# Max safe days set to zero bug

**URL:** <https://forum.beeminder.com/t/max-safe-days-set-to-zero-bug/246>\
**Category:** Bugabee\
**Created:** [November 19, 2014, 9:51pm UTC](https://forum.beeminder.com/t/max-safe-days-set-to-zero-bug/246 "2014-11-19T21:51:18Z")\
**Posts on this page:** 9\
**Page:** 1

<div class="post-metadata">

**Author:** ![kkyang](https://sea1.discourse-cdn.com/flex019/user_avatar/forum.beeminder.com/kkyang/32/108_2.png) [@kkyang](https://forum.beeminder.com/u/kkyang)\
**Post date:** [November 19, 2014, 9:51pm UTC](https://forum.beeminder.com/t/max-safe-days-set-to-zero-bug/246/1 "2014-11-19T21:51:18Z")

</div>

I have a goal of writing 750 words everyday. I’ve used a binary system of 1 or 0 to represent yes or no.

I’d like to have more data so I want to log my words written each day opposed to 1 or 0. So I created a goal with a daily rate of 750. I then set my max safe days to 0.

What I anticipated was for the goal to auto ratchet back to a minimum of 750 words after my derail time regardless of how much I wrote the day before. So if I write 3000 words today, I would still have to write 750 tomorrow.

However setting max safe days to 0 doesn’t seem to affect the derail time.

---

<div class="post-metadata">

**Author:** ![dutchie](https://avatars.discourse-cdn.com/v4/letter/d/91b2a8/32.png) [@dutchie](https://forum.beeminder.com/u/dutchie)\
**Post date:** [November 19, 2014, 10:53pm UTC](https://forum.beeminder.com/t/max-safe-days-set-to-zero-bug/246/2 "2014-11-19T22:53:54Z")

</div>

There has been lots of discussion on this sort of thing in [some](http://forum.beeminder.com/t/feature-request-store-safety-buffer/227) other [topics](http://forum.beeminder.com/t/thoughts-about-retroratchet/229).

---

<div class="post-metadata">

**Author:** ![spookane](https://avatars.discourse-cdn.com/v4/letter/s/96bed5/32.png) [@spookane](https://forum.beeminder.com/u/spookane)\
**Post date:** [November 20, 2014, 8:03am UTC](https://forum.beeminder.com/t/max-safe-days-set-to-zero-bug/246/3 "2014-11-20T08:03:30Z")

</div>

I would appreciate this feature too. There are many goals I want to persue where I just “show up” e.g. do at least 1 minute of Anki reviews everyday. Now if I’ve been lazy a few days and built quite the backlog, I can review for 20 minutes in one day and that gives me 20 days of buffer, which is what I don’t want! I want to study everyday. Conceivably, I can retroratchet to eep day everyday, but this gets kind of tedious for certain type of goals, especially if you have many of these “show up” goals.

The autoratchet feature is already there, but it can only autoratchet to a minimum of 2 days. I think this is a relic from when you were only able to retroratchet to a minimum 2 days, and the feature hasn’t been updated since? I might be wrong. But it would be nice if we could autoratchet it to 1 safety day (orange) or 0 (eep! day) if we wanted to.

Being able to autoratchet to emergency days solves the problem of habitualization ([http://blog.beeminder.com/habits/](http://blog.beeminder.com/habits/)) i.e. setting your YBR so low that the goal becomes harder because it’s not a habit anymore. Now that everyday’s an eep day, you have to make it a habit!

---

<div class="post-metadata">

**Author:** ![bee](https://sea1.discourse-cdn.com/flex019/user_avatar/forum.beeminder.com/bee/32/41_2.png) [@bee](https://forum.beeminder.com/u/bee)\
**Post date:** [March 4, 2015, 11:31pm UTC](https://forum.beeminder.com/t/max-safe-days-set-to-zero-bug/246/4 "2015-03-04T23:31:26Z")

</div>

Autoratchet was broken for a while, but I fixed it up a few weeks ago and it should work like manual ratchets now, which means that Do More goals ought to work with 0 safe days (slightly more detailed explanation [here](http://forum.beeminder.com/t/triangular-beeminding-for-tracking-alcohol-consumption/543/39?u=bee)).

There is one potential problem I anticipate, which is this might cause double derails, since it’s problematic ratcheting to an eep day immediately after derail. So keep an eye out for this, and drop us a line in support if you if you do set yourself up to an every-day’s-an-eep-day situation!

---

<div class="post-metadata">

**Author:** ![drtall](https://sea1.discourse-cdn.com/flex019/user_avatar/forum.beeminder.com/drtall/32/1341_2.png) [@drtall](https://forum.beeminder.com/u/drtall)\
**Post date:** [March 5, 2015, 12:01am UTC](https://forum.beeminder.com/t/max-safe-days-set-to-zero-bug/246/5 "2015-03-05T00:01:10Z")

</div>

Is this related to the issue I saw earlier when I used retroratchet (not auto-ratchet) to set 0 safe days and the goal derailed instead of ending up as an eep day? (The support ticket went to Jill on Feb 16).

---

<div class="post-metadata">

**Author:** ![dreev](https://sea1.discourse-cdn.com/flex019/user_avatar/forum.beeminder.com/dreev/32/4311_2.png) [@dreev](https://forum.beeminder.com/u/dreev)\
**Post date:** [March 6, 2015, 1:07am UTC](https://forum.beeminder.com/t/max-safe-days-set-to-zero-bug/246/6 "2015-03-06T01:07:18Z")

</div>

Yeah, let me paste the exchange here for reference:

> [@drtall](#):
>
> 1. Yesterday goal derails from $0 → $5 (legit).
> 2. Today I check the “no mercy recommit” box and then retroratchet to “0”. I wanted to simulate as if I had no-mercy checked yesterday.
> 3. Goal immediately derails from $5 → $10.

And an excerpt from @chelsea’s thorough reply:

> [@](#):
>
> No-mercy isn’t quite as no-mercy as that; we give you 2 safe days, but that is because of some technical fussiness described below…
> 
> The double-derail comes from the way the derail condition is currently set up. Two days in a row off the road = derailed, graph reset. When you try to retroratchet to 0 days immediately after derailing, the road resets as desired, but then sees that you’re now at _three_ days in a row off (two days for the legit, then today’s date as well), and it just derails you again.
> 
> So, not currently possible to do eep-ratchets immediately after derailment, but any other day is good. Support bees can make the appropriate changes to your road to make it work, though (we just need to shift the restart back a couple days). And the derail condition is certainly something that is on the list of “things to improve soon,” because of exactly this use case and people expecting _truly_ no-mercy and not getting it!
> 
> Does that help clear things up? Sorry for the bit of hassle here!! We’re totally happy to make these sorts of changes for you anytime, while the code behind things is still to-be-improved.

---

<div class="post-metadata">

**Author:** ![drtall](https://sea1.discourse-cdn.com/flex019/user_avatar/forum.beeminder.com/drtall/32/1341_2.png) [@drtall](https://forum.beeminder.com/u/drtall)\
**Post date:** [March 6, 2015, 1:46am UTC](https://forum.beeminder.com/t/max-safe-days-set-to-zero-bug/246/7 "2015-03-06T01:46:49Z")

</div>

Oh hey, that reply never made it to me! I’ve just searched Gmail for various snippets of that reply and it’s definitely not there, so maybe something got black-holed along the way?

In any case, yes that definitely clears things up. In the meantime it might be nice if the retroratchet tool would refuse to function in this corner case, rather than going ahead and derailing.

However, I was asking for retro to 0 because I forgot no-mercy gives you 2 days. I’m actually OK with that. 🙂

---

<div class="post-metadata">

**Author:** ![chelsea](https://avatars.discourse-cdn.com/v4/letter/c/a3d4f5/32.png) [@chelsea](https://forum.beeminder.com/u/chelsea)\
**Post date:** [March 6, 2015, 3:23am UTC](https://forum.beeminder.com/t/max-safe-days-set-to-zero-bug/246/8 "2015-03-06T03:23:21Z")

</div>

self fail! but also gmail fail! (but mostly self fail for not noticing.)

---

<div class="post-metadata">

**Author:** ![dreev](https://sea1.discourse-cdn.com/flex019/user_avatar/forum.beeminder.com/dreev/32/4311_2.png) [@dreev](https://forum.beeminder.com/u/dreev)\
**Post date:** [March 6, 2015, 5:59am UTC](https://forum.beeminder.com/t/max-safe-days-set-to-zero-bug/246/9 "2015-03-06T05:59:50Z")

</div>

> [@chelsea](#):
>
> mostly self fail for not noticing

Definitely technology fail despite the existence of a workaround requiring superhuman diligence.

Maybe it’s finally time to switch to one of these, if anyone has opinions:

- [HelpScout](http://helpscout.net "Seems nice")
- [SupportBee](http://supportbee.com "Though a friend says they tried this and it was terrible")
- [BeSnappy](http://besnappy.com "Possibly recommended by patio11")
- [FreshDesk](http://freshdesk.com "A friend says it's working great for GitTip")
- [Front](https://frontapp.com/ "Malcolm pointed us to this; lots of funding and big names behind it")
- [FogBugz](http://blog.fogcreek.com/fog-creeks-email-workflow/)
