mob justice

[This thread was resurrected from a couple month’s after Beeminder’s launch, in 2011!]

Our most lucrative user, Michael Merchant, is asking for some special treatment (see below) and gave permission for me to drag him in front of you all for arbitration here.

Here’s the goal in question: http://beeminder.com/mimercha/review (note the $30 pledged)

What do you think?

---------- Forwarded message ----------
From: Michael Merchant
Date: Wed, Dec 28, 2011 at 22:17
Subject: Break for NYE in Vegas
To: support@beeminder.com

Hey Guys,

So, I’m going to Vegas for the NYE weekend!!! :smiley: Yes, I’m really
excited XD Unfortunately, I have this “reviewing/meditating” goal that
might not quite make sense when I’m piss drunk :wink: I’m wondering if you
can flatten my road from tomorrow until Monday? :slight_smile:

What do you say? :slight_smile: [I probably should have done this a week ago when I had planned to go on the trip, but ya… didn’t think ahead]

  • Mike

Eh, I say flatten it. A week of not sticking to a goal isn’t too detrimental; sometimes one needs a break. Forgetting to flatten the road isn’t too bad of an offense. Things come up and we can’t always be accountable for our daily goals.

Oh all the goals I’ve failed because plans got in the way and I didn’t know a week before…

$30 is small compared to the expense of getting piss drunk in Vegas. I think the precedent-setting harm, to Marchant and everyone else, probably exceeds the cost (to Marchant) of resetting his goal.

I say don’t flatten it. He knew what he was getting into. Perhaps he just needs a higher threshold.

Maybe you’ll scare him away as a customer if you don’t flatten it - but even if that happens, it will be useful data for you.

I don’t understand the goal well enough to know why it cannot be achieved while drunk. Surely this is an opportunity for creative problem-solving, or inspired halfassery?

I guess I picked a biased forum to float this question! The first response probably exemplifies what a random jury of one’s peers would say. Alas, here we are!

Isaac’s “inspired halfassery” is cracking me up. But that sums it up nicely. Figure out a way to make it happen. It might end up being a little lame, but not as lame as if you weren’t bound by this yellow brick road at all!

Here’s what Bethany and I have been groping our way towards:

Be as lenient as possible but it has to be based on a generally applicable principle, not special treatment. For example, anything that could remotely be considered Beeminder’s fault should never cause you to derail. This includes missing features that the ideal Beeminder would have, like specific warnings first thing in the morning on emergency days, even if you normally have reminders come at night, or don’t use bot reminders at all.

In fact, we made an accommodation for Michael Merchant last month on that principle, when he went off a road because he didn’t realize it was an emergency day, when the bot told him so late that night:

What do you think of this principle: modify things retroactively to
make it as if Beeminder worked the way it ideally would. So we imagine
that if Beeminder had sent an extra “OMG Emergency Day!” reminder in
the morning, you would have dialed the road down, and then worked your
butt off to get back on the right side of the road. By that principle
we added a flat spot that puts you in the green today, but leaves you
on the hook for the 30/week rate until the 21st. Does this seem fair?

(In retrospect, even that was kind of pushing it! :))

So I think in the absence of a general principle to apply in this new year’s eve in vegas case, we have our verdict: No special flat spot inserted within the akrasia horizon!

And, as others are suggesting, lest it seem harsh, these commitment contracts are powerfully motivating and thus hugely valuable. The cost (either the cost of doing work while drunk in vegas, or just paying the $30, whichever’s less) is small in comparison.

In fact, we’d like to make that a principle in itself, but we don’t know how to operationalize it:
Whenever an $X pledge is paid, the person shall have derived at least $X worth of pants-kicking value out of Beeminder up until the point of derailment. We do think that that is turning out to almost always be true.

Having said all that, here’s an attempt at the other side of this argument:

The true point of a commitment contract is to eliminate akrasia. So anything your long-term self would’ve agreed to, there’s no need to be hard-nosed about. It’s sort of in the spirit of the SOS clause: emergencies or, more generally, truly unexpected circumstances, should be exempted. So if this had been a surprise last-minute trip, it would’ve been in that spirit. The fact that it wasn’t a surprise doesn’t change the equation much (other than making us feel like Merchant got what he deserved for forgetting to adjust his road). He obviously meant to and his long-term self would’ve approved of the exemption, so what’s the harm in pretending he did it?

Man, I’m almost convincing myself! But it’s a dangerously slippery slope – the value of bright lines is huge and “my long-term self totally would’ve been cool with this” is a very dim, fuzzy line!

If anyone has changed their mind or hasn’t chimed in yet, speak now or forever hold your peace! If the current consensus stands then Mike has to start busting his ass to get as much safety buffer as possible before he leaves. :slight_smile:

I’m leaning toward letting him have the flat week.
After all, all he did wrong was forgetting to flatten the road a week ago,
and that has nothing to do with his goal of meditating.
Shouldn’t one be punished only it he fails to keep up with his goal? Or are
we suddenly punishing people for forgetting things?

I mean, if the goal was “be more mindful” or “forget less things”, the
punishment would have been deserved but as it stands the goal is “meditate”
and remembering or forgetting to flatten the road a week before the trip
has nothing to do with meditation.

I’d also like to open a small parenthesis here: please reflect on the fact
that maybe this matter would have not been raised if Beeminder did follow
another business model. In my opinion, a lot of the responses here are
subconsciously “let someone else pay for a service i use”.
And even if anyone in this thread has been completely honest, you could
never tell the difference as there is no way to test for it.

Hi Everyone!

Thanks for some thoughtful comments on my situation.

I just wanted to clarify a little bit more about this graph and explain the
challenge of this particular graph in creating a larger buffer/threshold
before the trip. So unlike the other graphs that I’ll be working hard today
to create my weekend buffer for:

https://www.beeminder.com/mimercha/goals/data

  • Currently this tracks how many contacts I upload from networking events.
    This always seems to take forever! I wish there was better card reader
    technology, haha

https://www.beeminder.com/mimercha/goals/thl

  • I have an enormous To Do List on my computer that keeps growing. This
    tracks to do’s that I take off the top of the stack (ignoring those that
    are organized by project and I take on because of project deadlines)

https://www.beeminder.com/mimercha/goals/sideprojects

  • This is hours spent on coding projects. This one will be tough to keep up
    with :confused:

This graph is meant to remind me to be self-reflective each day. So I’ve
been awarding myself points based on 4 activities that keep my planning and
self reflection happening:

  1. Planning out and visualizing my day in 30 minute blocks before the
    day starts
  2. Tracking my day on my calendar for any deviations from the plan
    (doesn’t include moving things around. This is for if I end up chatting too
    long with a friend or spend too much time sending long winded emails like
    this seems in danger of becoming!)
  3. Reviewing the day with a journal entry
  4. In the last couple of days, I’ve picked up Emotional Intelligence 2.0
    and have been trying its self-reflection exercises

So, the challenge I’m having here is that I can’t really just “bust ass”
today in order to flatten the road. This may be my fault because I’ve given
myself only .5 points for each of the above activities (assuming they
should take less than 30 mins) and have my road sloped at 1 hr/day.

Maybe this isn’t the best way to use Beeminder which is where I’m having a
problem; however, Beeminder has been great way (up and until now) for
helping me perform these daily activities that I get a lot of value out of.
Yet, when on this holiday trip, I don’t want to be making a journal entry,
and I’ll be traveling with a group of friends so I don’t want to be
planning my day, and I’m not keen to pull out my smart phone periodically
and track what we’re doing throughout the day. I think all this
self-reflection is not meant for a Vegas trip. Unless of course, one is
writing a personal version of Fear and Loathing.

With that said, I of course, do like Beeminder, it really helped me
accomplish my thesis (where I spent a lot of my money), and don’t want to
jeopardize the potential value it provides to so many of us by setting a
bad precedent. So, I do appreciate some of the earlier comments made about
creating appropriate thresholds. At the same time, I think it’s important
to take into account the type of graph we’re talking about. And maybe
there’s a chance to learn something interesting here about beeminding
graphs for activities with maximum daily thresholds. Similarly, there might
be something to note about holiday seasons and preplanning flat roads in
the graphs.

Also, I agree with Valerio’s comment that it seems that much of the
conversation might be under a subcontext about who is paying for what. The
idea here being that my $30 is somehow $30 that someone else doesn’t spend
on subscription fees. There may be a much larger discussion to be had
around the relationship between a mindful user and Beeminder. It seems sort
of dicey to think that the guys helping me to accomplish my goals are
feeding themselves by seeing me or others like me fail goals that we take
large bets on.

Woops, sorry for what now looks like a rant. Guess that became more of an
argument than a little notice. Thanks to any of you who read the entire
thing! :stuck_out_tongue: Thanks again to Bethany and Danny for a great service. I’m
interested to see what comes from the convo and hope I win the popularity
contest and get my trip to vegas guilt free, haha.

  • Michael

I have always been worried about building a business on earning money only when your customers fail their goals. I know that true akratics should feel you’re doing them a favor when you take their money under these circumstances, but it’s going to cause nightmares with customers wanting to negotiate, who may then feel like you are interpreting the rules to your own advantage. You’re going to have to develop a reputation for being extremely fair, as insurance companies have to do to get customers to trust that they actually will pay out, and not withhold based on some obscure technicality.

The problem of this particular case seems to be “What is the appropriate horizon to define your long-term self?” Earlier this year, you somewhat arbitrarily decided this was a week. Perhaps for activities that you have to do every day (i.e., you can’t “bank” them as Dan and Bethany usually do), and you want to give yourself a break, a day or two ahead is sufficient to modify the line. Certainly Michael seems to be talking like his long-term self right now. So that’s the main change I would consider at this point.

David

Merchant’s request sounds genuine, and something that his long-term self
would likely approve of. I would probably give him the flat spot exception
despite the drawbacks. I see two arguments against this:

  1. He should just increase his incentive (and pay for his mistake). This
    might or might not be adequate, it’s a matter of opinion I guess. If he
    really needed a higher incentive, he’ll get it soon enough anyway the next
    time he goes off-road. Making a single exception shouldn’t make a big
    difference in the long run. Of course, this would really need to be an
    exception and not become the rule.
  2. Having no exceptions is simpler / bright lines are better than fuzzy
    lines/etc. True. Personally, however, I still favor some controlled
    flexibilty, and really like the idea of (a more systematic) mob justice.
    For example, what if these types of exception requests required approval of
    x% out of a minimum of y established users/voters? The requestor would
    present his/her case in a public forum, people would respond, and vote –
    just like this discussion thread, except that we would have clear rules and
    pre-specified percentages/number of voters to grant exceptions. This saves
    Beeminder LLC the pain of making potentially arbitrary decisions that get
    users upset (and reduces the perceived conflict of interest) while allowing
    exceptions for cases that deserve it (yes, we would be trusting people to
    be fair). At the same time, I think that this “social” component would be a
    good addition to Beeminder (there are other potentially interesting social
    features – we can discuss that in a separate thread). Convincing people
    would still be annoying, but you could always pay if you don’t want to make
    the effort. I really don’t know how often these exception requests occur…
    this is obviously something to consider and think about. But in the
    meantime, I do think that Merchant deserves a break.
    David’s idea of a more flexible horizon sounds interesting and worth
    considering – but I wonder if that would solve all potential exception
    requests.

I like the idea of figuring out how to implement mob justice as Julian suggests. You’ll have to figure out how to get participation from other users, as your users are likely to be busy people. But I think that you should definitely cultivate a sense of community spirit. This will certainly help you get around the problem of people mistrusting you because you stand to get money when they screw up.

One thing you’ll have to think about is whether users will collude against you to be “nice” to each other so that disputes are always resolved in the user’s favor. If you have the right sort of customers, who actually care about helping each other learn to reach their goals, then you’ll probably be OK, but this is something you want to think carefully about.

Good points. There are a number of ways to address concerns regarding
collusion. For example, you can easily limit the number of exceptions per
user in a given time frame (e.g. max one exception per year), or increase
the requirements for approval (i.e. votes) if you kept requesting
exceptions. In any case, if someone really wanted to “cheat”, they could
just lie and wouldn’t really need collusion to avoid paying (either way,
they would mostly be cheating themselves and shouldn’t be using Beeminder
in the first place). Plus, the owners of the site can always be watching
for potential misuse. There are also ways to encourage user voting and
“fairness” (it all depends on how complex you want to make this).

Wow, thanks, Julian and David, for these really smart ideas about
systematizing mob justice.

One other note: I’m pretty confident that the fears of perverse
incentives are unfounded.
No one is thinking that voting for Michael Merchant to pay up may help
save them money on subscriptions in the future, not even
subconsciously.
This is based on extensive interaction with customers who, well, trust
me, they don’t think that way!
And there’s no shred of hypocrisy from people: the ones who think Mike
should pay are even harder on themselves.

It’s of course meaningless for me to defend my own subconscious
motives, though, for the record, I don’t think that’s an issue either.
Maybe when we’re a big evil corporation, but we’ll have made so many
pivots by then that this particular question is likely to be moot. In
the meantime our overwhelming self-interest is well-aligned with happy
customers. This $30 we may or may not make off of Michael Merchant is
trivial compared to the bigger question of what the right policy is
for this sort of thing. In evaluating that we can think in terms of
what will make us the most money in the long run or what will best
motivate people to stay on their yellow brick roads. I contend that at
this stage those are essentially equivalent. Beeminder’s success
hinges entirely on how valuable it is to people.

So that’s the key question: does leniency in this case set a precedent
that makes Beeminder more valuable or less?

But more immediately, Mike needs his verdict, so let’s hurry and get
the official vote:


[contents of shared editing doc pasted below]

The following was posed on the Akratics Anonymous email list. (See http://blog.beeminder.com/akratics )

Should Michael Merchant have to pay up if he goes off his http://bmndr.com/mimercha/review road, that he forgot to flatten in time for his new year’s eve weekend in Vegas?

LET HIM OFF. It’s in the spirit of an Unexpected Thing that came up, he clearly meant to flatten the road, and his long-term self would agree that it’s a valid exemption. In short, it doesn’t harm the anti-akrasia power of Beeminder to grant the exemption. And the precedent makes it less scary to enter into commitment contracts in the first place. (As long as we have ways to automate or crowdsource the exemption process, and we will.)

Votes: ijn10, valerio, julian, reiley, eyesack, aj

MAKE HIM PAY. The cost is trivial compared to the weekend in Vegas and the “long-term self would agree” criterion is too fuzzy. There’s huge value in bright lines. Letting him off is a dangerous precedent because it dampens the motivational power of Beeminder. If you know in the back of your mind there will be wiggle room if push comes to shove, then you take the yellow brick road less seriously.

Votes: johnicholas, dreeves

Verdict: We inserted a 3-day flat spot on http://bmndr.com/mimercha/review despite it being within the akrasia horizon.

This is based on extensive interaction with customers who, well, trust
me, they don’t think that way!
And there’s no shred of hypocrisy from people: the ones who think Mike
should pay are even harder on themselves.

But you need to be thinking ahead. You may have found a special group of people, and that may not scale well. The important thing is to be creating social institutions now that will sustain this culture going forward. If you get up to tens of thousands of users, I guarantee you will find some who contemplate acting in narrow self interest. The question is how to establish the norms that will sustain you, and how to figure out how to discount or neutralize the votes of the jerks who will start to join your service.

This $30 we may or may not make off of Michael Merchant is
trivial compared to the bigger question of what the right policy is
for this sort of thing. In evaluating that we can think in terms of
what will make us the most money in the long run or what will best
motivate people to stay on their yellow brick roads. I contend that at
this stage those are essentially equivalent. Beeminder’s success
hinges entirely on how valuable it is to people.

Sure. But keep looking towards a future where you are going to have people act like jerks, try to be manipulative, and/or try to instill fear in other users that your site can’t be trusted.

So that’s the key question: does leniency in this case set a precedent
that makes Beeminder more valuable or less?

I now vote for leniency in this case because we are still talking about how to improve the service, and I agree that the exception does not violate the spirit of Beeminder. But in future I hope you will have better rules in place (such as my suggestion that we let this kind of decision be made two days out by a user) so that you won’t have to consider exceptions. And voting rules that let your users help decide future tricky cases.

How is this for a solution? An emergency road-dial change option You can change your road-dial immediately, but doing so ups your contract to the next level on the exponential scale. To avoid having people do this constantly and (in theory) owe tons of money, limit this to 3 times.

I do agree with David that there might be some goals where you can’t or don’t want to build up a safety buffer. I thought about setting up a beeminder goal of posting to our family blog once a week. Then I realized that I already do that (on average), but I often go in spurts. But this is not really what readers of a blog like. People prefer new articles on a consistent basis. But I can’t figure out how to do that with beeminder (I suppose I could only mark one post per week on beeminder, and ignore anything over that - okay - I just figured out how to do it)

This has been an insanely valuable discussion; thanks everyone!

(It’s now looking good for Mike at [shared document URL for voting; see below for snapshot] !)

One other idea that we could implement more quickly than true mob
justice with voting is this:

In short: When you create the contract you can choose to specify the
email addresses of 3 friends. If you want an exemption, they all (2 of
them?) have to email in corroborating it. Prima facie that’s kind of
toothless but remember that it’s your long-term self that creates the
contract and very much wants to make things difficult for their
subsequent weaseling, short-term self.

So I suspect many would decline to specify friends and just accept the
risk of the inflexible contract (or relish it). Or you’d specify
friends that would be pretty hard on you and you’d only ever ask them
to corroborate an exemption if you had damn good reason.

But it seems like it would’ve solved Michael Merchant’s dilemma elegantly.

(Ooh, and you could specify the 3 friends at any time, just that
there’s the usual one-week delay before they can rescue you.)

Robert: your emergency change option creates an interesting trade-off, but
I am not overly enthusiastic because it sounds pretty arbitrary and could
create false options, e.g., we give you the “option” to increase your
commitment beyond your motivating point, even to a level that you can’t
afford.

I think that Dan’s solution of involving 3 friends is reasonable from a
mechanics/incentives perspective, and worth testing if it is so much easier
to implement. Longer term (but sooner rather than later), I still think
that a larger mob justice system could be more fruitful, and help create a
true community of people helping each other achieve their goals as opposed
to offering “just” a great tool. This community could involve not only
“justice”, but also support groups, expert advise, moderated forums, etc –
in other words, Beeminder can give users a lot more value than that derived
from a great tool. Besides, Beeminder as a tool would be relatively easy
for someone else to copy (given time and $); a real community of involved
users, however, would be more difficult.

On Dec 29, 2011, at 7:15 PM, Julian Schvartzman wrote:

I think that Dan’s solution of involving 3 friends is reasonable from a mechanics/incentives perspective, and worth testing if it is so much easier to implement. Longer term (but sooner rather than later), I still think that a larger mob justice system could be more fruitful, and help create a true community of people helping each other achieve their goals as opposed to offering “just” a great tool. This community could involve not only “justice”, but also support groups, expert advise, moderated forums, etc – in other words, Beeminder can give users a lot more value than that derived from a great tool. Besides, Beeminder as a tool would be relatively easy for someone else to copy (given time and $); a real community of involved users, however, would be more difficult.

Hear hear. As Dan and Bethany know, I’ve been skeptical of Beeminder’s ability to become a sustainable business. This is the first idea I’ve heard that has given me reason to feel some optimism.

You can always find time to meditate, and if that is your goal you should do it even if you are on vacation.

I just went through something similar. I feel like there should be a “on vacation” or “away from internet” option. That being said I tried to keep up with my beeminders even though I couldn’t log them since I had no internet. I ended up gaming the system and I will update my data when I am online again.

I’m resurrecting this thread from nearly two years ago because, one,
it’s a fascinating blast from the past, and, two, we need some more
mob justice!
Check out the comments last night and this morning on
blog.beeminder.com/blogdog and chime in!

Short version is that we might owe someone $1000! I’ll repeat my
comments here, in reply to the person who claimed the booty at
midnight:

Ooh, yikes! We did the UVI and tweeted it [twitter.com/beemuvi] and
then I forgot to actually update the bmndr.com/meta/uvi graph till
now. So we might actually have to cough this up even though it’s a
technicality (we’re purposefully holding ourselves to higher standards
– for our users we always make sure you never pay up if it’s a
technicality). So we’ll go with the precise intent of the fine print
here. The only thing that may save us is how to interpret “you may
claim the bounty at midnight”.

So Somervta is the first to claim it and gets it if it is in fact
forfeited, but we did the work by midnight (including publicly
tweeting) and entered the +1 on Beeminder in time to keep the graph
from officially derailing. So, arguably, the money was not in fact
forfeited. (Our minds can be changed here if that sounds weaselly!)

Clearly we need to hurry and add a Twitter integration for automatic
Twitter minding to make this moot. (We actually can do that already
via IFTTT.com but our UVI goal is still odometer style so it’s easier
said than done.)

[and in another thread:]

Tentatively I’m thinking the right interpretation of the official
rules is that we have to do the work by midnight, you can claim the
money exactly at midnight, but the money isn’t paid unless it’s an
official derailment, per the standard rules, which means we have a
grace period till ~3am to get the data entered (if it’s a graph with
manual data entry).