I think I'm sold on this now. Thanks Mike and Hugo!
I'm not certain about your intuition about essentially a single submit for all the tabs, but maybe it should at least warn you before dropping your changes if you click away without submitting. Though that could be annoying if you're not hitting submit specifically because you decided not to submit those changes. I guess it's a question of how likely is it that you'd make changes and forget to hit submit. Back in the old days you always needed to hit submit so users were used to that -- maybe less so now...
We’re still thinking hard about this. Thanks so much for the insight here — it definitely cast this issue in a new light for us.
Here’s the link to the feedback page about deleting goals: http://beeminder.uservoice.com/forums/3011-general/suggestions/2353108-let-us-delete-goals
We've done a few things in the meantime to try to strike the right balance between these schools of thought:
1. When you create a goal you can mark it as ephemeral, pre-specifying that it's just for testing and you don't want it to stick around.
2. You can "archive" goals, shoving them below the fold on your gallery page.
3. If you painstakingly delete all the data points then the option to delete the graph will appear.
Are you thinking just in terms of grouping related goals together? We did recently add a very crude approximation of that: you can now partition goals into those you currently care about and those you don’t currently care about.
1) Done. Eg, http://bmndr.com/meta/uvi
2) The above would've made sense as a separate feedback item (if it weren't already done! great minds think alike :)) but discussions like this in the comments are fine too.
Hmm, the prefix trick would work nicely if we sorted alphabetically... :)
There's another feedback item about how best to sort. Maybe we should let you choose, though that involves adding clutter to the interface (there aren't just columns you can click on in the gallery to re-sort).
Ah, as to your second comment, not yet. We have a crude API in private beta but don't have goal creation available as part of it yet. We'll hurry!
Ah, this is smart. (Disregard my comment about grouping goals -- this is a much more powerful idea.)
Just streamlining the process of actually creating goals is also a good idea. We'll keep thinking about that. Thanks so much these suggestions!
Thanks for pointing this out! Sometimes it's nice when it's actually extrapolating reasonably, but you're totally right.
Do you think it's less egregious when it just extrapolates up to the akrasia horizon? (See your graph now.)
There's an obvious solution to this... :)
We've viewed data export as a mission-critical feature from the start. The link is right below your avatar on every graph page.
We'd love to see what you come up with, data-visualization-wise! If you make something awesome it probably won't be hard to convince us to build it in to Beeminder!
The latter is definitely less of a pain to implement. I'm eager to hear if there are others who want this. I feel like, realistically, despite the publicness of URLs like beeminder.com/alice/weight no one is going to see that if you don't point them to it. Certainly it depends how sensitive it is, though, whether you want to take that risk. But then if it's that sensitive maybe secret URLs (like beeminder.com/38f8afhs9yvgw337a) wouldn't be sufficient?
As Alice pointed out in the previous status update, we’ve got apps for both iOS and Android, and they’ve got super easy data entry methods. But for people using the SMS bot or who just prefer the email bot to the apps, I like some of these ideas. Keep them coming!
I like this idea of having free reign to experiment with a goal without having to worry about it getting frozen, and then hitting Commit whenever I'm ready.
Of course to some extent you can view the initial free yellow brick road that way. And you can even approximate this by creating extra dummy goals, though we want to discourage that.
We'll keep thinking about this, and please add more thoughts if you have any.
I did mean the latter, and it's more for people who really don't like the whole commitment contract concept. (It's like the business model for a personal trainer with a money back guarantee -- you pay the trainer if they help you succeed.)
However currently our hands are very full with users (like you!) who are delighted and excited about commitment contracts (or "pledging to stay on your yellow brick road" as we're thinking of calling it) so this is on the back burner for a while.
That’s a really good idea!
(The way we’re generating the graph images currently makes that easier said than done, so this might be a while, but we agree it’s worth doing.)