dreeves

Encouraging update on the status page:

"The issue is related to heating/cooling complications in the data center due to a power outage . The power outage has been fixed and we are working quickly to bring our services back online."

More assurance that it's not just us: https://downdetector.com/status/linode/

"Mal de muchos, consuela de tontos."

Some signs of life from some of our servers now. Still no word from Linode.

Is this hard downtime for anyone else? We're not sure what they mean by "emerging service issue" and they haven't replied to our support ticket yet.

My counterargument: https://agifriday.substack.com/p/jevons

Short version: Jevons's paradox means that the more coding you automate away for developers (like with compilers in the past), the more in-demand those developers are. (Until AGI when all bets are off, of course.)

That's our (Beeminder's) esteemed, if now somewhat moribund, competitor, StickK. See https://blog.beeminder.com/anticharity for our argument against anti-charities like that.

Hi! Beeminder cofounder here. We do have a charity option but only in our most expensive premium plan. My own feeling is that a commitment contract with a charity as a beneficiary is less effective because what kind of jerk is motivated to avoid donating to charity? Unless you set the stakes so high that you can't really afford it, I guess?

Congrats on the launch! I think this is powerful and handles the use cases you've listed better than Beeminder can currently. So that's exciting for us -- we love worthy competitors!

Speaking of which, here's a list of all such competitors we know of: https://blog.beeminder.com/competitors/

(Adding Discipline.io to the list now...)

Beeminder cofounder here! Thanks for the plug! So much to say here but maybe I'll start with a pointer to our philosophy on anti-charities: https://blog.beeminder.com/anticharity/ (short version: we hate them).

Ah, this continues to be good feedback. Thanks for continuing to hash it out with me! I see I made it sound like there were a lot of moving parts in my argument for why our incentives aren't so perverse. I don't think that's the case! In particular, I don't think my argument relies on what kind of people we are. I mean, it relies on us not turning totally evil and myopic, but that's true of any company. If we started effectively wrongly charging you, you'd cry foul and quit.

I'm worried I'm not really grokking your underlying argument though. Maybe it just feels gross to have this kind of setup with a third party as opposed to doing it with friends. That's the kind of thing I can't argue with so if it's something like that we can leave it at that. Thanks again for helping me think through how to convey our pitch for the general non-perverseness of it in any case.

Hi! Beeminder cofounder here! I'm pretty excited to see all the positive comments but of course I've homed in on this negative one first. I think Beeminder is incentivized to make you fail at your goals the same way eBay sellers are incentivized to not actually send you your stuff after you pay them.

Anyway, we have a whole elaborate essay on why there's very much the opposite of a conflict of interest: https://blog.beeminder.com/defail/ (about how Beeminder revenue is proportional to induced user awesomeness)

There's a key faulty assumption that may make it seem like our incentives are more perverse than they are. Namely, it's not the case that Beeminder goals are binary things that you either succeed or fail at. They're things you make long-term graphs of, like averaging 10k steps per day or working 40 hours per week. You pay Beeminder because your overall progress is much greater with Beeminder than without it, even though the specific moments you pay are kicks in the pants when you've deviated from your commitment.

I'm definitely interested to hear if any of this is persuasive. We hear the perverse incentives thing a lot so we need to figure out how to convey our apologia much more concisely in our intro material! (And thank you for voicing it!)

Thanks for the plug! See also https://blog.beeminder.com/competitors

Scott Aaronson adds the following in the comment on his blog post in response to a question about this:

the NDA is about OpenAI’s intellectual property, e.g. aspects of their models that give them a competitive advantage, which I don’t much care about and won’t be working on anyway. They want me to share the research I’ll do about complexity theory and AI safety.

In my opinion these automated solutions seldom work in the long run

Beeminder cofounder here. Can I hear more about why you think this? There are definitely people for whom Beeminder doesn't work at all but you sound like you're making a different claim -- that it may work for a while but then stop working. That's the opposite of our experience. Our churn numbers get really good for those who stick around for a year and anecdotally we have lots of people getting PhD theses written thanks to Beeminder, etc.

But if you've had short-term success with things like Beeminder -- https://blog.beeminder.com/competitors -- and then had it fail, that would be valuable to hear more about.

Oh, and I should mention that Beeminder isn't necessarily entirely automated. If you derail and are about to be charged money but don't agree that it was a legit derailment, you talk to a human about that.

Oh! Beeminder's not actually affiliated with TaskRatchet. We're just friends with the creator of it and have promoted it a lot and have an autodata integration with it, etc. See https://blog.beeminder.com/taskratchet -- which is a guest post by the TaskRatchet creator on the Beeminder blog (so, um, I can see where the confusion came from!).

I'm not sure if it ends up too in-the-weeds but we had a fun debate with one of our Beeminder-but-on-the-blockchain competitors recently: https://forum.beeminder.com/t/decentralization-debate-beemin...

Very short version: We don't think smart contracts add much. Your users trust you.

Here's our counterargument, or at least why Beeminder prefers punishment: https://blog.beeminder.com/contrapositive/

Relevant bit:

This sounds good but we’re not into it. I mean, first, we do have plenty of positive reinforcement in the form of pretty graphs and the satisfaction of adding datapoints. You can even spin the pledges as positive — they help you quantify the value of your goals. That can be powerful information for us rationality nerds.

But why not reframe Beeminder to focus on rewards? Well, paying money up front and getting it back unless you derail is a trick — it’s equivalent to getting stung. At least for me personally, the equivalency would always be at the back of my mind and bother me.

And there are more pragmatic problems. I like having scary high pledges on some of my goals. It would feel especially unreasonable to pay up front on those. Even more pragmatically, most goals are open-ended: get 10k steps (or work 40 hours, or practice piano for half an hour or whatever) per day forever. There’s typically no particular point when it makes sense to get your money back. It would be totally inefficient to have money always flowing back and forth and would really muddy the mental accounting in terms of how much you’re paying Beeminder for the motivation it’s giving you.

Not to mention the laws and accounting involved. We’d be kind of a bank and have revenue that wouldn’t count as revenue. I assume this part would be perfectly overcomeable if we were convinced the psychology / behavioral economics were right. But, again, we are not.

Happy to answer such questions! By disputes I believe @hackandtrip means customers disputing charges with their credit card company. Aka chargebacks.

Beeminder gets zero credit card disputes because we give people a chance to contest their derailments and cancel the charge before it goes through. They have to talk to a human workerbee but we make it as easy as possible. If that doesn't happen in time and the charge goes through but the user still doesn't think it was legit for whatever reason, we refund it. No need for it to ever get disputed with the credit card company.

Definitely don't be discouraged by the existence of StickK! They technically already have the referee feature (that being your key distinction from Beeminder, which has a much less powerful version of that feature) but everything else about StickK is... well, I think you've already surpassed them despite their decade+ head start. They seem to have sadly been in zombie mode for years, since the founders left. I guess they're still making money though, which should be encouraging for Kommit!

Btw, here's our argument against the anti-charity feature: https://blog.beeminder.com/anticharity/

(I'm a cofounder of Beeminder, if that wasn't obvious. Also I just added Kommit to https://blog.beeminder.com/competitors/ -- very excited to have you as a competitor!)

Good point! I mean, I don't know if I'd say that makes lifetime plans themselves a dark pattern. It's demonstrating a lot of faith in the company to buy one. Hopefully Beeminder is proving itself worthy of that faith!

We don't feel awkward about it! We're leaning into it hard. See https://blog.beeminder.com/focus and https://blog.beeminder.com/defail

But it's definitely not for everyone! If your reaction to Beeminder is "I would not do anything differently and just waste money" then you are probably right and should not use Beeminder. We've been around for about a decade which we think is evidence that there are people for whom it does work.

For anyone in the category you describe (tried Beeminder, found their expectations of themself to be unrealistic, quit Beeminder) we definitely want to talk to you.

Also calibrating self-expectations is one of things many users tell us is worth paying for.

Good questions; I don't have good answers! Various people have taken a stab at it, suffered 90%-done syndrome, and, yeah. As for myself, I have 70 ttc pings (52.5 hours of work) pretty much exclusively on project management type stuff, like trying out people's betas or refining the rules of the offer or other writing. I suppose I'd count writing this comment if TagTime pinged me right now... (it didn't).

To clarify, by "subsume the janky Perl version" I did not mean touching any Perl code!

Nice work! I have no idea if this could fit into your roadmap (and this is probably the kind of feature request you should absolutely say no to) but my partner and I made a thing called TagTime that does time-tracking via random sampling (pinging you at random times) and I'm offering the domain tagtime.com to whoever subsumes the janky Perl-scripts-plus-Vim version we use and makes a proper TagTime-in-the-cloud. There's a pretty decent community of users to tap into but, again, it would probably entail a substantial pivot.

Thanks for this recommendation! If it has a commitment device aspect to it, I'll add it to our list of competitors at http://blog.beeminder.com/competitors

I think there are a million tricks and my top meta-trick would be to ask that question in the Beeminder forum. Just asking it will get you engaged with the community there and serve as additional motivation to stay on the wagon.

Analytics-wise it seems that if you stick around for 9 months to a year you become markedly more likely to stick around indefinitely.

Another possible source of tips: http://blog.beeminder.com/burnout -- about not getting overzealous about beeminding all-the-things right off the bat.

First question: did you know that with Beeminder you can schedule arbitrary breaks in your commitment/graph? They do have to be scheduled with a week of lead time. (Of course you can also take the hit and get, by default, a week of respite after derailing. Our most hardcore users find that worthwhile sometimes and explains how Beeminder makes money even (especially) from our most successful users. I know that seems ironic or paradoxical for those first encountering Beeminder!)

I can rattle off half a dozen things to stop someone from doing that!

1. It ruins the Quantified Self aspect. You don't want to falsify your data.

2. You've set up an autodata goal and there isn't any (easy) way to cheat. Your Fitbit (or Habitica or Duolingo or Project Euler or whatever) just tells Beeminder how many steps got (or whatever your metric is).

3. Your graph is public, and ideally you've pointed friends/family to it. Are you going to lie to your friends and family?

4. It's just part of your identity as a Beeminder user. If you were the type to falsify your data to weasel out of a commitment, you wouldn't have signed up for Beeminder in the first place.

5. Beeminder does a lot of valuable things (reminders, graphs, community, generally making you awesomer) so arguably has earned the money by the time you first derail.

6. Cheating on Beeminder is a devastating precedent that ruins the power that it had as a tool to motivate you.

7. Think of the children! https://www.youtube.com/watch?v=jB62AyZ6gHk (Those kids are so much bigger now, because your derailments pretty literally put food in their mouths!)

Ok, that's all I've got. Here's the blog version: https://blog.beeminder.com/cheating/

You should check out the Beeminder community: http://forum.beeminder.com -- I think it will quickly convince you that we're a Very Serious Tool For Self-Improvement.

Good observation! Yes, you need a simple, graphable metric that's a good proxy for the underlying goal you care about. We've written about that a bit at http://blog.beeminder.com/whattomind

Beeminder encourages you to define and refine your metric to make sure it doesn't fall prey to Goodhart's Law. Of course it's your own goal and so it's much less susceptible to Goodhart's Law. You actually care about the underlying goal!