Gmzamz’s Beeminder journal

Life
1

I figure I’ll give this whole public Beeminder journal a try. My only guarantee is that I do it for at least 2 months, after that I’ll think on whether it helped and may or may not continue. I try to only do automatic data input type goals, since adding another thing I have to remember to do each day means Beeminder would just derail a ton then I’d just delete my account. This will cover the past week, except for this post which will cover up to today.

Goals

Basic health stuff

  • Brushie - Todoist - Working fairly well. This goal, the floss goal and screentime alone make Beeminder worth it. I’ve seen some people split day/night, not sure if I’ll eventually migrate to that or keep a single overall count.
    • dial: It’s currently at 10/week but I’m going to bump it to 12 a week. I haven’t derailed and I’d like to brush more. I don’t think I’ll ever go past that to account for days where I’m not home or are otherwise occupied
  • Floss - Todoist - Definitely skating the line on this one. It’s still at $5 and I haven’t derailed yet, which is surprising since I’d figured my aversion to doing it was higher. Maybe I just needed something more significant than a reminder notification.
    • dial: Not increasing yet. Maybe in a week or two to 6/wk
  • Moreprotein - Todoist - This was intended as a protein+creatine shake goal, but I generally treat this as ‘any protein >30g in a single sitting’. That however loses out on the creatine when I just slam a can of tuna/salmon. Maybe I should rename this to protein shake? Or make a separate creatine goal. Or consider making this a ‘grams of protein’ instead of just ‘specifically ate protein’ combined with a separate creatine goal. Something to think on.
    • dial: Not adjusting rate
    • derail: Derailed because my shaker bottle was dirty, solved by buying 4x $3 bottles :smiley:
  • Wakeup and earlybedtime - API + iOS shortcuts - right now these are both tracking only goals. I’m not entirely sure what ‘ideal’ looks like and how I should even track these. I have both set as ‘minutes before’ the ideal bedtime (including negative minutes) and both only count the last goal (so I can go to bed, get up to do something, and go back to bed without faffing around with data entry). Ideal bedtime is just a note in the ios shortcut so that might too easy to change once this has teeth, but I’ll tackle that when it happens.
    • wakeup - I will say that right now the wakeup goal is EXTRMELY weaselable. It’s set whenever sleep focus is turned off and it’s just waaaaay too easy to do that while still in bed and just roll over. Eventually I think I’ll migrate to bed detection (somehow) or only adding a datapoint when I first complete a todo in todoist. Or potentially using the garmin2database I saw somewhere and pulling wakeup time or a heartbeat analytic to determine when I actually start moving. ¯\_(ツ)_/¯
      • dial: not yet, need to see the data to see what’s reasonable
    • earlybedtime - working fairly well even though there’s zero chance I’ll derail right now. given that it only accepts the last datapoint, it’s fairly robust. I even get a neat little notification on my watch with my ‘score’ for instant feedback whenever I trigger the automation : )
      • dial: not yet, need to see the data to see what’s reasonable
  • snooze - iOS shortcut - too easy to weasel. Just cancel the alarm and ask siri to set a timer or something - see wakeup. going to archive this one next week until I figure out a way to make it more rigid. I do like the idea of it but the overlap with wakeup makes me think it might just be a bad way to track things
    • dial: N/A since I’m archiving
    • derail: I derailed once and it just felt bad because let’s be real, I was barely even conscious when making that choice.

Exercise (my enemy!)

  • Runmore - Garmin - created this goal simultaneously with actually deciding to run. Primarily been doing long slow runs to increase endurance. sometime next early next year I’ll add speedwork. It’s working well so far but mostly because I’ve actually wanted to run, and Beeminder hasn’t had to nag me. $5 probably won’t be enough when my motivation runs (hehe) out.
    • dial: Bumping to 2.5hr/week to encourage running 3x a week.
    • Since this uses the generic ‘time’ integration it catches my walking activities which I then have to delete the next day. I’ll write a script to fix this #eventually
  • gymtime - api + ios shortcuts - how much Time I spend in the gym/doing non-running exercise. Derailed once. set to derail today unless I do some pushups or something. getting the motivation to actually work out is tough. Definitely built an aversion to it after not working out for months. Might change to a +1 ‘walked into a gym/did one exercise’ type goal. or combine with Forfeit for a month or so to build the habit again.
    • dial: uuuuuugh leaving the same
    • derail: one time - got too late and I didn’t want to work out
    • tech wise, this pushes two timepoints before and after the gym focus is enabled. Then Beeminder auto sums it. would be way easier if I could just track a single garmin workout activitiy though. #eventually

Mental health

  • Journal - Todoist - self-explanatory
    • dial: rate feels a bit low, increasing to 6/wk to reduce how ‘chunky’ it is compared a an everyday habit
    • derail: looking back, I derailed on a Friday so I probably got very distracted that night and it caught up with me
  • Screentime - API + iOS shortcuts + a slightly suss app - iPhone screentime as measured by Jomo. PRAISE BE, this has actually reduced my phone usage. Though now I’m spending some of that time on my laptop/ipad. BUT NOT ALL OF IT.
    • dial: Rate seems good for now, though I’m skating HARD. I don’t want to increase it in case I convince myself to hate this goal and want to archive it.
    • derail: I derailed once this week.
    • truthfully this goal doesn’t really ‘fit’ how I expect it to. a derail gives me a bunch more time to waste? seems like the opposite of what I want it to do. I’m considering kicking around integrating the direct charge API endpoint to have escalating stings the more I use that day. I.E. break past daily rate → $5, next hour → $10 and so forth.
      • going to try reducing the post-rail respite for now, but that feels like it might be a bit too ‘sticky’?
    • tech wise: It works right now, no idea how long it’ll work since I suspect apple might come down on the app exposing screentime like it does.

Meta-productivity stuff

  • morepomodoro- todoist - I don’t like this. part of it is that I haven’t had the motivation to actually START a pomo in a while. and partly because friction is a bit high while at the same time being too easy to cheat since it’s just a todo to check off. friction is weird because literally all I have to do is ask siri to start a timer. archiving this
  • toggl - toggl - same as above. NOT archiving this in the hopes I use toggl more. but this single goal adds another app to the number of things I have to keep track of. I don’t like this but I NEED to be more intentional.
  • beemindjournal - rssminder - partially as a way to get access to the discord, partially as a way to try to stay accountable to more than just myself (since kmy track record with self accountability isn’t great). This is the first entry and like I mentioned earlier, going to do this for at least two months.
    • dial: N/A since it’ll only ever be once a week
  • checktodos - ios shortcuts - just a +1 everytime I open todoist. zero danger of derailing. since the point of this was ‘actually use todoist’ and I have goals that require me to use todoist I’m archiving this

New beeminds

  • nothing this week, adding for the template next week
  • ideas:
    • YNAB spend less on dining out
    • YNAB less non-approved/un-categorized transactions and un-reconciled accounts. I have more than a hundred transactions to approve T_T
    • garmin walk everyweekday
    • min one extra beemind a week (or a similiar ‘improve systems’ goal)
    • whittle-down beeminder blog
    • read less fiction - no idea how I’d track this

Misc thoughts

  • legit/nonlegit - all of my derails were 'legit’, though again, not really sure about the snooze one. All the others were deliberately choosing to go to bed without doing it. Looking back, WOW 9 derails. I’m thinking that’s mostly because $5 is well below my ‘this actually matters’ point and my lizard brain can’t multiply. I can see most of my stuff moving to $30 or $90 since that much money actually feels substantial
  • Progressive decreasing of the pledge - I’ve been considering when it’s okay to dial back how much money is on the line (I suspect my pledges will get very high soon, but if that actually changes my habits… #worthit). Right now, I’m not not going to be decreasing at all for at least 3 months. I’ve just got this vague feeling that it’s okay to decrease if you haven’t derailed in some time, probably 1-3 months. And maybe only decreasing if you’ve derailed less than 1-2 times at that pledge level to avoid going below your ‘actually matters’ point. but that’s for #futureme to work on. #presentme has enough on his plate
  • API - I love it. esp since there’s always going to be some wacky custom onesie-twosie integration that literally nobody else uses and there’s no way that be doable if only official integrations were supported.
    • But also it’d be nice if there was a spec for integrations so we could crowdsource them.
  • non-cumulative goals - I’m kicking this around for bedtime/wakeup. I know I’d lose out on ‘graph go up’ but I think these goals are similar to weight loss/maintain goals where you’re trying to stay under a number.
  • diet - I’d like to Beemind this since mine isn’t great, but I’m not really sure how to track this tbh. number of healthy meals, number of actual meals and not just fancy snacks, carb/fat/protein numbers, something else? Right now my diet sucks so clearly something has to change
  • This is way too long so I’m cutting it off here

Feel free to share thoughts/tips/congrats/condolences/etc

4 Likes
2

What techniques do you use to reduce screen time and what’s your daily / weekly target if I might ask.

3

For actual techniques to reduce my usage:

  • Shortcut that shows (4hrs - current screentime) every time I open a problematic app like reddit or safari.
  • Intentional unblocking in the Jomo app - I can only unblock for a max of 15 min at a time, and I’ve manually set the ‘waiting time’ to 1 min before I can unblock
    • Downside of this is that when doing any kind of research, opening a reddit link will automatically open the app and boom bad-scrolling. Need to stop web-reddit from auto-redirecting
  • Automatically enable greyscale when I open reddit
  • remove reddit from my homescreen. Since it’s blocked by jomo it also doesn’t show up in search (usually) until I intentionally unblock it

My current daily goal is 4 hours max phone screentime. I’m also going to implement a ipad limit too soon that’s separate and probably a meta goal for combined usage

2 Likes
4

Goals

Basic health stuff

  • Brushie - Todoist - Last weeks dial is hitting this week. NSTR otherwise

  • Floss - Todoist - increasing to 6/week. That seems doable and sufficient to build the habit while still giving a bit of building a bit of buffer. Haven’t derailed yet but still edge skating

  • Moreprotein - Todoist - Renaming to 'moreproteinshakes’ to delineate from protein load meals. Not adjusting rate. Didn’t derail at all; the extra shaker bottles are working well and my buffer was sufficient for off days

  • wakeup - shortcuts - still not great. Not dialing yet, need to see the data to see what’s reasonable. But ‘alarm turns off’ isn’t good enough.

  • earlybedtime - shortcuts - Not dialing yet, need to see the data to see what’s reasonable. Getting really close to derailing despite giving myself plenty of slack. I did find out that it’s pretty trivial to handle the ‘submit after midnight’ issue in shortcuts by just subtracting 12 hours from the current time, formatting to YYYY-MM-DD and setting it as a custom daystamp.

Exercise (my enemy!)

  • Runmore - Garmin - ugh. inability to only import only running activities resulted in going longer than I wanted to before getting bugged to run. Not sure if I should delete them and eat the derail (since I did derail technically) or leave them and yeah…

    • archiving and switching to strava since I can split based on activity

  • :star: new - stravarunmore - strava edition. thank goodness. I should have done this to start with. Starting with 2.5 hours a week, same as my garmin goal

  • :star: new - stravawalkies - strava edition - I already regularly walk so this is mostly to try out Strava. Starting at 5/wk. Might be a tad unrealistic but it’'ll be nice to get out more

  • gymtime - api + ios shortcuts - workingish. I find myself doing pushup/planks in my room for the 10-15 min instead of actually going to the gym.

Mental health

  • Journal - Todoist - Last weeks increase to 6/wk goes into effect todayish.

  • phonescreentime - API + iOS shortcuts + a slightly suss app - renamed screentime to phonescreentime. No derails technically? see misc thoughts for more

  • :star: new - ipadscreentime - API + iOS shortcuts + a slightly suss app - starting at 3hrs a day. see misc thoughts

  • SOON: totalscreentime - api - cumulative total of screentime. again, see misc thoughts

Meta-productivity stuff

  • toggl - toggl - archiving. I’m not really sure this a beemindable problem? or maybe I just need a better solution for ‘intentional action’. dreams of IoT physical pomodoro timer

  • beemindjournal - rssminder - ratcheting back to 7 days. Also going to setup auto-ratchet to 8 days. I wonder if I can scope it to this thread only?

Misc thoughts

  • For gym time due to how small the time period was, I found it way too easy to just do some small exercises instead of a real workout. I’m thinking that’s a sign that I need to dial it WAY up, or implement some way to make a minimum time

  • screentime - what a shitshow. I definitely should’ve derailed on this. I don’t know why, but jomo starting going wild with the estimated screentime during the week so I switched to device specific times. Naturally that lead to binging on ipad time since it wasn’t being tracked at all. Going to setup ipad screentime too since it’s almost entirely leisure time. Not sure how to handle totalscreentime yet. Going to have to do more research

  • Last week just kind of sucked in general

2 Likes
5

Slow week. For some reason my iphone screentime shortcut got mangled and didn’t actually report? Not a huge deal since I didn’t feel like I was over-using it, but I threw an approximate manual datapoint in anyways. I did change my sleep related goals to hours before/past midnight instead of based on an arbitrary time in the shortcut. I think I’m going to migrate them to ‘do less’ goals though. That way I get an actual limit for just how late I can go to bed instead of 0:01 in X days, which isn’t particularly helpful. Do-more w/ the daily limit would be even better but doesn’t seem to be supported afaict.

Derails

Just one. put off going out shopping and ran out of floss.

Adjustments/additions

Dialing down phone goal to 3hrs/day. Eventually I’d like to hit 1hr/day phone and ipad

2 Likes
6

I think it’s doable with custom goals, but those come with one of the paid plans :confused:

Alternatively you can develop some custom data insertion (like an iOS shortcut), that checks the current state and doesn’t allow you to add more than X.

7

I moreso meant how do-less goals will show you “limit X today” whereas do-more goals say “X due by Y date”. From what I can tell do-more goals determine the slack by essentially shooting a line to the right from the current datapoint and getting the date/amount wherever it intersects with the red line, and do-less goals shoot a line going straight up and get the length of the line as the limit.

The behavior I’m looking for is a do-more goal that gets the slack by shooting a line down from the current datapoint for “limit X today”. I don’t actually care when in the future my bedtime goal derails, I only care about today. I.E. can I go to sleep an hour later, or do I need to be in bed exactly on time. I took a look through all the docs and did a search through the forums/api and didn’t see anything, so I’m guessing this is just a case of me not using the tools quite right.

Really I’m just pondering on the best way to structure my bedtime/wakeup goals. I could make it work both ways, but at least right now it seems like switching to do-less will work better with regards to how slack is represented. In either case it’s pretty easily made in shortcuts, since I can just track it differently: do-more “hours before midnight (2400 - 2130 bedtime= 2.5hrs)” with a rate of 2 or w/e or do-less “hours after midnight (just decimal time: 21.5)” and a rate of 22.

1 Like
8

I suspect that using a do-more for trying to limit the number of minutes you’re up ‘late’ is always going to have some mental gymnastics because a required number of minutes before a particular time is not human-friendly — not to my brain, at least!

Looking at your current do-less, I like the strategy of entering a -22:00 (for 10pm target bedtime) automatically in the morning and then recording actual bedtime e.g. +22:15 whenever it happens. This means that the net change to the goal will be the difference between ideal and actual bedtime.

This approach has the super-nice property of the goal’s limit being the latest possible bedtime today (with times > 24 being after midnight). I suspect that should be part of the design requirements of whatever scheme you settle on.

You can also set an auto-ratchet to impose a hard limit. e.g. never more than 3 hours after the nominal bedtime, no matter how often you go to bed on time. An auto-ratchet isn’t calculated until just after the goal’s deadline each day, so won’t be triggered by the -22:00.

And of course, if weekends are special you could auto-enter a -23:00 or whatever on those days and everything would continue to work without needing to adjust the goal’s slope.

2 Likes
9

That’s a pretty good idea, right now I’m just subtracting bedtime in the iOS shortcut that populates the data, but doing it this way is much cleaner.

The only adjustment I think I’d add would be to script a ‘ghost’ datapoint that gets updated every ~15 min until the first actual datapoint comes in. That would allow beeminder to notify me once I stay up late enough to qualify as a beemergency. Instead of thinking I have some time, then going to bed only to find out I’ve derailed. Sort of like how PPDs work but on a much shorter timeline

1 Like
10

Derails

  • I derailed on my brush goal, I thought it was only a +1 needed but turned out to be a +2 when I brushed at night. I suppose I could have just brushed again but that doesn’t really follow the spirit of the goal.
  • Running. no real excuse, just didn’t feel like running

New goals and/or adjustments

  • Split my brush goal into morning and night goals :D. RIP to a month of history
  • Re-made my bedtime/wakeup goals into do-less earlier this week. It just makes more sense
  • Reduced post-derail respites across the board. 7 days is simply way too long. I set the default to 2, but I might reduce it further. We’ll have to see next time I derail. For ‘daily habit’ style goals 1 or even 0 days feels ideal, following the ‘never miss twice’ ideal.
  • Moving my walkie deadline up. My initial intent was to walk after work and looking at my data… 9pm isn’t ‘after work’. Setting it for 7pm but adding a note in the fine-print that if I get out of work late it’s non-legit. :man_facepalming: Doesn’t work. That only affects beemergency days. Not sure how I’d setup “only log a datapoint if it’s before X” :man_shrugging:
  • ipadscreentime and phonescreentime now use #SELDESTRUCT so that I don’t have 40+ datapoints a day. Which also hoists them to the top of the urgency list? Interesting, that’s not what I would have expected but would make sense if they were PPDs :man_facepalming: That was dumb and I immediately turned it off. It being at the top of the list really bugged me.

Thoughts

I went and made breaks for upcoming holidays. Do-more was easy but for do-less I can only say ‘add X each day during the break’? It’d make more sense if it was ‘maintain whatever slack’ until the break ended. Looking forward to seeing family soon, but I’m not sure how well my habits will survive the holidays lol. In the spirit of new years, I’m thinking next year’s theme will be the year of little improvements to my systems.

On the slightly less exciting side, screentime continues to be an issue. Jomo is too easy to disable. I made my rules too optimistic blocking stuff on my ipad so it very quickly got disabled, fixing that now. I also setup the auto-racthet so I don’t get more than 6 hours a day max. Need to make a combined goal for both screentimes, I found myself migrating from ipad → phone when I ran low.

Intentionality/agency seems to be my underlying issue with a lot of my problems. It’s too easy to get sucked into doing things without making the choice to do things. Scrolling is an issue but I can’t really blame the mega-corps optimizing for engagement when it happens when I pick up a regular, physical book too. I’m really not sure how to ‘train’ agency tbh. I got a watch with a repeating timer and that helps, but (A) it’s easy to forget to turn it on, and (B) I’ve already picked up the habit of mindlessly skipping it when it does go off.

2 Likes
11

I got a burst of motivation and implemented @skorytnicki’s idea about having notifications only show a certain percentage of time. I may have perhaps gone a bit overboard, but I went and had fun with making the shortcut. Any datapoint greater than 80% of your allotted daily rate will always alert. The formula below is how I calculate the probability. It will also adjust for if you don’t have any buffer.

P(\text{notification}) = 100*\left(\dfrac{\text{screentime}}{0.8 \lfloor \text{current rate}, \text{today's safe limit} \rfloor}\right)^3

Example image where Y = chance to show a notification, X = screentime, far right is = current daily rate. The shaded area is ‘show a notification’

image
image1229×841 50.8 KB

With a blue bar at 4.5 hours representing screentime, using the green shaded area there’s a <20% chance of showing a notification if you have a week of buffer, but if you’ve burnt all your buffer and only have 5.5 hours left it’ll use the orange shaded area (with red = immediate derail). Because of the way the safe limit will decrease as you consume screentime, it will basically notify 100% of the time when you have no buffer, but I don’t plan on fixing it since if I’m skating I need the reminder

You can download the iOS shortcut for the screentime notification here, the helper floor function here and the get goal shortcut here. It does make some assumptions about what you call your goals, I.E. a variant of <devicename>screentime. You can change it in the text box in show screentime shortcut.

#feedback: This would’ve been a lot easier if a get goal shortcut w/ associated data returned as a dict (preferably locally cached so there’s no lag) was native to the app but eh, I suspect I’m one the very few people that use the shortcuts so the value might just not be there

2 Likes
12

Derails

  • Running, again. Hopefully staking $30 will make me. I’m NOT going to archive this. I WILL start actually running

New goals and/or dialing rates

  • New goal: YNAB-transactions - This tracks the oldest cleared, un-approved transaction without a flag as ‘how many days old is it’
  • New goal: ynab-more-dates This tracks the same thing, but as ‘days since epoch’. See the financial section for why I’m doing both
  • New goal: system-improvements In the theme of ‘year of little improvements to my systems’, I’ve created a goal to track them.
  • Dial: Wakeup - adjusted rate down to 1 hour/day, and ratcheted it

Tech stuff and automation

YNAB Script (Expand me!) 
#!/path/to/your/python/venv
import ynab
import datetime
import logging
from pyminder.pyminder import Pyminder
from ynab.models.patch_month_category_wrapper import PatchMonthCategoryWrapper
from ynab.models.save_month_category import SaveMonthCategory

budget_id = 'last-used'
cur_month = datetime.date.today().isoformat()
epoch = datetime.date.fromtimestamp(0)

configuration = ynab.Configuration(
    access_token = "YNAB_ACCESS_TOKEN"
)
pyminder = Pyminder(user='BEEMINDER_USERNAME', token='BEEMINDER_ACCESS_TOKEN')
bm_trans = pyminder.get_goal(slug='ynab-transactions')
bm_more = pyminder.get_goal(slug='ynab-more-dates')
date_lowest = 999999999
print(f"ynab-more-dates: Initial: {bm_more._data['curval']}")

with ynab.ApiClient(configuration) as api_client:
    api_instance = ynab.TransactionsApi(api_client)
    tran_type = "unapproved"
    transactions = api_instance.get_transactions(budget_id, type=tran_type).data.transactions
    today = datetime.date.today()
    highest = today - today
    for t in transactions:
        if (t.cleared == 'cleared' and t.flag_color == None):
            if ((t.var_date - epoch).days < date_lowest):
                date_lowest = (t.var_date - epoch).days
            if (today - t.var_date > highest):
                highest = today - t.var_date

if (date_lowest > bm_more._data['curval']):
    bm_more.stage_datapoint(date_lowest)
    bm_more.commit_datapoints()
    print(f"ynab-more-dates: Creating new datapoint: {date_lowest}")
else:
    print("ynab-more-dates: same data, skipping")

if (bm_trans._data['curval'] != highest.days):
    print(f"ynab-transactions: Creating datapoint {highest.days}")
    bm_trans.stage_datapoint(highest.days)
    bm_trans.commit_datapoints()
else:
    print("ynab-transactions: same data, skipping")

I fixed my bedtime shortcut to use the adjusted date[1] as the determinator of whether to log a datapoint. Since I gave myself Fri and Sat night off, it shouldn’t log datapoints even if it’s Sun at 1am. This was as simple as hoisting the adjusted date I use to determine what day to log under and using that date for the logic on if I should log as well.

I’m increasingly convinced I should just setup node-red and put all my workflows and automation in there instead of having everything live in scripts. It’d make it easier to write them, and easier to keep track of them. Plus I could use for other automations like Home-Assistant stuff or RSS->emails

Thoughts

Tracking financials

YNAB-transactions is my goal of the week. I’ve been super lax with staying fiscally responsible lately and kinda let YNAB go wild. I’ve got it set to have the oldest allowable transaction to be 8 days old (plus a small window at the beginning). Not shown in the graph is the previous 2 months of transactions I went through… before getting distracted halfway through and making the script to automate tracking this (see above expando). Also planning to do the same for reconciling accounts later.

That being said, I’m not super happy with how it’s structured. Right now I’ve got it setup as a weight-loss goal, but it won’t use the green>blue>orange>red pattern since the red line is flat. So it’s looking like I’ll only ever see it when it’s a Beemergency rather than during ‘blue means do’. I’m open to suggestions here. Maybe a do-more goal somehow? A rate of 1day per day sounds ideal if the Y units make sense. If I use ‘date of oldest tran.’ rather than ‘how many days old is oldest tran.’ it kinda makes sense. I’m not sure how high Beeminder goals can go, but for sure they can’t currently be rendered as dates. So 2025-11-25 would have to be represented as ‘20,417 days since epoch’ [2] . Then just have the starting datapoint be set to epochdays(today) - MAXOLDESTDAYS.

Actually, writing that out helped crystalize what I want, so I’m going to try doing it this way too and see which works better. Thus was ynab-more-dates was born. This will solve the green>blue>orange>red issue, so unless there’s some weird pitfall I think this will be the winner.

Systems

I’m working my way back through the forum to see what people have tried. I was reading through @narthur’s journal, his systems-upgrade goal inspired me to actually track my own systems improvements. Thanks! The rate is super conservative right now and matches my Beeminder journal rate. Basically meaning if I haven’t done anything lately, do something while I’m journaling. It’s just a Todoist label goal so it’ll integrate nicely into my current workflow.

How many Beeminder goals is too many? I could never imagine manually adding data to tens of goals like I’ve read some people do. But if all my goals are automated… then it seems like it’s just a matter of self-expectations?

Intentionality

At one point I used to Pomodoro’s a lot. I tried using toggl to actually track this, but it fell through because it just didn’t feel right, so I archived it awhile back. This past Friday I happened to use the timer aspect of it before playing games to limit myself and it just kinda stuck. I’m not going to Beemind it (yet) but if using it for fun things helps adoption, I might be able to shift it to slightly less fun things later on once get in the habit. It’ll still have the issue of being yet another app[3] I need to track though

Screentime

Now for the !FUN! one: screentime. I think my goals are too lax, so I’m dialing them down waiting until after the new years to dial them down. The shortcut showing my screentime is INVALUABLE, as without the instant feedback I know I would have derailed. I’m more and more convinced Jomo premium will be worth it, since I only care about useless scrolling time, and limiting time with useful things like documentation or YNAB is counterproductive. Plus it’ll let me lock my reddit rule so I can’t pull a sneaky and set it to only be active a week from now… again

  1. Adjusted date = current datetime - 12 hours. This accounts for doing things after midnight. ↩︎

  2. I’m just using unix epoch as an example since everything uses it, but any ‘days since’ counter would work ↩︎

  3. I’m fairly sensitive to the ‘too many apps’ cycle. If I start using multiple apps for my productivity system then it all falls apart because I forget to check them, then have to pick up the pieces months later and rebuild. Right now I just have Todoist. Since Beeminder is entirely automated for me, it doesn’t really have this issue ↩︎

2 Likes
13

Derails

  • None!

New goals, Dialing, and breaks

  • Dial: Wakeup → 0.25hrs/day
  • Dial: Bedtime → 1hr/day
  • Break: Big break for xmas season :christmas_tree:
  • Archive: ynab-transactions. The do-more goal was just better in every way

Tech stuff and automation

Set a base case for ynab-more-transaction. When there’s no transactions it just defaults to submitting today’s date as the ‘latest’. This combined with the auto-dialer means this goal’s automation is ‘done’

Thoughts

It’s xmas season yay!

Running

I do not like how my strava goal is setup. instead of three long runs per week, it’s asking for ~20 min a day. I know I could still do the long runs, but my hate for running is high enough that I need financial consequences to run. I zeroed it out until after new years (or rather, it will once the akrasia horizon passes). I’m planning on writing a custom dialer over xmas so it can handle forcing me to run for a long time instead of this. Basically a 45min rate on mon/wed and 20 min on fri. All other days will be zero. I also put a future rate increase juuuust in case I procrastinate setting it up forget to set it up.

Screentime

Goals are working fairly well. Perhaps a bit too much phone time, but I was able to actually put it down in the nick of time to not derail. With xmas coming up I anticipate it’ll get bad but w/e, it’s xmas. I’m thinking next year I’ll dial it down by 15 min every week until I hit 1 hour.

That being said, jomo glitched today and didn’t report the screentime right for a solid hour, so I gained an hour of screentime randomly once it caught up. Luckily I knew it might happen since the notifications weren’t updating so was able to put it down just in time. My search for a better way to handle this continues.

1 Like
14

I’ve got one pair of goals that sound similar to this requirement. The raw data goal enforces the average, and a secondary goal only gets a +1 when the day’s data exceeds a threshold.

My implementation is to naively send every datapoint to Make using the webhook, and filtering on that side before adding +1 to the secondary goal. This is wasteful of my Make quota of operations per month, of course.

Someone here wrote a custom scheduler app that lets you specify different rates on different parts of the week.

2 Likes
15

Derails

  • Iphonescreentime - Friday I went over by accident, and decided to just keep going for some dumb reason

New goals, Dialing, and breaks

  • None – all of my goals are already dialed for xmas.

Thoughts

Squeaking this entry in before the ‘extra day’ I gave it expires. expect the next couple to be pretty bare bones too. Excited for the post-xmas ‘all the new goals’ season though’!

Screentime

Technically I went over several times, but it was always due to going up riiiiight to the limit then jomo adding like 5-10 minutes. I just delete the latest datapoint with the understanding that if I open safari/reddit again, my shortcut will auto-report and I DON’T get to delete that.

There’s probably better ways to handle it, both tech-wise and ‘just stop scrolling earlier’-wise, but w/e, this is working for me right now

2 Likes
16

Derails

  • 6x non-legit from a Todoist sync bug resulting in 2x more non-legit because I eyeballed the new break instead of checking it. Whoops
  • 1x actual legit one because I’m still recovering from jet lag and slept in, which messed up my routine

Thoughts

None because it’s Xmas time

1 Like
17

Been having fun with family and not been very productive

No derails, though I’m going to pause my walkie goal until feb

2 Likes
18

Still on vacation! I set my post minder for a break until feb 1. In retrospect I should have done that earlier

1 Like
19

Derails

  • Running - just didn’t want to run that late.
  • Beeminder journal - I dunno, I could have smashed out a quick post (like this one), but was just too tired to

New goals, Dialing, and breaks

  • Archive: wakeup - more to follow on why
  • Dial: running, way down to 30 min a week

Thoughts

Going to to think on what I want to say. It’s a lot since it’s been so long

2 Likes
20

Beeminder Journal

Derails

  • four type 2 non-legits. stravawalkies, system-improvements, moreproteinshakes, and journal

New Goals, Dialing and Breaks

  • Dialing a quite a few goals back
  • Enabled weekend breaks on my stravawalkies goal. It was originally for post-work walks

Thoughts

Beemergency Days

All I can say is UGH. it seems like the whole of the last month has been barely hanging on. every single day is 4-6 beemergencies. Hence me dialing way back on some goals. There should be a middle ground between ugh and not doing it enough. I’ll come back in a month and see if this helped

Types of Derails

The way I see it there’s a couple kind of derails

  1. You didn’t do the thing. Nothing prevented you, you just didn’t do it. These are legitimate derails
    1. You didn’t do it because you forget
    2. You didn’t do it because you decided you didn’t want to
      • This is a choice I’ve made some days. I’d rather spend 5-10 dollars to avoid doing something. I don’t think I’ve done it yet for 30, that’s just too high
  2. Something prevented you from doing the thing, but you only want some of your goals to be affected. This is something like work running late, bad enough weather, etc.
    1. Legitimate derail
    2. Non-legitimate derail
      • These are goals that I don’t consider more important than whatever happened
  3. Something prevented you from doing the thing and it’s way more important than doing ANY of your things. Things like car crash, sudden hospitalization, etc.
    • I always consider these non-legit. The kinds of things that would fall into this category will pretty much always be more important than any of my goals and I’ll likely be frantically trying to fix said emergency

Type 2 Derail Matrix

I’m getting back into Obsidian so putting my goals in and making a goal tracker base was pretty easy. Here’s what I’ve got right now. It makes adding both new goals and new types of type 2 derails easy. Any time I add either, the matrix will be blank depending on what I added, which means I need to update my fine print.

image
image1198×647 51.7 KB

Global Fine Print

I also created a global fine print, but there’s not really a good way to add it to all my notes at once, or a setting for a global fine print : /. oh well, it’s basically just ‘once you hit a new type 2 derail you have to add it to the matrix’

1 Like