Beeminding by k1rsty

[k1rsty]

I love reading other people’s threads about their Beeminder goals so here’s my contribution to the genre. I’m aiming to post once a week about one of my goals. I currently have 25 active public goals - and a handful private ones, and most of them could do with me writing myself a manual for what they are for, how they help me, how they work etc. And of course: here’s my meta goal for posting to this thread.

bookreviews

The first goal I’m highlighting is one that helps me keep on top of writing book reviews on the Goodreads website.

why I do it: I only do this for fun, not profit or even free books (which would be stressful), and have reviews dating back to 1999 that got imported to Goodreads at some point. I really appreciate the time past-me has taken to tell current-me why they did or didn’t enjoy a book. I don’t read super fast - to average a book a week is my usual aim and I miss that quite often - but it’s easy to end up with a backlog of books I want to write a bit about but by the time I get around to it I can’t quite remember everything I wanted to say.

how it works: I set up an IFTTT applet to watch the RSS feed produced by Goodreads and add datapoints to my beeminder goal, so I only need to publish a post on Goodreads and Beeminder gets updated by magic. To find my RSS feed at Goodreads (and there may be an easier way to do this!) I navigate to the bookshelf page I want and look at the page source code and search for a link with “rss” in it, it’s on a line that begins <link rel="alternate"

settings: the goal is set to 3/month which is slower than the rate I usually read at, which stops the goal from stressing me out. If I miss a book it doesn’t really matter, I just don’t want to miss every book.

current state: the goal has over four months of safety buffer as I do actually get around to posting about pretty much every book I read eventually because I like having the complete how-many-books-I’ve-read stats at Goodreads as well. I’ve ratcheted the goal occassionally - and it looks like I gave myself a break another time, probably because I wasn’t reading much at that point - but I’ve never actually derailed on this goal and it’s been very successful at making me do what I want to do.

changes: I’ve only got one book I’ve finished reading that hasn’t made it to Goodreads yet, and another couple that’ll be finished this week or next so I’m ratcheting the goal back to having four days of safety buffer so I get that book written up in the next few days. And I’m also raising the slope to 3.5/month so I don’t get so far above the line quite so quickly, but I want this to stay as a goal that is easy to stay ahead of. It’s been a successful goal so far and I just want it to keep doing what it has been doing.

[k1rsty]

Next up is a goal that I’ve just reached the target on.

blanket2021

why I do it: I have a loooong history of starting craft projects… and a lot of them don’t get finished. I have a general craft beeminder that makes me do stuff regularly, but that’s more for the mental health benefits of doing something than to finish things. So I also like to make individual goals that are focused on actually finishing what I start, especially for big projects. This goal was to knit a blanket composed of 49 knitted squares.

how it works: Manual entry of how much of a square I’ve knitted. Each square was about 50 rows so it was easy to enter a decimal value of how much I’d done that day. As a bonus it also tracks what row I’m on which has got me out of trouble with the patterns a few times.

settings: Initially I set the slope to 20 squares/year, and needed 49 squares to make a blanket.

I really like the story the graph tells. I knit about ten squares and put the project down for few weeks. In pre-Beeminder days this would have been where the project got put in a cupboard and forgotten about. You can see how in August I get to the “pick it up again or pay” point and decide to pick it up again (even though the commitment is for $0). Then I get all enthusiastic about it - it’s like starting a new project but someone has already done some of it! - and I get ahead of the line, ratchet away the safety buffer and increase the slope. Enthusiasm wanes and I skate on the edge of the line for a while at the end of the year but keep going. And then at the beginning of 2022 I can see the end in sight and race for the finish line. I have a new major knitting project scheduled to start at the beginning of March so that’s why I moved the last segment of the line to make sure I’d have finished knitting this before then.

I also think this one looks nice like this (graph generation info is in this thread)

current state: Finished…

changes: but what I have now is a pile of 49 knitted squares, which is not a blanket. The squares need all the ends sewing in, sewing together, and a border needs knitting round the edge. So I need another goal to actually finish this. I’ve got no idea how long it’ll take me to do all that (back of an envelope calculation: “yonks”) and I want to be free to concentrate on the incoming major project so I’m just going to commit to spending 2 hours/week on working on this blanket until it’s done.

New goal: blanket2021finishing which is going to be a manual entry time graph. I’ve already added the 2 hours I spent sewing in ends yesterday evening and I’m hoping to build up some safety buffer during the rest of the month so I can dive into my new project on March 1st without feeling guilty about abandoning this one a little.

[k1rsty]

ebike

why I do it: In August 2020 @bee posted that…

We decided on a new family rule that any new equipment aquisitions should be backed by a beeminder goal.

At the time I was umming and ahhing over whether I should buy myself an electric bike and had a list of pros and cons written down. One of which was something like “but will it just end up sitting in the garage with my old (non-electric) bike”. Of course the answer is beeminding.

how it works: Straight forward link from Apple Health via the Beeminder app tracking the distance cycled. Distance cycled tracked using Apple Watch apps, both the default workout one and WorkOutdoors.

settings: I had lots of info on how much I’d ridden my last bike which I’d bought just over ten years before. The electric bike was four times the price that that one had been. I worked out how far I’d ridden around home in ten years. I excluded holidays as I thought they skewed the data a bit much - I live in a hilly region and take my bike on holiday to flat places for a reason! So something with a factor of four made sense - I did a quick calculation of what four times the distance in a quarter of the time would be - laughed at myself because that’s a factor of sixteen - but the answer wasn’t stupid, about 45km a week, so I went with that as an initial goal.

When I first bought the bike I got ahead of the line pretty fast, and ratcheted the line up, only to run into mechanical problems that made me derail. I got ahead again, had more mechanical troubles, so I haven’t done any more ratcheting.

current state: I’ve cycled a lot less this winter than the first, and the weather has been awful lately but I have enough buffer that I think I can stay on track with my original goal.

changes: A couple of months back I realised that I was going to cycle a lot less in the second year I had the bike if I just kept to the original goal, which seemed a bit disappointing. I had cycled just about 2000 miles in the first year. So I added an extra goal to cycle another 2000 miles in the second year - and I’m sticking breaks in this one to keep it on track for when the days are longer and the weather is nicer. This is my stretch goal so if it becomes clear it’s failing then I don’t really mind that much, but I want to try and keep at it as long as I can. It’ll probably involve more effort than last year as lockdowns very much worked in my favour last year encouraging me out onto my bike for exercise when the swimming pools were closed. This year I will need to do both. I’m just putting another break in this one, leaving the target date and distance the same, while I wait for a spell of storm-free weather to arrive!

And, yes, my mile based goal is in km because Britain is a mixed up place that continually frustrates my attempts to be an SI unit person.

[skorytnicki]

how it works: I set up an IFTTT applet to watch the RSS feed produced by Goodreads and add datapoints to my beeminder goal

Clever; I wrote my own script to do that, but IFTTT integration is much better idea.

how it works: Straight forward link from Apple Health via the Beeminder app tracking the distance cycled. Distance cycled tracked using Apple Watch apps, both the default workout one and WorkOutdoors.

Do you use Strava? It produces nice charts and has beeminder integration.

[k1rsty]

@skorytnicki

Do you use Strava? It produces nice charts and has beeminder integration.

I signed up a thousand years ago and tracked one ride that led to a disparaging comment so I didn’t carry on using it. It seemed to be aimed at speedy riders only then. That was 2012 apparently, and while searching through my email for that date I found a friend request on there from my mother, so I guess it must be friendlier/more private/more tolerant of old slow people than it was then. And I know I’m missing out on other cool tools by not using it so I might try it out again, since my account is still there.

[k1rsty]

weeklydraw

why I do it: Admin work task that needs doing weekly and takes a maximum of ten minutes but I end up not doing it and end up with a huge backlog that’s then a pain to deal with because it’s not fair on other people to batch the task up, it just needs to be spread out.

how it works: Manual entry of data to beeminder.

settings: Straight forward 1/week setting. The only thing interesting here is that, when I set the goal up last year, beeminder helped me get this task back on track to actually happening once a week by steadily clearing the backlog over the course of a few weeks, that’s the steeper slope at the start of the graph. Its derailed once which then put me on a schedule that was really a week behind for a long time. I ratcheted to remove that leeway earlier this year.

current state: Up to date and working well.

changes: None required.

[k1rsty]

update on things above: It occured to me after that last post that saying “changes: None required.” was missing something. What I should be thinking is: I have a goal that’s working well, how can I apply this to something else and make that better? So I’ve just set up another goal for a similar kind of once a week work task that I always leave to the last moment, which should encourage me to get ahead of myself: pdprep

---

After reading the latest Beeminder blog post Control Systems for Backlogs I thought it’d be a good idea to highlight my gmailzero goal and how it works, and often really hasn’t worked for me. But I think I’ve made it better now.

why I do it: Because the integration is there. Because email is terrible. Because I hate email. Because I ignore things in email. Because I like getting nice email. Because I like feeling on top of things.

how it works: Standard beeminder integration. Which only counts the read messages in your inbox. Which is a problem. Over the years this goal has steadily encouraged me to never read my email. If I don’t read it, then I never have to deal with it. Result! Hmm, no, not really. And then people start texting me or phoning me with things that would have been easier by email which is a really bad result. I think the concept of this goal is good, but my usage of it has been awful.

settings: Recently this has been set to keep the messages in the inbox with below 20, it’s gone up to 40 after a derailment but I have it coming back down again.

current state: Oh, boy. I wanted to break the whole “don’t read your mail” thing, and I came across an article on setting up Gmail filters to automatically delete/archive mail which seemed like a good idea. So I can read things and have them tidy themselves up without my needing to click the pesky “archive” or “delete” button which is, for no good reason, a task that is beyond me.

But then I realised I was spending a lot of time setting up filters for emails and a lot of things were getting past the filters that should have been tagging them as archivable or deleteable anyway. I just seemed to be creating more work to have a simple task automated than actually training myself to do the task would be, and I’d already determined that training myself to do the task was too hard.

But then I realised I had it backwards. Whilst the “delete me” tags on email were kind of useful for clearing things out of my Gmail storage space entirely, the “archive me” ones were pointless. If I’ve read it then it ought to archive itself unless I’ve marked it to stay. So I hacked the scripts from the link above to archive any mail I’ve read that hasn’t been marked as important or starred. And now that runs in the evening shortly before Beeminder checks my inbox count for the new day and again in the middle of the night, and that seems to be keeping on top of things much better.

I now actually go through my email box and look at every message. Which means I see things in a timely manner. And I’m enjoying reading the newsletters and other updates that I have signed up for. And it’s easier to unsubscribe from things I don’t want. And email seems to have become good again! But maybe I should leave it awhile before concluding that. I’m finding that Gmail has started to mark things as important for me that really aren’t, probably because I’ve read lots of messages like them, which might become a problem.

changes: I’m very happy with this at the moment, and, following on from last week’s postscript, wondering what else I can apply similar techniques to.

[k1rsty]

@k1rsty

I’m very happy with this at the moment, and, following on from last week’s postscript, wondering what else I can apply similar techniques to.

Thinking about this was remarkably enlightening!

• I have lots of digital inboxes with lots of items (articles, videos, tabs etc) and I really don’t care about the amount of things in them at all, they are no more of a problem than the library having more books than I can read. It’s nice to realise what the things are that don’t need beeminding!
• There are a few things that could do with regular clean ups - e.g. deleting spam answerphone messages - but they just aren’t important enough to bother beeminding.
• I keep a to do list that gets clogged up with stuff I need to do but don’t do - okay, I’ve found something it would be useful to beemind!

But then I instantly found a load of reasons that would be a bad idea…

1. I’m the only person who puts stuff on my to do list, I could just cheat by never writing stuff down
2. I can just mark things off, or shift them to the next day rather than doing them.
3. There’s no way to automate the beeminding with the software I’m currently using
4. I’d have to actually do things
5. Maybe it’ll go badly and Beeminder will stop working for me at all

To which my better self responds…

1. But I know writing stuff down is a really, really good idea. Stuff makes more sense whens it’s out of my head. I’m not going to forego that just to save myself a beeminder charge.
2. Marking things off that don’t actually need doing is fine! Shifting things to the next day is a bit more shady, and might require some clearer rules but let’s wait and see if it’s an actual problem.
3. If all goes to plan then counting the items left on the list at the end of the day shouldn’t be very onerous.
4. That’s not a bug, is it?
5. As ever with Beeminder, if it all gets too much you just pay up (or stick with it a week) and get out of there, and/or start again.

I did spend a little time going down the rabbit hole of “ooh, I need a shiny new to do list app that I can link to Beeminder” but I somehow remembered Gall’s Law before I got too far down. My current app is a long way from perfect, and I’ve taken to putting important work stuff in a different app, but I’m better off trying to get something I already have going to work a little better.

So, now I have a new goal: todoboxzero

how it works: Manual entry of data from my to do app. I found the app will give me a badge on my phone home screen telling me how many of today’s items are left so I don’t need to count them. I discovered you can’t have “pessimistic presumptive” points on an “inbox-fewer” type goal as the system has decided “they don’t make sense”. I think they do, and so does this thread. As suggested in that thread I set up an IFTTT applet to add a big datapoint to my goal automatically each weekday morning so this wouldn’t be a goal I could forget about.

settings: To start with I’ve made the goal completely unchallenging. Over the next few weeks it’s going to step down until it only lets me leave three items on my list each day. I plan to step it right down to zero once I get better at getting to three!

current state: So far - a few days - it’s going well.

• I’ve made myself do awkward annoying things rather than leave them until the next day as I haven’t wanted there to be a batch of awkward annoying things to deal with at once.
• I’ve found myself adding things to my list so that I will go and do them straight away, which means I have a better record of what I’ve done.
• So far I haven’t noticed myself refusing to add an item to the list, if anything I’ve added extra stuff and then done it because I want to show myself that this system will work for me.
• I’ve also been going through tomorrow’s list in the evening and checking it is all stuff that seems doable tomorrow and putting it in a reasonable order to attack it in, and I think that’s being really helpful. I feel like not shifting any items off today’s list except to complete or write them off entirely is probably the most sensible approach. It’s like a mini-akrasia horizon.

changes: I want to make sure the work tasks get back on the main list, and maybe look at that shiny new to do app, but it’s early days and really I want to get the system running for a while before I try and make it do more.

[k1rsty]

Two very similar goals that I set up in the same way at the same time: ynab keeps me up to date with my personal finances and sage does the same for my business ones.

why I do it: Both tasks are just a bit tedious when done regularly, but if they get very behind then they turn into real tangles, so these goals help me not get behind.

how it works: I struggled to find a good way to beemind “don’t get more than x days behind” and eventually went with “I’m up to date as of y days after 14th Nov 2019”, ie day 0 of the goal. I have a couple of iOS shortcuts that calculate that number for me and post it to beeminder. I like that I don’t have to be completely up to date, just not too far behind.

settings: Both goals originally let me be 14 days behind. The ynab goal has never derailed and so that setting is still the same. The sage one has derailed once and now lets me get 21 days behind as a result.

current state: I’ve thought of ratcheting the sage goal back to 14 days but I think I’m leaving it where it is as the extra buffer isn’t a problem.

changes: Leaving these exactly as they are, unless I think of a better way to beemind “x days behind the present day”. They work just fine. I do feel like I’ve got in the habit of keeping both things updated and that I don’t really need them any more but knowing I’ll get to update some beeminder goals when I’m done is an extra incentive to keep on top of the tasks so they are staying.

[k1rsty]

clearout

why I do it: Because I am bad at tidying up, lazy at organising and completely useless getting rid of things. I very much have a “but it might be useful for something one day” attitude and I don’t like wasting things. My partner can just throw things in the rubbish and never think about them again, when I try that I make bad decisions about how important things are to me and regret them. But I also don’t want one of those houses that’s floor to ceiling piles of junk!

how it works: Just manual entry of a data point every time I clear something out. I deliberately made this goal very vaguely defined so that I’d be able to do what felt good in the moment, it just has to be something bigger than what I’d do on an everyday basis. So, for example, throwing odd bits of mouldy veg away don’t count but clearing out the whole fridge of out of date products does count. Throwing a pair of socks with holes away doesn’t count, clearing out a shelf of wardrobe would count. Sometimes I count things where nothing has really been ‘cleared out’, tidying and organising so I know what I have and can find it is a good thing.

settings: Twice a month. If I made it more often I think I’d just find myself finding smaller things to do to count towards the goal rather than getting more done. I’ve derailed a couple of times, but also ratcheted the safety buffer back when I’ve got ahead of myself. I often use the trick of adding “0.1” to the goal to avoid derailing while I think and figure out what to do. Sometimes several times, but then when I clear something out I only make the goal up to a round number.

current state: Working pretty well in that I really can’t be bothered with it at the moment but am making myself do things anyway.

changes: I’m keeping the goal the same but what I am going to do is start keeping a list of things that I could clear out to satisfy this goal to make it easier to just do things instead of "0.1"ing. The comments I’ve left with the Beeminder data points will help here. Sometimes I have loads of ideas, other times, like now, I just draw a blank in a “can’t see the wood for the trees” kind of sense.

[k1rsty]

rescuetimegoal and productivitygame

why I do it: To make myself do actual productive things on my computer.

how it works: Both these [beeminder] goals actually use IFTTT to add datapoints to Beeminder when I reach various different [rescuetime] goals for amounts of productive work done in various timeframes. I’m not sure why I’ve never[1] used the built in Beeminder/Rescuetime integration which I think allows you to just add time to a Beeminder goal rather than using Rescuetime’s goal system.

settings: The rescuetimegoal goal is older and can get up to two datapoints a day added to it, one for a certain amount of productive time and one for a certain amount of very productive time. The productivitygame one I set up in Jan 2020 when I was feeling I was spending too much time on my mobile devices and I tried to set it up so it gave me points for productive hours on the computer (at any time of day) but also subtracted points from me if I racked up hours on my mobile devices during my self-defined working hours.

current state: Neither goal is holding me to anything these days.

• I’ve set the rescuetime one on autodial, just as a test thing to see where the slope went. It has autodialled itself to about 6/week at the moment, which seems low for a goal that can be 2/day, but I feel I’m working pretty consistently and productively at the moment.
• The productivity game one… I still think the idea is good but RescueTime on iOS only seems to work if you check it a lot, whereas the beauty of RescueTime on the computer is that it keeps on tracking no matter what. I pretty much forgot that I had it running for over a year once and it was a delight to find the data after all that time. But the “subtract points for things you don’t want to do” doesn’t work reliably.

changes:

• I’m going to archive the productivitygame goal, it got me out of a rut when I set it up but doesn’t have ongoing usefulness.
• The rescuetimegoal one I’m going to leave for now. I like seeing the slope of the datapoints change over time. It gives me info that I don’t see directly in RescueTime. And it’s a blunt instrument that I can keep in reserve for a time when I need it, if I ratchet away the safety buffer then it’s ready to spring straight into use.

I do think that I’ve pretty much exhausted the usefulness of RescueTime, I think I’ve had it running since 2008 or so. It has trained me not to do non-productive stuff on my computer in working hours pretty well, which I think 2008ish-me would be happy with. I like seeing the trends in the data. I have a subscription that runs for another six months or more but I’m wondering if it might be time to move on.

On the whole I think more specific short term goals are better for getting me doing productive work than these goals have been. It would probably be a good idea to beemind something like giving a regular answer to the question “do you have a work thing that you are beeminding at the moment, that is actually making you do things, and if not, do you have a good reason why not?”

[1] Apparently I did use the Beeminder-RescueTime integration directly before I set these goals up. I found this in my archive:

That has hours tracked directly rather than goals reached. I think I prefer the “goals reached” metric because it encourages me to work in smaller chunks. 9 hours on one day are not equivalent to 3 days of working 3 hours. I gravitate towards the one long day but I know my work is better quality if I do the three short days.

[k1rsty]

risebedtime

why I do it: Because I’ve slept badly for years and bought a quite expensive app subscription last year which I wanted to help me get better at sleeping. The point in paying that much for an app was that it’d make me actually use it and follow its suggestions. And then I’ve beeminded it just to make sure.

how it works: This is manual entry for lying down to sleep in the window given to me by the app. What I really wanted was to use an iOS shortcut to get my sleep times (from the Health app) and check them against the sleep window which the app can put on my calendar. One problem was that iOS shortcuts involving health data do not like being automated - I could have coped with that, but I also had trouble with the shortcuts to get the sleep data seeming to feed me garbage data so I gave up on even a manually run shortcut. This is just completely manual entry.

settings: I set the goal off at three times a week, which is about my usual “that seems reasonable” level. I put it on autodial after a while. Which dialled the slope up as I got in the habit of using the app, and down as I got in the habit of forgetting to enter the data.

current state: I am sleeping really well! But I am not entering the data for this goal regularly.

changes: I’m archiving this goal. I’m doing the thing, and beeminding it without automation is just turning out to be too much trouble.

[k1rsty]

instagram

why I do it: At the end of 2019 I went to look at the “your top 9 instagram posts” thing and found I had only just about made 9 posts in the whole year. And I found that a bit disappointing as instagram is my favourite social media kind of thing and, for better or worse, it ends up being one of the main ways I share things with family and friends.

how it works: A straightforward IFTTT applet that posts a datapoint to beeminder every time I post on Instagram. The applet was very flakey for a while and I was manually updating the beeminder goal but it’s been going since last August without a problem now.

settings: I set it up as one a week, that was about 120 weeks ago and I’ve made 115 posts. I’ve had it using the autodialer recently but that isn’t the right thing for this goal.

current state: I like it how it is. Often enough that I have to remember to document things, infrequent enough that I don’t post total junk.

changes: I had 55 days buffer thanks to a recent run of posts, I’ve ratcheted it back to ten days. I’ve switched off the autodialer and the rate is back at 1/week.

[k1rsty]

read

why I do it: Because from time to time I forget that reading is one of the things that makes me me, and reading regularly is one of the best mental health workouts there is. It’s challenge and rest all at the same time.

how it works: This is one of my oldest Beeminder goals, and considering it’s just manual entry of the number of pages I’ve read I’m surprised it’s lasted so long, but also, that tells me how important this is to me. As you can see from the graph there was a spot in 2016-2017 when I decided it was too much trouble to update but I came back around to it and have kept it up more or less ever since.

settings: It looks like when I started in 2014 I set the slope to 13000 pages/year, which I’d say was about 40 books/year. It’s been as low as 10000 pages/year (excluding the flat bits). And now it’s set at 16000 pages/year which is probably more like 50 books/year.

current state: I still love having this goal but it’s a bit of a pain when I get close to the edge and am having to update the pages read data every day to stay on track. I prefer to keep plenty of safety buffer and just update the page counts of books when I finish them. I tend to wait until I run out of safety buffer to update the data though and then I have a mad scrabble around trying to remember what I’ve read and looking up the page counts.

changes: I realised recently that I could tie this goal in with some automation (via IFTTT) to work in conjunction with my bookreviews goal from the top of this thread. I put pretty much every book I read on that goal already so I now have it posting a zero data point to this goal whenever I review a book. Then all I have to do is pop in and change the page count from zero to the right number. I’m sure there is a way to find the page counts automatically but this way I remove the hurdle of having to remember what books to add at least. I think doing the page counts automatically might bring up annoying issues of differing editions having different page counts for example so I’m not worrying about further automation for now. I’m happy with my about 31 days of buffer too. I am just going to push the pages/years commitment up a little again though.

[skorytnicki]

On tracking pages read:
Exactly, this is possible with IFTTT. I have an automation that checks if my goodreads RSS feed has an update this day and adds 1 if I read something. This does not check how many pages I read yet, but I work on adding this missing bit.
This setup however, still makes you log data via goodreads. Similar pain, but I want to keep my data scattered across services, so that I don’t rely only on Beeminder or only on Goodreads.

I wonder what’s the best scenario of tracking pages read. If I switch to 100% ebook, maybe it’s possible to hack things around and make it fully automated.

[k1rsty]

I was about to derail on posting to this thread. I’ve been super interestingly productive in other places and my read-the-beeminder-forum-its-useful-and-totally-not-procrastination levels have been low as a result. This is a good thing, although reading the beeminder forum is always a good thing too!

Instead of derailing here’s the world’s most boring kind of beeminder graph, which is also a good kind of beeminder graph:

existmood

why I do it: I love tracking all kinds of stuff over at Exist and finding patterns in my behaviour and correlations between different aspects of my life. I can easily forget to do it though, especially when I’m in busybusybusy mode as I am at the moment and this is a beeminder goal to remind me to keep my details there up to date.

how it works: This is just a shell script that runs regularly and looks to see if I’ve posted a mood rating on Exist and posts a single data point for the day to Beeminder if I have done. I want to do more than post a mood rating on exist each day, but the mood rating acts as a proxy to show I’ve been into the app and done things there. While I’m there putting the mood rating in, I’ll always at least add my custom tags about what I’ve done on the day. Most of Exist’s tracking is automated but the bit of extra effort to add my own extra data to the automated stuff adds a lot of additional value to the service for me, especially now you can have actual data values in custom tags. (And hopefully soon they’ll link that into their API so I can track stuff on Beeminder and copy those values to Exist (or vice versa) and figure out the perfect levels of everything . )

settings: Hmm, the beeminder page for this goal is telling me it’s set to 0/week which is clearly wrong from looking at the graph! graph.beeminder.com says 6.5/week though so we’ll believe that one.

current state: I’ve built up 4 days of buffer.

changes: I’m cutting the safety buffer to a single day, updating the post-derail respite to be two days rather than 7, but adding a break in for an upcoming holiday where I might be out of internet range, when I get back I’ll ratchet my safety buffer away again.

[k1rsty]

gamevalue

I think this is the loopiest of my beeminder goals!

why I do it: In mid-2020 I realised I had just bought several new board games, and that had made me feel miserable. My previous strategy was to limit my board game spending by budget and these were within budget; the problem was that I wasn’t getting any board game playing in with the ~waves hands~ whole 2020 thing going on. I decided I needed a way to link the budget and whether games were getting played together, and beeminder is my go-to hammer whether or not the thing I’m trying to fix is actually a nail or not.

how it works: Each board game play is recorded as a data point, and each currency unit of board game purchase costs a certain amount of plays. I already recorded my board game plays on Board Game Geek so I wrote a script to pull that data into Beeminder. Purchases are negative data points that I calculate and add by hand.

settings: I accidentally pulled in far more data from BGG than I intended to when I set the goal up, but I also had records of what I’d spent on board games over the same period so I added that as well, and I decided that an exchange rate of one play of a board game putting £0.75 in the savings pot for new games worked out about right. I was initially intending the beeminder graph to have a flat line, but beeminder isn’t very keen on that, and I like the idea that my savings pot will slowly empty itself if I stop playing for a long time. So the road has a slope that amounts to one play getting cancelled out every three weeks. It’s kind of stupidly complicated but quite simple to operate in practice. I log my plays like I always do, and then I have an iOS shortcut that queries the buffer from beeminder and converts it into £ for me.

current state: It’s working exactly as described above and I haven’t changed the settings at all.

changes: I’m not minded to change anything. This is a case where I’m certainly not going to cut my safety buffer because that’s my savings pot and I have games I want to buy with it, I’m just not buying them all at once. The exchange rate I began with seems to be working about right, and I’m very happy to be back playing games in person again. Actually the one thing I should probably change is to write an iOS shortcut to quickly update the goal when I spend money, since I did that today for the first time in ages and I had to go back through the data to remember what the exchange rate was and whether to multiply or divide the cost to get the right figure to input.

[k1rsty]

craft

why I do it: I find regularly sitting down to do something non-tech related like sewing or knitting is just an all round good idea. I like that I make things so it’s kind of productive but its also just a good thing for my relaxation and mental health.

how it works: When I set it up in 2015 I set the slope to 3 times a week, and I manually added the data points. At some point I started having other goals for craft projects so I used IFTTT to update this goal when I added a point to one of those other goals. And changed this goal to only accept one data point a day so that I couldn’t cheat the 3 times a week thing. I don’t have a minimum amount of craftwork that needs to be done to count as I find getting over the inertia of just picking up a project is the hard bit. Telling myself I’ll fulfill the goal if I just sit down and knit a handful of stitches usually results in me doing a lot more than that.

settings: The goal has mostly been at three times a week all the way through, with occasionaI reductions. I set this goal up to use the autodialer earlier this year as it’s really a secondary goal with the goals that feed into it as the main ones. I can’t imagine derailing on this now as it’s such an ingrained habit, but it’s taken me numerous derails to reach that level. Success has been as much about carving out the time as anything else, and about always having something ready to work on without having to think about that.

current state: The current slope is 2.6ish, mostly because I often forget to update the goal that feeds into it every time I do something on it. I ended up with a flat spot recently as I’d forgotten to hook up my new craft goals to feed this. There’s actually only my blanket2022 knitting goal feeding into it at the moment.

changes: I considered switching it back to a straight three times a week as the autodialer is just masking mistakes I’m making, but actually I don’t think that’s the right answer. I’m going to trim the safety buffer back as 90 days is ludicrous, but mostly I’ve realised I need to set up goals for other craft projects so I’m encouraging myself to work on those too. So I’ve just set up a sewing goal with a very minimal commitment, and a sashiko goal to get me to finish a project I was given last Christmas by next Christmas. And both of those are already linked up to join up to this goal via IFTTT.

[k1rsty]

Beeminder

why I do it: Last year I grabbed myself a free trial of the Apple Fitness Plus service when I needed an indoor fitness option. It was a winter day when I’d tried to go for a walk but had been defeated by ice. I enjoyed it and subscribed and set up a Beeminder goal to make sure I used the subscription. Well, it wasn’t my most successful Beeminder goal ever.

I was struggling to find the time to use it when the outdoor world got a bit more hospitable. I kept derailing and making the road less steep. At the same time I was using the short sessions to kinda-cheat the goal, and some data points look like they are outright-cheats where I exercised elsewhere but gave myself credit here. So I gave up, gave up the subscription and was kinda-sad.

Then we had the Hammertime club thing in November and I looked at it again. I compared the cost of the service with other fitness related things I pay for and concluded that the service was worth it to me if I got the annual subscription and used it for 20 minutes a week. Which seemed reasonable. So I thought I’d have another go and said I’d keep the goal going for a year which could have been a lot of derailments if I’d got the parameters wrong again.

how it works: It’s just a manual entry goal that I update the number of minutes on everytime I use the service. It’s kind of annoying that I can’t get an iOS shortcut to update the goal automatically, I might have to look at the workflow again but iOS health data is locked in so it is only accessible when your phone is unlocked. Which is understandable, but it would be nice if you could power-user override this stuff sometimes!

settings: Originally 20 mins/week and the autodialler has raised that a bit.

current state: I’m feeling like I’ve successfully carved out one chunk of time a week for a yoga session and am working on carving out a couple more slots without impinging on the other exercise I do. I have a load of safety buffer.

changes: Whilst I love safety buffer - I like doing things so tomorrow-me can be lazy - I don’t need more than a couple of weeks worth for any reason so that’s going to get ratcheted away a bit now.

[k1rsty]

Hmm, I derailed on my beeminderthread goal last week and am posting now so I don’t derail again, but I feel like I’ve run out of goals I fancy sharing info on. I may turn this thread into an “updates on how the goals are going” type of thing but I’m not sure about that at the moment either. So, I’m posting now then putting this on hiatus for a while.