This is from around 2012. Historical interest only!
Status quo: It’s purposefully difficult to delete a goal unless you created the goal as ephemeral (as a “temporary test” goal). See the FAQ item on deleting goals.
Current proposal: Add a delete button right on the goal page that respects the akrasia horizon (similar UI to pledge decay, where the button turns into a countdown to deletion in a week). But for newly created goals in the trial period, the delete button is instantaneous.
We call a goal a Freebee if it has no pledge on it.
All goals start out as freebees with no limits on how many you can create. You don’t have to take any action unless you explicitly created a goal as ephemeral but then change your mind. If you derail on a freebee you can resurrect it by pledging. If you derail with 7 or fewer datapoints (or within the first 7 days) then you can re-rail without pledging.
All goals start in a don’t-fall-in-love-with-me-yet trial state (watermark says “TRIAL” instead of “$0”). Furthermore, you get 3 true freebees — goals that can stay pledgeless indefinitely as long as you stay on the road.
In case you’re not quite ready to commit, you’ll have 7 days with this goal to decide if you’re really made for each other.
After a week of data (7 datapoints or 7 days, whichever’s longer) the goal either automatically becomes a freebee (watermark: “$0”) if you have freebees left or it freezes till you add a pledge.
Once you use a freebee you don’t get it back, even if you delete the goal. [1] You do get the freebee back if you add a pledge to a freebee.
Similar to Magnetic Fields but there’s no such thing as freebees. All goals start in a “trial” state but at the end of the trial period, if you like it then you better put a pledge on it, because otherwise it stays frozen forever, even if you’re still on the road.
The “frozen goal” screen would emphasize that you are not immediately paying the $5 but rather committing to pay it IFF you go off you road.
What happens to goals that you don’t add a pledge to? Can you elect to delete them or do we keep “cluttering your goal gallery” as a disincentive for creating goals willy nilly? The new model of “All goals can be deleted subject to the akrasia horizon” probably works.
You get 7 freebees and there’s none of the complexity of trial periods and choices to make when the trial period ends. Creating a goal just always uses up a freebee unless you explicitly add a pledge.
This one might be naively simplistic. It might also be exactly the stepping stone that we’re looking for. And the simplest thing that could possibly work.
Given:
So, the alternative proposal: change nothing about how pledging works today, except to rework the labels & associated wording. $0 becomes a “$5 freebee”, no credit card required. The dashboard will show as $5, etc. Just as now, I could make a real $5 pledge instead of taking the freebee. No limits on freebees, etc, just a firm psychological nudge in the thinking-it’s-money direction.
People will quickly forget which goals have real chargeable pledges and which have freebee pledges. They will behave toward their graphs as though they have real money at stake. In turn, if the historical inclination holds, they will more willingly ante-up to the next pledge level. (That last point needs monitoring, but the others I’m certain of!) They should also realise the value of staking money because they will act as though they have already done so.
Dreeves adds: One non-cosmetic change this might imply is that derailing on a “$5 freebee” means you have to pledge $10 to unfreeze even though you didn’t actually ever pledge or pay that $5. We’ve thought about dropping the $5 level anyway — no one would really mind that.
Philip adds: I have recently noticed that ticking the ‘temporary’ box is detrimental to my Beeminding. I’ve let a few expire that I shouldn’t have. I would have been better off creating as many new goals and later decisively abandoning them. I had argued for ephemeral goals because I often can’t imagine the effect of a goal until I’ve lived with it for a bit. That’s a false argument for ephemerality, because I can decisively cease tracking a goal at any time, subject to the akrasia horizon. I am much better off when I control my goals through acts of commission, rather than passively.
I don’t object to dropping the $5 paid pledge level, it was always at odds with the pledge schedule. We could monitor the number of data points entered before derailing at various pledgel levels. This would help tell us whether there is additional commitment effect from pledging $10 over $5.
Note that for either Magnetic Fields or Simply Seven we’ll need to rewrite beeminder.com/money to not be all “first try’s free! no obligation and no credit card!”. The steady state (or the platonic ideal of the bee) is that starting goals means pledging money. Just that we need trial periods and freebees to show people how great we are before they’re willing to do that.
Glenn adds:
The no-take-backs rule complicates Magnetic Fields a bit, because without it you could just autopromote every expiring test goal to a freebie as long as more freebies are available.
Perhaps there’s some better formulation of no-take-backs that would let you do that.
Removing that choice makes Magnetic Fields look a lot less complicated.
(Datapoint from Glenn: Of my current 10 goals, under Magnetic Fields I would probably pledge 6 of them and keep another 3 around as freebies. Under Simply Seven I would probably pledge about 7 and drop the rest. I am in the habit of gratuitously creating lots of goals just to see what it feels like to track those things. Thirty days’ grace would likely be enough to eliminate my desire for freebies; 7 days is not.)
Another radical idea that could potentially lower the stakes on these tricky freebee-related decisions: When you delete a goal we could just be like, “Should that count as one of your freebees? Say no if anything went wrong in the setup or anything. Be honest!” Maybe that’s unnecessary with the Magnetic Fields proposal where you have enough of a trial period, but in general this kind of honor system thing will probably work surprisingly well. For example, you could just click something to get more freebees (except, as Andy points out, akratic folks would just keep clicking that indefinitely; but perhaps as a one-time thing).
[1] The no-take-backs rule prevents people from endlessly deleting and restarting pledgeless graphs. Currently people do it without deleting, like “weight2”, “weight3”, “lose-weight”, “for-real-this-time-fatso”, etc. When we make deleting easy then people can keep recreating “weight”.