes for each work shall be published in descending order of number “reject” votes. The names of each work shall NOT be published alongside this list.

Mercy panel

Moved, to add Section 3.BA (Mercy panel) after section 3.B, and to amend Section 3.C.1 (Nominee eligibility voting) as follows:

Section 3.BA: Mercy panel

3.BA.1 Each Worldcon Committee shall appoint at least 3 natural persons and 2 alternates, and no more than 6 of each, to serve as a Mercy Panel for the following Worldcon. Presence on the Mercy Panel shall not necessarily preclude assuming other functions for any Worldcon. The procedure for such appointment shall be at the discretion of the Committee, and may or may not include voting.

3.BA.2 If for any reason there are not at least the minimum number of members and alternates from the prior year, the current Worldcon Committee shall appoint any which are lacking, using the same degree of discretion.

3.BA.3 The Worldcon Committee and the Mercy Panel shall remain in contact during the first round of voting, and the WorldCon Committee may elect to share preliminary voting counts as appropriate and practical. This information shall be treated as private, maintained with reasonable security precautions, and erased as soon as practical by the Mercy Panel.

3.BA.4 As soon as the longlist is published, the Mercy Panel shall proceed with all reasonable haste to rule on whether each nominee on the longlist should be subject to an eligibility vote as described in section 3.C.

3.BA.5 Any member of the Mercy Panel who has a conflict of interest regarding a given work shall allow an alternate to vote in their stead on that work. Any work which for this reason cannot be reviewed by at least 3 members of the Mercy Panel shall be subject to an eligibility vote.

3.BA.6 Nominees that should be subject to eligibility votes include any in the following categories:

3.BA.6.1 Those for which there is a reasonable argument that they contain harassment or libel against a living person or any past member of WSFS.

3.BA.6.2 Those which any member of the Mercy Panel considers to be likely to be offensive to the standards of at least 20% of WSFS members.

3.BA.6.3 Those for which there is some clearly-defined credible reason(s) to suppose that over 20% of the nominations were made based more on outside influences than on the nominator’s judgment of the nominee’s merit.

3.BA.7 Any nominees which are unanimously considered by the Mercy Panel not to fall in any of the categories enumerated in 3.BA.6 shall not be subject to an eligibility vote.

3.C.1: After the longlist is published, and after the Mercy Panel has ruled on all members of a category, each member of the administering Worldcon who had been eligible to vote in the nominations poll as described in 3.7.1 may vote on the eligibility of any of the longlist members not exempt under section 3.BA.

+2 against trolls

Moved, to amend Section 3.7.1 (Tallying of Nominations), Section 3.9 (Notification and Acceptance), and Section 3.11.4 (Tallying of Votes) as shown above in “3SV Base”; and also to add section 3.D as follows:

Section 3.D Addition of nominations during round 2

3.D.1 During the second round of voting (that is, simultaneous with the process described in 3.C, if present), all persons who were eligible to nominate in the first round of voting may add nominations as follows.

3.D.2 Each voter may add up to two (2) nominations per category.

3.D.3 Any additional nominations for the same nominee a voter has already nominated in the same category shall be ignored.

3.D.4 All new nominations must be for nominees that are on the longlist in the same category.

3.D.5 Postal mail ballots should be accepted insofar as it is practical given the schedule, but the Worldcon committee may if it chooses schedule this voting round in such a way that some members may not have ample time to submit such physical ballots.

3.D.6 All nominations added during this period shall be treated in all ways as if they had come during the first round of voting, except that they shall not be counted as violating the allowed number of nominations per category.

Extend finalists

Moved, to amend Section 3.C.3 and 3.C.7 of the 3SV proposal above, and to add section 3.C.A after 3.C.7, as follows:

3.C.3: The ballot for this round of voting shall present the voter with the following options for each longlist nominee: “Accept”, “Accept with extension”, “Reject”, and “Abstain”.

3.C.4: Any voter who does not submit a ballot will be considered to have voted “Abstain” on all nominees.

3.C.5: Any voter who does submit a ballot but does not explicitly choose one of the three options for a given nominee will be considered to have voted “Accept” on that nominee. This default choice should be made clear on the ballot insofar as practical.

3.C.6: Postal mail ballots should be accepted insofar as it is practical given the schedule, but the Worldcon committee may if it chooses schedule this voting round in such a way that some members may not have ample time to submit such physical ballots.

3.C.7: A nominee shall be eliminated and considered ineligible if it meets the following two criteria: the number of “Reject” votes it receives is greater than 20% of the pool of voters eligible under 3.C.1 and 3.C.2; and the number of “Reject” votes it receives is also greater than the number of “Accept” votes or “Accept with extension” votes it receives.

3.C.A: If the number of “Reject” votes and “Accept with extension” votes a nominee receives is greater than 10% of the number of voters eligible under 3.C.1 and 3.C.2; and the number of “Reject” votes and “Accept with extension” votes the nominee receives is also greater than the number of “Accept” votes it receives; and if the nominee is selected as a finalist; then the number of finalists in that category shall be increased by 1, up to a maximum of 7.

120 thoughts on “To Say Nothing of the Dogs; or, How We Confound the Hugos’ Third Slump (Hugo voting proposal discussion 5)

  1. The post has a number of proposals, and discusses each of them. But even for those of us who think each of the individual proposals is a reasonably well-thought-out idea on its own, it’s obvious that there will be a lot of healthy skepticism at the Business Meeting if they are all on the agenda at once. And some of that skepticism will (also healthily) focus on my own role in all of this; I doubt there’s a precedent for a relative outsider like me helping to craft multiple proposals over two years.

    For the stuff about me personally: I’m not going to leap to defend myself if people talk negatively about me in the third person. If you want to know how I respond, please ask me directly.

    For the stuff about the proposals: I think it might be useful if lurkers post quick comments stating where they stand on all of this. Anything from “I hate all of the proposals” to “I love all of them” to “I’m still not sure” would be interesting and I think these proposals are mature enough that it’s productive even if you have nothing “new” to add besides your opinion.

  2. Errata: 3.BA.7 should reference 3.BA.6 instead of itself, I think.

    Other than that, a swift read through gets an approval from this moose.

    I’m on the wrong continent, so can’t come to the business meeting.

    ETA: first pre-fifth.

  3. So, about my fundraiser.

    Last year, due to the generosity of many donors, I was able to attend Sasquan, help explain EPH, and see it pass by a 3:1 margin. I also went over my fundraising goal and stayed under my budget, so the remaining $380 of what I raised went to support Electology.Org (aka the Center for Election Science, CES).

    The CES is an organization dedicated to helping promote improved voting systems and voting reform. Aside from my work with E Pluribus Hugo, we’ve supported other awards such as the Webby awards; but our main goal is to see better election systems, such as approval voting, used in political elections. The English-speaking world has some of the worst election systems around, using horrible systems like plurality voting (aka first past the post) extensively. We’re dedicated to the belief that better voting systems can lead to healthier politics and better outcomes, not just from the point of view of one partisan ideology, but for the overwhelming majority of citizens.

    So, obviously, I think it’s a good cause.

    This year, I think it’s becoming clearer that EPH+ will be on the agenda. I hope it passes, but even more so, I think it will be useful to have an expert like myself at MAC II to help clarify how this and would work and how it will interact with the other proposals.

    So, I’m fundraising to do both of those at once. If I meet the initial $1400 goal, I’ll use that money to go to Worldcon. But honestly, after the amount of work I’ve devoted to this issue, I hope that I’ll be able to raise more than that, so that I can give the extra amount to the CES.

    I don’t think it’s a good idea to have “stretch goals”, though. I’ve done what I’ve done so far without any quid pro quo, and I think it’s best to continue on that basis. Whether or not I raise the money to attend MAC II and/or extra money, I will continue to participate in this discussion, and do the same extra work making graphs, reports, etc. that I would have anyway.

  4. The moose is right. 3.BA.7 should read:

    3.BA.7 Any nominees which are unanimously considered by the Mercy Panel not to fall in any of the categories enumerated in 3.BA.6 shall not be subject to an eligibility vote.

  5. So anyway, as I was saying: the fundraiser doesn’t have any stretch goals. But it does have rewards! Check them out! Including electoral fanfiction!

    (By the way, if you’re not a fan of seriously-amateur fanfiction, but you still want to give at that level, your reward can be that I don’t do that. You can protect one fictional election from me touching it with my grubby paws.)

  6. Kevin Standlee has noted that the standard for an amendment between passage and ratification is “no greater change” – and that this is not quite the same thing as “not a lesser change”. I find it quite plausible to say that “adopting EPH with Sainte-Lagüe weightings” is not a greater change to what we’re doing now than “adopting EPH with D’Hondt weightings” is…and that we could therefore go straight to EPH+ without having to have a year of just EPH first.

    Do people disagree?

  7. It seems to me more likely that EPH+ will be counted as a greater change; after all, there’s a reason it got named with a “+”. But if it could pass this year, all the better.

  8. Lurker chiming in here with my FWIW:

    EPH: Definitely yes.

    EPH+: Sure but mostly because I trust all ya’ll’s math.

    3SV: Strongly support. I think this is an excellent idea and empowers the voting membership to act.

    +2 vs Trolls: Originally I was “nice to have” on this, but the more I think about it, the more I like it especially beyond having it as a slate-limiting tool. Best Short Story is the most obvious category that this would be great for. Given the long tail problem well discussed before, having all of one’s nominations not even close to being a finalist can easily feel (unfairly) like you don’t have a say. This would definitely allow everyone to have a strong voice in the nomination of these categories. Plus, there is the great symmetry that categories where it would be more appreciated are often the ones with shorter works and an easier time becoming familiar with all the works on the long list. So where it is more beneficial, it has the synergy bonus of also being the categories where we can be the most informed.

    Mercy Panel: I am far weaker on this one. I can see the benefit, but, personally, having any sort of panel involved in making judgements pushes against the spirit of the awards. When I first saw it proposed, it felt like a “worst of both worlds” pseudo-jury. I don’t hate it as much anymore, but I’m certainly not excited about this one at all.

    Extend finalists: Have to spend more time thinking about this one, but when in past years I have seen how difficult it is for people to understand how “No Award” works, I see this as far more confusion than it’s worth. Not necessarily to administer, but for the population of voters to understand it and it’s implications and how best to use it (or not) to convey what they feel.

    I will keep an eye on discussion of these, but as of right now, if I were an attending rather than supporting member (full disclosure), I would not support Mercy Panel or Extend Finalists but would whole heartedly support the others.

    FWIW.

  9. I think EPH+ is a greater change in a mathematical sense. It’s not clear to me one way or another if it is a greater change in a parliamentary sense. I wouldn’t try to guess how Jared Dashoff would rule. Either way, somebody would probably call for a vote to overrule him.

  10. Mainly a lurker and I didn’t read the last couple of threads so this might already have been discussed, but you did ask for opinions.

    The mercy panel idea sounds terrible. Based on the proposed amendment, it’s not just about exempting a couple of works, it’s about completely subverting 3sv into a way for the worldcon administrators to call out works they don’t like and accuse them of being offensive / libelous / manipulated. “Panel of Shame” might be a more appropriate name.

    In that situation there’s no way a work named (or “not spared”) on that list wouldn’t get 20% of voters agreeing to ban it. It’s just a fig leaf over the previously rejected” administrator veto” option.

    It also removes one of the other useful functions of 3sv, ie identifying ineligible works.

    I’m still slightly negative on 3sv as a whole (I don’t share your confidence that people will only use it to screen out abusive nominees), but the mercy panel idea is just downright terrible.

  11. As for the other proposals, I think the +2 idea is a definite improvement to the 3sv idea (although I would still weakly oppose 3sv as above), but the extend finalists option is just adding complexity for no strong reason

  12. My two cents:

    EPH/EPH Plus: Definitely. Get Plus this year if at all possible.

    3SV: Yes. Get in in the pipeline and get it going. (Am I reading it right that you had from TWELVE to 15 finalists, all the way up to 18? Would that apply to certain categories, or all categories?)

    +2 Against Trolls: Yes. I think that would be especially effective in categories like Short Story and Related Work, which (as we’ve seen) are particularly susceptible to puppy poo.

    Mercy Panel: Unsure. If I understand this correctly, and applying it to this year’s Best Novel category, The Fifth Season, Ancillary Mercy, and Uprooted would be “mercified” by the panel, because they’re non-slate nominees, while Seveneves and The Aeronaut’s Windlass would be held over for 3SV? And then the second-round voters would give the final thumbs up or down?

    That just seems…eh. Making things more complicated than they need to be, when the first 3 proposals will do the same thing just as well.

    Extend finalists: This definitely seems to be more trouble and confusion than it’s worth.

    So: the first three proposals, an enthusiastic yes. Fourth, on the bubble, not terribly excited. Fifth, no.

  13. I’m for EPH and 3SV makes a lot of sense to me. I like the idea of there being a long list and then trying to narrow it down from there.

    I suggest rolling EPH+/3SV/2+ into one draft and calling it the Love Is Real amendment.

  14. Also, I tossed some money in your pot. If all these proposals are going to be introduced (or even 3 out of 5) I think you need to be there.

  15. @David Shallcross: EPH uses 1, 1/2, 1/3, 1/4, 1/5 as the divisors, and somebody pointed out a few days ago that if the divisors used in EPH were 1, 1, 1, 1, 1, that would generate the exact same results that the current voting system does. That implies that using 1, 1/3, 1/5, 1/7, 1/9 as the divisors in EPH+ would take us farther away from the current voting system, and thus adopting EPH+ would be a greater change than EPH itself would be.

  16. 2+ seems like a gap that a slate could easily exploit.

    I’d easily support a Highlander ballot (two or more identical ballots = one vote). I’m tempted to go so far as to propose the appointment of a Roman-style dictator for the suspension of the two-year rule for the duration of the Puppies.

    I’ll be at the business meeting this year, and I’ll be voting for everything and anything that’s brought up for a vote. I’m willing to throw the kitchen sink at this thing.

  17. So will the Hugo Packets come with a Captain Midnight Decoder Ring so we can figure all this out?

  18. I am not convinced that a rational strategic response to SDV by non-slate voters is to bullet-vote.

    It is certainly not rational if you consider informational limitations; and it is not clear to me why anyone would think it a particularly good idea in that position.

    Consider a true independent, not affiliated with the equivalent of a political party. This voter is characterized by low levels of use of information used to coordinate behavior, and uncertainty about the number of voters that share their precise opinions about which works are good. (High responsiveness to information used to coordinate would result in coordinated behavior, i.e., acting as a de facto party voter.)

    Most works are non-viable for the award. If you do not know which works are viable or non-viable, you can anticipate that most or all of the eligible works you nominate will be eliminated early; which means rational strategic behavior entails nominating as many eligible works as possible in order to maximize the chance of creating what will be (after the hasty elimination of all the completely non-viable candidates) a bullet vote for the most viable work on one’s ballot.

    Thus, the assumption of non-coordination and the reality of diversity of eligible work would make bullet voting highly irrational. I can construct a few scenarios in which a bullet vote is rational, but the ones that come to mind all involve coordination and good information about extant party structures.

    Commenting as an expert on the matter, I will say that a long list voting stage in between open nominations and the short lists is mathematically desirable, but that voting in more stages tends to create / compound turnout issues. YMMV.

    I will take a longer poke through the paper later and may have more to say then.

  19. Ken Marable: The role “No Award” plays in the Hugo voting system is complicated, as voting systems go. I am not surprised that you have seen difficulty in explaining it to people.

    I have found that for the most part, it is not easy to explain unfamiliar voting systems to people even when the system is very simple in mathematical terms.

  20. In my opinion, saddling us with an opaque, complicated system that, let’s be honest, at least half of us do not really understand and that nine people out of ten cannot accurately describe, is not the right way to go. In it’s own way, it’s as bad as what it is purporting to cure. And it won’t even fix the problem.

    I propose that if EPH is adopted, it has a one year sunset clause. That will give us time to get Kevin’s three-tiered approach put into place. That has a much better chance of working, and it is something that practically all of us will understand.

  21. EPH: Yes, absolutely. Without this we have two more years of hell.

    EPH+: Maybe. I want to see how an ordinary year is affected first.

    3SV: Absolutely. EPH/EPH+ aren’t enough by themselves. We need some kind of troll defense and this seems to be the only one people find acceptable.

    +2 vs Trolls: Undecided.

    Mercy Panel: Yes. This, pre-filled boxes on ballot or some other feedback from admin. 3SV is not used to let people determine eligibility as someone mentioned. If they aren’t eligible, they should be removed from ballot. Directly without voting. This panel is to let admins give feedback of which items that couldn’t have made it to the ballot without block voting (i.e, to differentiate between popular items on slates which would have made it without slated and those that only made it because of slates. Seveneves and The Aeronaut’s Windlass would only be “mercified” if they couldn’t have made it without the slates.

    Extend finalists: Have to think about it, but most likely no.

    Captain Midnight Decoder Ring: We need this.

  22. I feel like this suite of proposals is too tightly tailored around the problems of the last couple of years.
    Jameson, you say the multiplicity of proposals is an indication of robustness, because the mechanisms support each other and back each other up.
    But what I’m strongly reminded of is the way that software bugfixes pile up one on top of the other, calcifying the system into something that works just so, for increasingly arcane reasons that nevertheless can’t be argued with. The problem with those is not that they fail to achieve their function; it’s that they’re incredibly difficult to work with. They require expertise just to understand what’s going on — and modifications, no matter how minor, grow overwhelmingly difficult. When half the system’s design is going to circumventing one particular bug, that’s a very difficult design to work with any further.

    I really dislike the proposal for a Mercy Panel.
    As @KevinP noted, it fundamentally changes the nature of the 3SV rejection phase; particularly, it’s undermining 3SV’s contention that the community is the moderation mechanism.
    I understand the appeal, but this has all the strong-man conflict-sparking issues of a nomination-rejection jury, with none of the benefits of actually getting rid of bad stuff.

    I’m still gravely concerned about the Hornet’s Nest scenario.
    In a nutshell, the Hornet’s Nest scenario would be for griefers to slate previously-uninvolved parties, particularly volatile creators prone to outrage. This would prompt WorldCon membership to publicly reject those creations, probably amidst some degree of internet outrage, which has a strong potential to bring in these previously-uninvolved parties as newly offended and antagonistic to the Hugos.
    It’s a scenario that’s applicable to the Hugos even without any rules modifications, but 3SV could potentially make even worse, due to the longer list and lower entry barrier.
    I’m preparing a longer post where I describe this scenario in more detail and explain why I think it merits our attention.

  23. @Rich Lynch

    I propose that if EPH is adopted, it has a one year sunset clause. That will give us time to get Kevin’s three-tiered approach put into place. That has a much better chance of working, and it is something that practically all of us will understand.

    While I think 3SV is a mistake, without EPH it becomes worse.

    The nomination tallies of non block voted works are small. Sasquan had a record year. In a lot of categories organic long list works were still only receiving 30 to 70 nominations each.

    I believe 250 is a good estimate for block votes last year. There are both higher and lower estimates out there. I suspect, with imperfect slate discipline, 200 to 250 is probably a reasonable effective estimate though. If true, without EPH, you could afford to split slates two to three times and still dominate the long list. In about half the categories you could outright sweep, in many more you could lock 10 to 13 works.

    EPH mitigates that risk. Without it, or some other mechanism, we potentially end up right where we are at, 3 stages or not.

  24. @Standback

    It’s hard to imagine any system, including current, that isn’t susceptible to the Hornets Nest Attack. Best mitigation I can see is allowing a single one year eligibility extension for works voted off in the long list phase. It would allow works creators to make their case if a mistake were made, give a cooling off period, and speak to a commitment to fairness. Down side I can immediately see is works creators campaigning for a year knowing they had a second bite and possibly solidifying a block of voters. Any other ideas for mitigation though?

  25. Giving EPH a one-year sunset clause seems absurd when it takes two years to effect any change at all (must be ratified at two consecutive years’ business meetings). It would be impossible to ratify its continuation without necessitating a hiatus in its efficacy. That seems needlessly disruptive.

    I’m not against sunset clauses on principal, but they need to be structured so that the choice as to whether to let the sun set on the measure is a real choice, and not a fait accompli.

    (Or am I missing something?)

  26. The work that has been done is impressive and a credit to everybody involved. I’m not really convinced by the proposals compared to a panel that just chucks out the obviously harassing works (with maybe +2) and lets EPH and No Award do the rest. But…there isn’t much of a consensus on that and people have genuine concerns about whether such a panel could work or whether anybody would be willing to staff it.

  27. My lurkerish thoughts-

    EPH – Yes, absolutely. Even is it only mitigates the harm slightly, it doesn’t increase the complexity at the voter’s end.

    EPH+ – Yes. More harm mitigation, still no more complexity for the voter.

    3SV – Yes, but. Here we start requiring more of the voter, with a process that requires high-information and highly motivated voters. These can be thin on the ground. However, I can’t think of a way to fix the defamatory finalists problem that doesn’t require unprecedented and unwelcome admin powers *or* very engaged voters, and the voters have shown they care.

    +2 vs trolls – Yes, but. My reasoning is the same as for 3SV.

    Mercy Committee – I am not sold on this. Is it mostly there as a hedge against bad actors in 3SV? I might accept it with a sunset clause, if it doesn’t prove necessary.

    Extend Finalist – No. This is adding complexity to the voter’s task for only a questionable gain, I think.

  28. summarizing thoughts I’ve had throughout the discussion/past two years (but without rigorous analysis) I favor naked 3SV as I believe it is the most effective, as well as the least complicated.

    I’m not so worried about how the system will be gamed following the intro of 3SV as I believe that the moral imperative has been amply demonstrated through No Awarding, and most, if not all of those who ‘had’ to take that route will happily shift their focus the second round.

    Further, gaming attempts have been, by their nature, open and public and easily identified; 3SV may drive some underground (to mailing lists for example) and that’s a good thing as it limits their reach and impact and works nominated by such ought to be fairly easily identifiable if they make it to the ballot.

    I do think that EPH needs to be sunset; withdrawn once 3SV is implemented; I don’t think we need the complication.

  29. @Kevin P: You think that the Mercy Panel is a bad idea because once they’d decided, the outcome of 3SV would be a foregone conclusion. You have a right to your opinion, but I do disagree with this reasoning. It would take a unanimous decision to grant mercy. If the panel couldn’t agree on a work, why should the voters’ decision be a foregone conclusion?

    @Bonnie McDaniel: Thank you for your donation. Note: if somebody has an unused attending membership lying around, that would be a useful donation too.

    @LZ: You say that a slate could easily exploit +2. My simulations suggest that +N would clearly help non-slate voters more than a slate for N up to 2. (Note: Under the NDA, I am not allowed to use the real ballots I have for simulating ideas unrelated to the ones up for ratification this year, such as +N; so these simulations are much cruder than the ones that make the pretty graphs.)

    @tomas: You argue that bullet voting would not be the rational response to SDV unless you were a slate voter. I agree that you’re right for most voters. But if there’s one work you clearly care more about than others, then bullet voting might become rational under SDV, significantly more so than under EPH or the current system. Such voters are a minority, but not all of them are slate voters, and getting the full range of their opinions is useful (for instance: some of them may be authors!).

    @Rick Lynch, etc: sunset clause. Others have said this, but I’d like to emphasize it: 3SV without EPH is vulnerable to a slate taking over the longlist. Would the puppies manage it? Who knows. But the system would remain clearly vulnerable.

    @standback: you make some good points.

  30. 3SV + EPH are of a piece. We need both for best effect.

    Mercy Panel: A panel will be the target of every grievance, rational or otherwise. I’m not against the idea (or against banning malicious parties outright), but there’s enough division within the community to make a panel an exercise in argument. Any argument will be exploited. I don’t see any good coming from a panel, however constrained.

    +2 against trolls: 3SV addresses the long tail issue, and also gives us an opportunity to remove unqualified work before the final vote. I’m wary of adding anything that may dilute that effort.

  31. The document should make it more clear that 3SV will not work without EPH. Otherwise some people will get the idea that 3SV can replace EPH.

    I’ll vote for EPH+ at the Business Meeting and for 3SV, but against the other proposals. And I’ll vote against treating EPH+ as a minor modification to EPH, for the mathematical reasons I’ve given before (and which @Joshua K. repeated above).

    From last-year’s items, I’ll vote to ratify EPH but against 4/6 unless it’s modified to 5/6. I think EPH with 5/6 is the absolute best we can do to help in time for Helsinki.

  32. 3SV does not address the long tail issue. It addresses the troll issue.

    +2 is the only thing that addresses the long tail.

    3SV is the only thing that addresses the troll nomination issue.

    EPH / EPH+ give broad-spectrum help but do not definitively resolve either of the above issues.

    Extend finalists is the only thing that fully addresses “kingmakers” (or “area defense”). Other things help, but this can fully solve the issue.

    The mercy panel, it would seem, is a solution without a clear problem. “Negativity” is too vague as a problem, and thus not something that can be clearly solved.

  33. Negativity among the majority of Hugo voters is only a theoretical problem in most of our minds, I think. (As opposed to the actual actions of griefers.) We like to think better of ourselves than perhaps is justified; also, I have *no idea* at all who ought to be on such a committee. Who would accept being on such a committee. But I *can* imagine the amount of vitriol that committee would spawn.

    Still, I could be persuaded, especially if there was a sunset clause.

  34. @Jameson Quinn: 3SV does not address the long tail issue. It addresses the troll issue. +2 is the only thing that addresses the long tail.

    3SV takes a set of works published in a year and reduces it to the top ~15. That provides a strong set of nominees going into the final vote and covers the long tail. Being able to reject unqualified nominees covers the trolls.

    I misread +2 against trolls as being two new works rather than two from the long list. Hm.

  35. 3SV takes a set of works published in a year and reduces it to the top ~15. That provides a strong set of nominees going into the final vote and covers the long tail. Being able to reject unqualified nominees covers the trolls.

    You might just as well say that the current system “takes a set of works published in a year and reduces it to the top 5. That provides a strong set of nominees going into the final vote and covers the long tail.”

    Without +2, there is no way for the vote to consolidate before the finalists are chosen. If none of the top 5 are eliminated, they will be the same as with the current system, except that round 1 would be shorter. Only +2 (or +1) would let voters who’d chosen only long-tail works in round 1 help decide which of the longlist should be finalists.

  36. Rich Lynch on May 31, 2016 at 9:04 pm said:

    I propose that if EPH is adopted, it has a one year sunset clause. That will give us time to get Kevin’s three-tiered approach put into place. That has a much better chance of working, and it is something that practically all of us will understand.

    I expect EPH as it stands now to be saddled with a “must re-ratify annually for three years or it sunsets at the end of this year” clause. This will allow each year’s WSFS Business Meeting for the next three years to review the situation as it stands and take action immediately. Assuming EPH is ratified in that way this year, if any of the next three WSFS Business Meetings votes against re-ratification, EPH disappears immediately and is not voted upon again the following year. The Constitution would then revert back to the existing first-five-past-the-post nominating system.

    Greg Hullender on June 1, 2016 at 8:31 am said:

    The document should make it more clear that 3SV will not work without EPH.

    I am unconvinced by this. The arguments that appear to be in the form of “but Griefers will just split into three groups and sweep everything” have IMO an unstated assumption that Griefers are as much as a majority of the entire electorate. This is demonstratively untrue, and if they are, they should win anyway.

  37. @Jameson Quinn:

    You think that the Mercy Panel is a bad idea because once they’d decided, the outcome of 3SV would be a foregone conclusion. You have a right to your opinion, but I do disagree with this reasoning. It would take a unanimous decision to grant mercy. If the panel couldn’t agree on a work, why should the voters’ decision be a foregone conclusion?

    I think it would be a foregone conclusion because any work that failed to pass the Mercy Panel, and thus was sent on to an eligibility vote, would be presumed by many voters to be harassing, libelous, offensive, or nominated based on outside influences.

    I would not be surprised if most of the Mercy Panel’s votes were either unanimously to allow a work onto the ballot, or unanimously not to pass the work and send it to an eligibility vote.

  38. @Kevin Standlee

    If the top 15 are selected the way they are currently, then you have a lot of small clusters of nominations around organic works. Splitting slates against a fractured pool of nominations allows dominating the long list given sufficient griefers. It doesn’t take any big grouping of griefers or even close to a majority. It’s a fraction. It’s the same situation we have right now where a small group can dominate a short list of 5 nominations by concentrating their votes. Maybe I’m being particularly dim (seriously) but I don’t see how 3SV protects against that. If a small group can still dominate the long list it doesn’t matter whether every WSFS member votes in the second round. If the 15 long list nominees are the same old griefer dog food does it matter that we get to pick and choose which flavor of Alpo suits?

  39. I’ve mentioned this before, but I’ll repeat here:

    I’m uneasy about the +2 proposal, for consolidating nominations around the longlist.

    My issue with it is that it turns the longlist into much, much more than a spam filter. In a normal year, with no significant trolling, it’s just going to be a good list, attracting voters and attention and campaigning, and that’s awesome. It’ll be an honor to get on the longlist. BUT: that’s an honor that’s going to be reeeeally easy to get in on.

    I love the idea of a longlist; the long tail is probably where all the stuff I love best lies. But I think, particularly with our current experience, that we need to be very wary of introducing a new Hugo honor that’s trivially easy to influence.

  40. I expect EPH as it stands now to be saddled with a “must re-ratify annually for three years or it sunsets at the end of this year” clause. This will allow each year’s WSFS Business Meeting for the next three years to review the situation as it stands and take action immediately. Assuming EPH is ratified in that way this year, if any of the next three WSFS Business Meetings votes against re-ratification, EPH disappears immediately and is not voted upon again the following year. The Constitution would then revert back to the existing first-five-past-the-post nominating system.

    This is a good idea if you expect that in the long run we will not want EPH. But I don’t see how it’s a good idea otherwise. Annual re-ratification, sure. But why would you structure this so that if it’s ever not ratified, and that turns out to have been a mistake, it would take 2 years to get it back in place? If it gets re-voted every year, that should happen either way, until the result is the same for 3 years in a row.

  41. @Stoic Cynic: To dominate the longlist, the griefers would need to be able to marshal 3*N nominators, where N is the number of nominations given to the most popular non-slated piece.

    For Best Novel, that’s probably a lot.

    For short fiction and Best Related Work, that’s not a lot at all.

    @Kevin Standlee: For the record, I strongly favor EPH with or without griefers, with or without slates. It’s a desirable mechanism which improves the balance and diversity of the awards, with no added complexity to the nominating membership.

    It’s a great measure against stuff like BDP:SF being full of episodes from one popular show, and in general it nudges the shortlist to maximize the number of participants who are pleased to have gotten something in.

    I wouldn’t want to have it sunset, even if the griefers all disappear tomorrow.

  42. I am unconvinced by this. The arguments that appear to be in the form of “but Griefers will just split into three groups and sweep everything” have IMO an unstated assumption that Griefers are as much as a majority of the entire electorate. This is demonstratively untrue, and if they are, they should win anyway.

    No. We can look at the published data to show this is not true.

    Let’s say that the griefers win if they can get at least 12/15 of the longlist; that is, if they can ensure that there will be 3 or fewer non-griefer works among the finalists. In order to do that, they need 2 and a half to 3 times as many voters as the 4th place organic work in a category. Look back to prior years and you’ll see that, aside from Best Novel, that would take only a couple hundred voters in most categories; far from a majority of ballots.

  43. To dominate the longlist, the griefers would need to be able to marshal 3*N nominators, where N is the number of nominations given to the most popular non-slated piece.

    Depends what you mean by “dominate”. If you mean 15/15, then you’re right. If 12/15 still counts as “domination”, then N can be the number given to the 4th place non-slated piece, which is usually dramatically lower than the first place tally (under 20% as high in some cases, and more commonly around half).

  44. @tomas: You argue that bullet voting would not be the rational response to SDV unless you were a slate voter. I agree that you’re right for most voters. But if there’s one work you clearly care more about than others, then bullet voting might become rational under SDV, significantly more so than under EPH or the current system. Such voters are a minority, but not all of them are slate voters, and getting the full range of their opinions is useful (for instance: some of them may be authors!).

    If you care about one work sufficiently more than others, bullet voting becomes rational in nearly any system. However, provided you care at all about the downballot works, good information is required to make the expected payout positive.

    Once you have coordination, you can make large leaps in point value (from X/n or X/(2n+1) to X, in the most extreme case); which in turn can produce large changes in the elimination process. That is similarly true for SDV, EPH, or EPH+, and the circumstances are generally going to look quite similar: You need to know that your coordinated group of voters is in a weak position, and you need to be able to get them to change gears in the same way.

    If your information is off, this can backfire badly, and it’s not going to be a priori obvious when you are in that situation. (OTOH, while the risk of backfire is still there, it’s going to be very obvious to anyone trying to break EPH+ that they can increase slate power by boosting their total number of points on the board in the earlier rounds.)

    You should expect that most strategic voting will be in terms of which works to name at all – trying to make sure all of your votes are for viable candidates, rather than your most preferred eligible candidates.

  45. “I am unconvinced by this. The arguments that appear to be in the form of “but Griefers will just split into three groups and sweep everything” have IMO an unstated assumption that Griefers are as much as a majority of the entire electorate. This is demonstratively untrue, and if they are, they should win anyway.”

    I do not believe that they need to be a majority or even close to a majority of the entire electorate for the categories with a very long tail and few clear favourites.

Comments are closed.