@DavidZ Also, why not? Isn't that how community policy-making is supposed to work?

@DanielSank No, it's not. E.g. that leaves open the door for a bunch of terrible suggestions that are worse than the current policy, in which case you wouldn't want to implement any of them. Anyway, policy changes are made by consensus, not by popular vote. (At least that's how we've historically done it most of the time, and I think this case also calls for the consensus approach)

@DavidZ I see. You're saying that there could be ten answers each with +1 score and one with +2, and all of them have ten downvotes, for example?

@DavidZ How do we measure/define consensus?

@DanielSank Forget about score. Scores don't tell you whether a proposal is good or not, only whether it's popular (which is jsut a contributing factor). But if the scores did directly reflect how much better than the status quo a proposal is, then yes, that would be exactly what I'm talking about.

@DavidZ How do you judge goodness if not by votes?

@DanielSank Consensus is when most people can agree on something. Like a compromise.

I have to say I also think votes are not a good measure.

@DavidZ And how do you measure that on Stack Exchange?

@Danu What is a good measure?

> But if the scores did directly reflect how much better than the status quo a proposal is, then yes, that would be exactly what I'm talking about.

@DanielSank Mostly comments and similar feedback, in both cases.

@DavidZ Wait a second. If a meta answer with some policy proposal has +30 upvotes and -5 downvotes, but there are three comments criticizing it and zero comments supporting it, how do you judge the situation?

@DanielSank Depends on the comments. If the comments bring up valid objections pointing out the negative merits of the proposal, then it looks like a popular but bad proposal, so I likely wouldn't support it.

@DavidZ (assuming they can't be fixed etc.), right?

wait, so you are judging these objections as "valid"?

that doesn't seem like community decision. that seems like community suggestion and moderator decision.

@DavidZ Who judges that. You?

Isn't there a separate room for berating of moderators by now? :P

@Danu If they get fixed then I might change to supporting the proposal

@DavidZ Quit acting like a king

@DanielSank Everybody

@DavidZ How?

@BernardoMeurer This is such an unconstructive comment.

What mechanism is used to indicate "everybody's" opinion?

that escalated quickly

If you don't have anything useful to say, just don't interfere with the people who do.

^

@BalarkaSen, because it doesn't make sense. though you're right, things have been coming to a bit of a head recently.

@DanielSank Again, comments. Everybody goes through the process I described. When you see other people bringing up valid objections to a proposal, you endorse those objections.

wait, quick rewind:

@DavidZ You are dodging my question.

how in the universe do upvotes on meta not constitute community approval!?

^

you guys always, always, always, always say that upvotes on meta are about whether or not someone agrees on a stance.

@DavidZ And once again, you do?

@DavidZ And what measures whether the proposal is good or not?

Could you please answer this question without using circular logic?

@heather I often find that they are not thoroughly considered (from my own experiences).

@DavidZ community approval is what matters. not what the mods think about whether or not a proposal is good. because that's really what you're saying here.

@Danu That's your judgement, which is fine, but we need some way to quantitatively measure approval of the entire user base.

@Danu Why does that matter? A vote cast thoughtlessly is a vote nonetheless

you're saying: yeah, community approval's great and all, but it's not always right, so i'm gonna step in and make sure it is.

@heather That's what it sounds like, but I'd like to try to understand DZ's thoughts more carefully.

@heather Yes, it sounds that way. I am puzzled and a little worried.

I think some of you (heather, Bernardo) are willingly misinterpreting DavidZ's comments as if he will singlehandedly decide anything at all. He's pretty clearly not saying that.

@DanielSank okay, I'll take a step back.

I think it should be a good mix of community approval/popularity and community well-being/admin-approval.

@BalarkaSen What's an admin?

@BernardoMeurer But one that should not weigh as much in deciding on policy.

administration, not administrators. Aka mods.

@DanielSank It's like making a measurement: there is a true value, and then there are imperfect procedures we can go through to try to uncover that true value to some level of precision. Similarly, there is a "true goodness" of a proposal, and meta debate and consensus-building is the imperfect process we go through to try to determine the true goodness of that proposal.

I mean it's potentially possible we all want PSE destroyed; does that mean the moderators should conduct that operation?

@Danu And how will you measure the amount of thought that goes in an anonymous vote?

@Danu for the record, I'm not willingly misinterpreting everything. DavidZ is saying he/the mods will decide.

@Danu I don't know about "willingly" but that does seem to be true.

@DavidZ You are not answering my question.

that's the only interpretation i can come up with from reading his messages.

@heather You should read back on the conversation.

I will ask it again.

Though, @DanielSank's question is crucial.

@DanielSank Dat is the question

what is this? i interpret the comments in a way that's fairly non-negotiable, and people jump on me for "willingly misinterpreting" and misreading the conversation? Please, can we all just take each other seriously? I have no reason to willingly misinterpret anything. If I'm wrong, what did DavidZ mean?

@heather I am not (except to the extent that some changes can only be implemented by moderators because of technical restrictions in the system, so those probably aren't going to happen if the mods are overall against them).

Anyways, this is exactly the kind of behavior that I think should not be weighed as heavily when making big decisions. I think Daniel is the only one of you who has said anything useful and I feel like heather and Bernardo are not contributing constructively. That's all I'll say for now, I'm not interested in wasting more time on this.

@DavidZ I must also ask, do you notice that several users are finding your comments here confusing and even objectionable?

@DanielSank As I said, comments. If you're looking for a formula where you can plug in numbers and determine whether a proposal is good, I can't give you that. There isn't one.

@Danu JD seemed to think no one was constructive as well, interesting

There's no reason to make this all unnecessarily heated. Y'all should chill out.

@DavidZ "so those probably aren't going to happen if the mods are overall against them" - you have to accept that as mods you are carrying out community policy. if you all disagree, sorry, but the community agreed on it. that seems directly against the purpose of mods.

@DavidZ So again, if a post has 30 upvotes, 5 downvotes, and three critical comments, how do you decide what to do?

@Danu without saying anything about what we're discussing, what exactly am I doing? I'm just trying to understand what DavidZ's trying to say. How am I coming across?

@BernardoMeurer Daniel is being constructive. This is another comment of yours that is honestly nothing more than a childish "poke". You should probably take a step back and get over it.

@Danu Oh no, you called me childish :(

@heather We can discuss it some other time.

@heather Well, the mods also form part of the community. There are six of us. If six active members are strongly opposed to a proposal, it would be unlikely that that proposal reflects community consensus, right?

@Danu okay. I'm just asking because I'd like to continue in this discussion, without being rude.

@heather This comment, eg, is overly heated.

@DanielSank I read the comments.

IMO

That's the issue with any variation on the consensus process. It's always unclear as to why decides when a consensus has been reached, or what that intangible threshold is on any case. And that's a problem on maybe 75% of meta discussions, when things aren't clear.

I think DavidZ is trying to be like a Quaker-based facilitator, which is probably the optimal SE strategy, but the boundaries of that are not clear.

@BalarkaSen fair enough.

@heather Yeah, sure. The atmosphere has now been ruined for the time being so I don't see a point in continuing right now.

@HDE226868 I don't trust Quakers

@DavidZ if there are enough upvotes/comments/what have you on a question for there to be a community consensus, i would consider that "higher up" than 6 active mods, yes.

@HDE226868 Eh, maybe something like that

so no, I think that all mods can disagree but a proposal can still be community consensus.

@heather And what I'm saying is that there cannot be a community consensus that six active participants in the discussion are opposed to.

Stuff about site development is a good philosophical problem. Discussions about that need not necessarily turn into a "fight for freedom against the ruthless moderators". Who am I talking to...

@DavidZ Not really a perfect analogy in this case.

@DavidZ there most certainly can if enough people are in the discussion (in my opinion, anyway).

@BernardoMeurer I go to a school founded by Quakers, so I'll ignore that remark.

@BalarkaSen I most sincerely hope this discussion does not come across like that.

@HDE226868 I don't know what they are, so I'll ignore your remark

@HDE226868 Perhaps. In an online situation where communication bandwidth is limited, I suspect the Quaker model is less effective. Some kind of direct quantitative measure of consensus would seem rather helpful.

@BernardoMeurer, Quakers are a religious sect.

@heather Several messages certainly do, to me.

@heather That much I could've guessed

@Danu are any of them mine? if so, which ones?

@heather Well, OK, you can hold that opinion, but my opinion is that that is incorrect. Furthermore your opinion clashes with the decision-making process I'm used to on this site. So we seem to have a more fundamental issue to resolve here.

::makes popcorn::

@DavidZ First of all, presuming that all of the mods oppose a proposal seems like a strong limit unlikely to pertain to the real world.

@DanielSank That's true. I'm using an extreme case to make a point.

@DanielSank I view the meta system as a kind of discussion like that that's smushed out in time. People post answers at different points, which then get discussed, and eventually people agree that the options they like have been posted, at which point things move on. Over long periods of discussion, which seem to be the norm on SE, it's a bit harder to follow the original Quaker structure.

@DavidZ yes but your point boils down to an argument about inferring majority opinion from a sample.

At any rate, I was emphasizing the role of the facilitator.

@DanielSank Not majority opinion, exactly; I'm talking about consensus.

@DavidZ No you're not. You used the example of the six mods to illustrate your point. If you were interested in consensus, then whether those six users were mods or not wouldn't matter.

Who wants some popcorn?

@DanielSank It doesn't matter.

me

@skillpatrol, me

Me!

:P

I can finally eat it again

It's so good. I had two servings at Chicago

@DavidZ um...i kind of think it does. Let me put it this way: would you listen equally to an objection from John Duffield as one from John Rennie?

@heather Is John Duffield an active participant in the discussion?

Is his objection sensible?

@DavidZ sure, both are

@JohnDuffield Come from the shadows

@heather Then yes

@DavidZ is it? is John Rennie's? who's judging "sensible" here?

@heather Everybody involved

@DavidZ how?

what is the exact medium people are judging sensibility through?

@DavidZ Oh come on that's ridiculous. There isn't enough traffic in meta to gauge whether the community agrees with a comment!

@heather I'm not sure what you mean there.

You can't downvote comments and long comment chains are actively discouraged.

@DanielSank Not on meta

@DavidZ comments, votes, questions, answers - how are people judging the sensibility of a suggestion on meta?

what are people using to convey their impressions of a suggestion?

@heather As I've said, mostly comments

@DavidZ okay. so we have comments. but that's not how meta works. if people agree, they upvote. if people have a disagreement or want to discuss more, they comment and/or downvote. your system is predisposed towards people who disagree with the idea. If you see three negative comments, you get a negative picture. But what if those were three out of four negative downvotes on an answer that got 20 upvotes?

@heather Then, if I understand the situation you're getting at, there are three valid objections to that proposal. Three reasons not to implement it. That is not a consensus.

@heather This is a very, very important comment.

Thanks, Heather.

I note that DZ tends to take a very conservative approach to these things.

@DavidZ uh...what? 20 vs. 4 is a clear consensus in favor of the idea.

DZ, it seems that given a proposal that improves on the status quo in 4 ways, but has 1 flaw, you'd rather keep the status quo.

As I've said, I disagree with that approach.

@heather No it's not. It's a vote, it's not a consensus. You can't can't measure consensus by X vs Y.

@DanielSank Well, that's not what I'm saying here.

@DavidZ Actually I think your calculus is wrong there.

@DavidZ erm...than you're measuring consensus by comments which are predisposed negatively? i am very much confused.

^

Suppose we have an issue with facets A, B and C.

Suppose we make a proposal which improves on A and B, but has a problem with C. Suppose there was already a different problem with C.

I would say that as long as the new problem with C is not worse than the old one, the proposal should be accepted.

I have repeatedly had the impression that you @DavidZ implicitly disagree with this.

Yes/no?

@heather Consensus is closer to an absence of strong objections to a proposal, than it is to a majority of support for a proposal.

hmm, that reminds me of that software maxim: perfect is the enemy of good @DanielSank

Yes, that's why I keep repeating it.

@DanielSank No, I do not disagree with that.

@DavidZ If that's our working definition of consensus, then I think consensus is not a good criterion for policy changes.

Here's why:

Suppose we have an issue with facets A, B, and C.

Suppose, under the status quo, I have a strong objection to facet C.

Now someone comes and proposes a change that improves on A and B.

Suppose another user has a strong, good objection to facet C of the proposal.

You guys are going insane

It is clear, to me, that the proposal should be accepted as long as the objection to the new facet C is not much more important than my objection to the old facet C.

Have some popcorn @BernardoMeurer

@skillpatrol I have alcohol, keep your cereal

@DavidZ in short, I think your calculus undervalues users' present objections to the status quo. In this way, your calculus tends to preserve the status quo, even if it is known to be problematic.

@DanielSank Sure, and in a consensus-based decision-making approach, that's exactly what would happen.

@DavidZ Not under your definition of consensus. I encourage you to re-read the chat log at this point.

@DanielSank No, that's not correct. If that's the impression you've gotten, something has gotten lost in the communication of what I mean by consensus-based decision making.

@DavidZ from your definition of consensus i must admit that you come across with the logical conclusion being that that proposal would be rejected due to the objection.

@heather That's why I said consensus is closer to the absence of objections than to a majority vote. But it is not the same as the absence of objections.

further, i must admit your distinguishment between consensus and majority opinion bewilders me.

@DavidZ that just made it even more unclear, i'm afraid.

argh reading transcript, feel DS/H are into challenging authority no matter how reasonable & getting hung up on hypothetical situations, just let the meta posts on HW jihad policy play out, see the votes, & cross the bridge when you come to it. also wish this HW thing would move to meta room :(

@vzn Ah, yes, good idea to move to the meta room - @heather and @DanielSank (if you're still around) would you agree? I can migrate the rest of the conversation too

@DavidZ sure, yeah.

Moderators are chosen by popular vote, not by consensus

@BernardoMeurer The nomination phase is supposed to be the consensus-building

I gotta take off for a few minutes, sorry, but when I come back I'll move this discussion to the meta room

and we can pick it up there

@DavidZ consensus is a difficult word, thx for mentioning it, think the resistance/ opposition should think carefully about (achieving) it :|

← 186 messages moved from The h Bar

Did this migration automatically invite me? :P

@Danu I dunno, maybe

(you didn't invite me by hand?)

I did not

OK :)

Then it did.

Sorry about all the "popcorn" comments.

and the gif

@skillpatrol, no need to apologize as I see it - it was funny =)

Nothing wrong with injecting a bit of comic relief into a policy discussion

Anyway, where were we? Or if you'd rather leave this for now and continue it another time, that's fine with me too

eh...unless @DanielSank is back/willing to talk, i think it's best we left it for another time...he brings up good points i think.

also, @DavidZ, I have an apology to make.

late to the party, I just got home

Welcome.

some of the statements I made were overly provocative and detracted from the conversation. I will try in the future to avoid that, and hope that what I meant to convey came through even though I said it poorly. @DavidZ

@DavidZ maybe your point is that a populist proposal may get tons of upvotes from people that rarely use SE, and they dont know how things work

they dont understand the big picture

for example, a proposal to make all questions on-topic might seem attractive to low-rep users that just want to post their homework

and get it answered

and such a proposal might get >100 upvotes, and <20 downvotes

so it would look popular, but it would be something bad, something we dont want

however @AccidentalFourierTransform, I'm not sure that's right, because those low-rep users don't come to meta.

@heather Well, apology accepted, though I don't think that was anything out of line that really requires apologizing for. This tends to happen in any intense discussion. It is hard to keep everything on point when discussing in real time.

@AccidentalFourierTransform I wouldn't put it quite that way.

Like @heather said, the real low-rep users who have very little investment in the site probably don't come to meta, for the most part. (It would be great if more of them would, but reality is not like that.)

@heather well such a hypothetical post could end up in the hot meta posts tab

but it wouldn't, i don't think. the people who do come to meta are the ones who'd downvote into oblivion, and so it wouldn't end up anywhere near the hot meta posts tab.

in fact, it'd probably end up deleted.

yeah, perhaps

I was trying to understand under what circumstances a bad proposal could attract many upvotes

i.e., I was trying to understand D.'s point that many upvotes in meta might not be correlated to general consensus

It's more that people tend to upvote (or downvote, though I suspect less commonly) proposals that serve (or run counter to) their own self-interest, without realizing all the consequences for the community at large. This is where the comments come in. Each individual voter is likely not to identify the reasons a proposal might be good or bad for the community, but in a large group, there's a decent chance that most of those reasons will be pointed out by someone.

yes, i think that is the thing i really don't understand along with his definitions of consensus vs popular opinion etc

@heather What I'm getting at with "consensus" is the idea that, in theory, everyone should be able to agree on a certain proposal being the apparent best way forward for the community as a whole, even if some (or many) of those people would individually prefer something other than that proposal.

This requires discussing the reasons that the proposal might be good or bad for the community. If we just go by votes, the votes often don't reflect whether the proposal will truly be good or bad, they only reflect whether people like it or not.

@DavidZ Of course, I do agree that comments are a very important part of the discussions in meta

In a consensus-based approach, people will start out by posting comments that point out their objections to the proposal, and other people respond with comments that point out why those objections are not so bad (if that is the case).

And after some back and forth, we should wind up with the people who objected commenting that "well, I think this is a problem, but I can live with it" or the people who defended the proposal saying that "okay, you're right, this is probably more trouble than it's worth" or stuff like that.

Basically, everyone comes to accept the same position, even if they don't really prefer it.

Sounds reasonable. I guess we'll resume the discussion later on, when D.S. comes back.

Yeah, guess so.

I have nothing else to say right now

I do want to mention that, in reality, it can happen that people are stubborn and simply refuse to come around to the other side, even when that other side is supported by everybody else - so you get one person with a strong objection vs everybody else who has already come to a consensus. In these cases, everyone should really consider whether the objection is valid, but if not, as a last resort, the rest of the community can tell that one person "well, sorry, we're going on without you"

What about "popular opinion"?

i think consensus is a bit of a poor approach then for community - because that's a unanimous vote, basically, after discussion. that doesn't really happen in real life.

@heather Not exactly, but it comes reasonably close.

As I recall, that's how the chat ban on politics got overturned, for example.

i might also point out that this approach hasn't always been carried out, i.e., I'd say Daniel Sank's post about changing the wording is pretty close to consensus, and no action has been taken yet.

@DavidZ mmm, i wouldn't - there were people in favor of the ban. However, many disagreed and came up with good arguments as to why the ban was bad.

@heather Yes, and then the people in favor of the ban largely relented and decided that they could live without it. Presto, consensus. That was my understanding, anyway, though it was kind of hard to tell what was really going on at some points.

What happened here^

I can't be sure, but I suspect my attempts to explain what I mean by "consensus" didn't get through to Daniel. I'm not sure at what point in the communication process it broke down.

@heather consensus is not the same as unanimity. consensus is about finding the key "sticking points" and trying to compromise on them. its kind of the opposite of "trying to whack each other over the head" :P

good point.

@vzn That sounds like a good way of putting it (well definitely the first part at least)

so, we reached a consensus on the meaning of "consensus"?

Hah :-P I don't know if we're there yet, but that probably is worth going for

I concede.

have been following this HW debate from afar/ sidelines so far. it even consumes major time on the recurring meeting, maybe more than once. it does not seem fully resolvable. its similar to a problem that recurs across stack exchange, different sites. Computer Science has similar issues. not sure that a meta policy can fix a recurring-bordering-on "chronic" issue. not sure why DS apparently thinks so vehemently it could.

you can't blame him for trying

from own experience, have found that most main site users are probably rather minimally influenced by meta! many rarely/ never visit. its for users who really care about the site. so metas big purpose/ potential is to get the "big players" (high rep users/ mods) on the "same page"...

@vzn I don't know that anyone thinks we can fix the homework problem, in the sense of getting people to stop posting bad homework questions outright. But what we can do is provide better guidance to the community as to how to identify bad homework questions (or similar, bad non-homework questions) and better guidance to people posting such questions on how to edit them in a way that makes them on topic.

onecommunityranch.org/…

@DavidZ sounds like a reasonable goal. then, my question is, what are the points of contention? DS did write a very comprehensive policy idea, think it would be good if mods/ high rep users engaged with issue clarified what is potentially not ideal about it. think he has a point, 30 votes on a meta post is high, "nearly" a consensus by votes at least...

@vzn well, like I've been saying, you can't establish a consensus by votes alone.

Hey guys, my prof recently wrote $\int dx x$ and meant $\int x dx$ is this just a different notation? Did you already encounter the first one?

try here

@Felix.C but yes: it is just different notation

they both mean the same thing

mathematicians tend to use $\int x\ \mathrm dx$, while physicists tend to use $\int\mathrm dx\ x$

but conventions vary

There's no consensus :P

lol

@Felix.C see

@DavidZ ok. just read DSs meta Q/A from oct 2015. o_O it all sounds reasonable to me. hes thought long & hard about it, maybe more than anyone. kind of agree with his point that users like his basic idea & think its certainly enough to guide suite administrators in just implementing it, subject to their own reasonable adjustments. (not saying this will actually happen!)

Ok, thank you! Imho $\int x dx$ is much easier to read since you know where the integral ends, but well....

I agree.

My main point of contention is that Daniel's proposal doesn't actually seem to change the policy - or, I mean, it doesn't change the set of questions that it designates as off topic. It just changes the wording, but the way it partitions questions into on-topic vs off-topic is basically the same as the current policy. I agree that it's probably better wording than what we have now, but I think the wording doesn't reflect the partitioning that people actually want.

@Felix.C you're welcome. Im a physicist, so I like $\int\mathrm dx\ x$ way better ;-P

@DavidZ DS will probably be happy if some language on the site changes along the lines of his proposal & think theres strong support its at least worth "experimenting" with different language along the lines he proposes. as in "try something to see if it works, no change will certainly have no effect"... & also DS might start to realize also that mere wording changes may not have much effect...

@DavidZ I might be wrong, but I was under the impression that Daniel's concerns were precisely about the wording!

@AccidentalFourierTransform Yes, that is my impression too

Basically, I think we need a much bigger change than what Daniel is proposing. I would rank his proposal slightly better than doing nothing, but I would not like to see it become the new, long-term permanent policy. I think if that is the case, we will still have most of the same problems we do now.

@DavidZ lets think about it maybe this way. (1) there is a policy wording challenge and (2) there is a site culture change challenge. suggest just starting with (1), which it seems DS has largely tackled. some of (2) is not entirely achievable, its about getting people (both low/ high rep users) to conform to stated policy. but at least one could get a new policy "officially recognized" to go far towards both (1) (2)

Sure, but are you saying that Daniel's proposal should be that new policy you want to get officially recognized to go towards (1) and (2)?

@DavidZ there has been no change in "official" HW policy really probably for very long time. suggest just getting mods to agree on a policy along DSs lines & also changing close reason, that would be most of what is achievable right now. if one of the mods posted something, "here is what we are now doing based on prior meta posts & discussion with DS" think that would be fine, after first running the basic idea(s)/ wording by DS to see if its in the spirit of his proposal.

That doesn't answer my question though....

@DavidZ (sigh) what do you or other mods/ high rep users object to in his meta post? that is my question. let me go look at comments...

@vzn For myself, what I object to is stated in the preceding chat messages.

Maybe I should add those as comments. I guess my previous comments on the matter may not have been clear.

@DavidZ it seems abstract, youre mainly talking about wanting consensus. consensus on what? what are the sticking points for mods/ high rep users against DS proposal?

@vzn I can't speak for the mods as a whole on this one, only for myself. The sticking point is what I mentioned here and here

@DavidZ lies! we all know that the rest of moderators are actually your sockpuppets!

@DavidZ ok what "partitioning do people want"?

@vzn As far as I can tell, low-effort questions should be off topic, and tedious calculation requests should be off topic, or something like that. But getting more detailed than that, I'm not sure. The point of this meta post and its followup(s) was precisely to determine what partitioning people do want.

> Prepare and edit a draft of the new policy that implements the consensus on the examples as well as possible. This may take a while.

↑ lol

having some of this feeling re "analysis paralysis"/ "bike shedding" en.wikipedia.org/wiki/Law_of_triviality :|

Surely it's a judgement call as to whether a short answer is posted as a comment or an asnwer; obviously long answers can't be comments.

@DavidZ: I don't see why they should be deleted as a matter of course.