First things first, I was mistaken in today’s weekly beemail about scheduling breaks:
It’s now a bit easier to schedule breaks on all your goals at once, thanks to a “Schedule all” button that just expands everything all at once. You still have to click Save on each goal individually. Half the total clicks though! (This is a compromise — we really think you should think about each goal you’re setting a break on rather than schedule indiscriminately.)
That was doubly incorrect. I forgot we’ve long had a Save All button at the bottom of the breaks page. You haven’t, since forever ago, needed to both expand and click Save for each goal.
So my blithe aside about not making it too easy to schedule breaks on literally all your goals willy-nilly, is, as of this new “Schedule all” button, no longer true. It’s suddenly super easy.
Is that good or bad? I think it’s going to be popular but I’m not sure yet if I want to let it stand. Here are the reasons I don’t like it:
Speaking for myself as a user, I’m horrified at the idea of indiscriminately setting breaks on everything. That would break so much of my stuff! Like goals I’m counting on to reach specific values on specific dates.
There are infrastructure reasons that gratuitous break-setting is not great, as segments of the bright red line proliferate too much. (This is our own fault and we’ll get it sorted; it mostly shouldn’t count as a reason.)
Too much willy-nilly break setting on all your goals could let Beeminder go out of sight, out of mind? (I’m not sure this reason is particularly valid from a user perspective. That would be the wrong kind of paternalism to second-guess how much Beeminder you’re choosing to have in your life.)
It makes it too easy to make things too easy, goal-wise. (Also questionable.)
We view it as preferrable, when possible, to build up enough safety buffer to not need to schedule a break.
You might be simply wrong to not consider whether and how much of a break you want on each goal. (For example, if you’re going on vacation you’ll probably get more steps than usual; no need for a flat spot on that one.)
Relatedly, if you’re setting breaks due to a trip next week but you have a goal with a month of safety buffer, maybe pushing that goal further out of sight, out of mind isn’t what you really want.
This brings us to the title of this blog post. We’re pretty on board with paternalism, aka Opinionated Software, as a design principle for Beeminder. We’ve been talking about it since the very early days, in the context of setting good defaults and making a user’s path of least resistance a path of ever-increasing awesomeness:
Given the uncanny power of defaults, we need to embrace the inevitable paternalism. What’s actually in the user’s best interest? What would they choose if we could reach through the computer screen, grab them by the shirt collar, and hold their attention until they made a considered decision?
Examples of this abound in Beeminderland:
- Many gung-ho users think they want to jump straight to $1,000 at stake and we learned the hard way not to let you do that.
- Sometimes people want to cram a whole sentence into their goalname, and we don’t let them. Long goalnames are gross and make Beeminder subtly worse in ways newbees don’t anticipate (e.g., using the SMS bot).
- No deadline snoozing for you. “Let us save you from yourselves — backing yourselves into corners where all your deadlines end up at 6am, night owl mode.”
- No more sorting by deadline, just by “urgency” where that means a whole list of things we impose on you.
- You have to type a phrase to prove you’re eyes-open opting in to No-Excuses Mode.
- We turned Anti-Settings into a whole guiding principle.
As we said when pitching the new, less forgiving rules for adjusting goal deadlines, the general rule of thumb is that things that make staying on the good side of your bright red line easier make reaching your underlying goal harder.
Maybe that applies to how hard the UI of the breaks page should nudge you towards considering your goals one at a time when setting breaks?
Or maybe you can help us find the best of both worlds? That might mean (a) you can reasonably quickly set a break on literally everything — or everything up to some amount of safety buffer? — when you’re sure that makes sense for you. And also (b(ee)) there’s just enough friction that you’ll normally choose to consider each goal separately when setting breaks, as the Omnibenevolent Infinibee intended.
