Scarabaea's beeminder journal

[scarabaea]

I found others’ beeminder journals interesting and inspiring to follow, so I thought I might start my own as well. I also hope that imposed regular check-ins will help against goal staleness. Since my life has been quite hectic recently, I have dialed down many of the goals, but a mechanism to revisit them and gradually introduce them back seems to be a useful thing for optimal awesomeness (I mean, a mechanism to prevent such staleness in the future).

Next January will mark my 10th anniversary with beeminder. I was thinking that to celebrate that, I might do an overview of the most helpful goal setups I had over the years. But then, I thought, why wait to start the journal just because of some random calendar date?

What I am planning to do here: several first posts will be dedicated to describing my current setup, by thematic groups of goals. After that, I am planning on weekly (or so) updates that would prompt me to save some insights for the future, as well as revise the setup based on recent/upcoming life changes.

And then, perhaps, an overview of the highlights from my archive around the 10th anniversary in January? I recently looked through the archive and found some pearls there that I completely forgot about (restarted a couple but mostly admired the ingenuity of the setups that had tremendously helped me with specific things at specific periods of my life that are not relevant anymore).

Any follow-up questions are very welcome!

[scarabaea]

I will start the description of my current setup with this most meta goal of mine. It tracks the sum of the “due in X days” for the soonest-due 10 goals.

Well, when I say, “it tracks”, it’s disinformation because for the almost 8 years that I have had this, I didn’t think of a proper way to automate it. But I am now used to counting it manually real quickly.

I am just summing the number of days in which the goal is due (actually, “min” of that and “10”) for the 10 goals that are due the soonest. Well, not all, because some are deliberately set to edge-skate, but I am using the tags and the AltBee interface to group the goals in a way that I get the goals of the type together. Mostly those where I might want to do the required thing once every several days but I don’t care on which day exactly (yoga, blogging). And those where I want to do almost every day but it doesn’t matter if I skip one (mostly set to 5.5-6/7 to compensate for the skipped days over time) (anki reviews). And those where it doesn’t matter if I do one big binge or tiny bites every day (processing my photos backlog).

So, without this meta goal, I would just let all those goals skate to “due today” and only do them when they are in the red. Which is a bit too much stress for me, having >10 goals in the red every day. With the meta goal, on the other hand, I still get to do the average of what is to be done for those goals every day, but since I have some buffer on most of them every single day, I can choose more of what to do on that specific day. Like, work through more of my photos backlog or spend more minutes on language revision?

The slope is set to 0.1/day, just because beeminder doesn’t work well with 0/day odometer-type goals. The effect of pushing to do more is mostly not from the slope but from the fact that the sum naturally deteriorates from day to day. As you can see, every single day I start with -8 to -10 from where the red line is. And if I go above the red line, it ratchets to keep only 2 points in the piggy bank.

Do I sometimes abuse the fact that it’s not autodata? Yes, I do. But it’s still one goal that helped with my beeminder use to maximize the awesomeness without unnecessarily overwhelming me, over all those years.

Initially, I thought that the aim would be to strive to reach 100. But reaching the 90s, the choice of “which of my goals I will work on more today” disappears, as I have to bring every single one of them to “due in 10 days” at least. So, it turned out to work most effectively when I am dancing in the 50s-80s.

[scarabaea]

Scholarship / Work
For years, this sphere of my life was called “study/phd” but there’s been a recent upgrade, now that I am a full-ass researcher in a post-doc position (/sarcasm)

• work_and_study
  feeds on Toggl and measures time spent on projects marked as academic- or work-related. That’s mostly been my post-doc project recently, but also some side projects are counted as well.
• focused_work_early
  This one, however, is my single most helpful goal for the year. It is my choice for Ride-or-Die challenge this year.
  Briefly, the reason behind it is that if I only track time, I am prone to procrastination and getting to do everything later in the day. So I set up a system in which tagtime pings received before 2pm count x3 and those between 2pm and 6pm x2 than those after 6pm. That has been really helpful in encouraging me to work on my project earlier in the day so far. I have been posting additional thoughts on this process in the challenge thread, will link them below:
  Goal setup
  Update for March
  Explanation how toggl-based and tagtime-based goals work together
  Update for April
  Update for May
  Update for June
  Update for July
  Update for August
• sept_drafts
  A short-term goal to work on the drafts of the papers I need to produce this fall. Odometer, based on word count. Fortunately, I started it early enough that it’s not hard to hit the daily word counts; I already have a significant buffer after just several days, and I will be able to spend time revising/shortening the text afterwards.

All work-related goals are weekends-off. I learned about the need for that the hard way over the years, unfortunately.

Languages
My language acquisition focus is now on Lithuanian, which is supported by two goals:

• lietuviukalba - time spent, tracked with toggl. Currently set to 1 hour a day. That includes everything, online classes with the tutor, homework, additional time spent watching videos/reading in Lithuanian, reviewing anki cards.
• lit_anki - boolean, whether or now I revised any anki cards that day. (Literally any number of cards. Because the problem is remembering to do that, not any aversion to finishing the decks once I have the app open)

My time-based goals for German and Spanish have been drastically dialed-down to focus on Lithuanian. Currently 6 min/day, which is probably even below retention level. But I wanted to keep a least this tiny bit of practice (Duolingo mostly and some youtube videos) because otherwise a new language tends to “override the settings” for the previous language in my brain. Like, the brain thinks, “oh, so this is now the correct way to say that!” The reality of ninth-language acquisition…
I had similar goal for French and Hebrew but they are somewhere in the archive or with flattened road in the lower part of the list.
And for some reason, I still have a duolingo_spanish goal, which I do not really care about, but it is mostly under the part of the list that I do care about anyways. (That is, the time-based goal covers for the required number of Duolingo points in excess).

[scarabaea]

Reading / Book reviews
Now, this requires a bit of context. I am a literary scholar, and most of the time spent on work, as outlined above, is about reading books anyways. But after all that time of reading only for classes/dissertation, I found myself not having read anything for fun in, literally, years. So since I finished my dissertation draft (it took 1.5 more years to formally graduate), I have been working on returning reading for fun to my life. I am mostly reading library books, and I do not track that part with beeminder in any way, because the possibility to DNF and return the book with no sunken cost is such a pleasure. But I have set up some goals to prevent backlogs from piling up.

• reviews_goodreads - +1 whenever I write a review for a recently read book, or there are no not-reviewed books that I have read recently (small print specifies the date, not important now). The aim was to encourage myself to write reviews on the finished books before getting distracted by new shiny ones. I think I firmly got into the habit since then and mostly keep the goal for QS considerations. Also cross-post the reviews at storygraph now.
• purchased_tbr - I observed this tendency of mine to read the library books since they come with a due date but to let the purchased ones to pile because they are, like, mine for good? So I recently added this backlog-reduction goal. And I really made a dent in the backlog since creating it! Just by directing the attention that way. It should be added, though, that this doesn’t account for all the books I’ve ever purchased and haven’t read, of course. All those I still have in my parental home that I purchased in my undergrad years… well, can’t really effectively work on that backlog, since I am literally in a different country now, right? This backlog that I am currently working through is mostly newly purchased ebooks, hence the relatively low numbers. Of course, I have hundreds more elsewhere.
• arcs_tbr - became a reviewer of eARCs at Netgalley and Booksirens about a year ago. But I am so prone to grab (=request) everything that looks nice and then get overwhelmed by the pile. Hence the whittle-down goal and the promise to myself to only request new things when the graph allows.

Physical activity
This sphere of my life has recently suffered a lot, as this is something that I find hardest to build into my routines when all the routines are turned upside down every several weeks. (I am kind of a nomad right now). So all has been dialed down for a long time, and I am only now trying to gradually return things to my life. Was it me who could do 3-5K runs on a treadmill several times a week not even ten years ago? *sigh *
Currently only have these ones active:

• physicalactivity - total time, 15 min/day
• yoga-stretch - instances, 3/week. Mostly yoga but at some point I widened the goal to include the possibility of other strenchy-rather-than-cardio-or-strength exercise. (cardio and strength were parallel goals, long gone)
• steps - this goal is mostly QS by purpose, as I never skate the edge. 6000/day.

Other self-care/health-related routines

Here are three of the goals that I initially started using Beeminder with, all those nearly 10 years ago: one for taking Iron supplements, one for Vitamin E supplements, and one for showering. Ha-ha, the latter one not becasue I tended to go dirty without a reminder (not that it is anything bad if I did or if anybody reading this does), but mostly because I tended to just stare into the computer/phone screen when it was time to go to sleep, and a reminder that I still needed to shower that day was an effective prompt to encourage me to start wrapping up for the day. I think I am mostly keeping this goal for historic value now, like, 9.5 years and running!

Being organized/admin stuff

• Two “gmail-fewer” goals: one for total in inbox, another one for read items in inbox. I paused them before some trips earlier this year and only very recently restarted. So they are now in an active whittling-down phase rather than normal upkeep.
• morning_setup - I get a +0.5 for each of the following: making a list of to-do item for the day; brushing my teeth; both before 12 pm. From childhood, I internalized that I brush my teeth in the morning before I leave home, not after breakfast. Which is kind of the same if you eat your breakfast and leave your home to school/work right after that. But I would just skip on the days I am working from home so often. Well, Beeminder to the rescue (of my teeth)! The to-do-list part is optional on weekends or when I am travelling, I get the entire +1 for just the teeth on those days.
• Two goals related to my photos backlog: a whittle-down one accounting for the entire backlog (at 18,501 right now, ugh) and a do-more one for the number of those that I dealt with. They are set to 10/day and 15/day, correspondingly. The difference allows to, well, take new photos before having worked through the entire backlog.
• Two support goals for the spheres of life outline about: one to remind to refill the vitamin holder every week and one to reorganize the Anki decks every month.

Not sure what group do these goals belong to, let them be here as well:

• wordpress_entries - making posts to my personal diary using the wordpress integration, 3/week
• contact_alp - keep in touch with my granma (those are her initials), usually 1 Zoom call/week, but if both of our schedules don’t work in accordance on a given week, a series of back-and-forth text messages also covers the weekly minimum.

---

And that’s it for my current setup! Will report back on how everything is working in a week or so

[poisson]

A lot to absorb, but quite interesting!

It would probably be relatively straightforward to automate the buffer thing with a script running on your computer, but… at the same time, it seems like it’s working quite well already, so perhaps not worth the effort!

The tagtime thing is interesting to me because I also have a toggl work-hours goal. Currently trying to augment it with tocks to improve the amount of “focusing on getting something done” during the hours worked, but the tagtime method also sounds like it could work well.

I should probably have something analogous to sept_drafts to push me to finish things up, as well. When I was writing my dissertation I did have a word count goal as well. And then later, I tried some kind of "number-of-checklist items’’ goal when I was writing up a big fellowship application. Nowadays, I have no such beeminder goals, but I do also have a draft perpetually languishing and another project I should be trying to finish up, so…

Anyway, I could babble forever but I’ll stop now. Looking forward to next week’s report!

[scarabaea]

@poisson

It would probably be relatively straightforward to automate the buffer thing with a script running on your computer

Now that you wrote this, it dawned on me that nowadays it wouldn’t be such a hard thing to do, my coding skills already go about halfway of what is necessary for that (and most importantly, I imagine the entire “way”). But initially, it seemed like a very complicated and daunting endeavor, so I preferred to “shirk and turk” (as Danny put it on the blog) and just got used to it. Well, that’s very useful to revisit those year-long habits.

Anyway, I am glad that you found some of these relatable and probably inspiring to implement for yourself!

[skorytnicki]

Hey, nice work.

I’m curious how do you manage so many goals, how much time does it take to manually insert the data? Isn’t it overwhelming?

Second question: Why do you learn so many languages at once?

[scarabaea]

Thanks!

@skorytnicki

how do you manage so many goals, how much time does it take to manually insert the data? Isn’t it overwhelming?

After ten years’ experience, I don’t find it overwhelming. In any case, there have been periods when the number of active goals was way more overwhelming. That “meta” goal I described first really helps.
Today, for example, I have “touched” 14 goals (=have ticks that suggest there’s data entered for today)
Three of them are supplementary tagtime goals that feed into the fourth. That’s four less to deal with manually.
3 more are autodata goals that feed off toggl, and 1 more that feeds off wordpress integration.
So, that’s only 6 goals to enter the data manually, and I am pretty used to just doing it throughout the day, as I do the required things.

@skorytnicki

Why do you learn so many languages at once?

I actually don’t learn several at once, otherwise, they mix into a mixture in my head. I usually wait at least a year between starting new languages. And then I usually focus on one, keeping a couple more in the backburner just for the sake of retention, to prevent the neural pathways from being overwritten.
Lithuanian is my focus right now, since my new home institution for the post-doc position is in Lithuania, and (apart from the regular considerations of learning the language of the place where I spend significant time) I would like to be able to include some Lithuanian material into my research for contrastive purposes sometime in the future.
What I am doing to stop German and Spanish from totally slipping from my memory (6 minutes a day on average) I wouldn’t call “learning,” really. Language retention is like running up a downward escalator, and I feel like the escalator is faster than my running here. I used to be pretty proficient in Hebrew, taught it for years… and then I had this conversation several weeks ago when I realized that even simplest words (like, first-semester-level vocabulary) have slipped from my memory. Such a shame. But I am fully aware that I don’t have the mental capacity to add the efforts necessary to recover those skills right now.
Sorry for a lot of rambling - that’s a sore point with me.

[skorytnicki]

@scarabaea

and I am pretty used to just doing it throughout the day, as I do the required things.

This is why I ask, as it doesn’t work for me at all. I wanted to find out how can one have over 20 goals and keep up with them. I try to automate everything.

Have a nice time in Lithuania!

[scarabaea]

General impressions

• This week was a slower one, as far as my work/postdoc goals are concerned, as at some point I discovered that I am very uncomfortable reading on screen and that continued for a couple of days (must have been some new hair care stuff getting into my eyes, but clearly I didn’t run repeat tests to check this hypothesis). Fortunately, all three of my work-related goals had enough buffer to allow that. (I didn’t completely stack off, did some reading ahead on my e-reader, which was more comfortable for my eyes than computer or phone screen. But that meant less progress toward the nearest expected output - an article draft, as well as only 9.3 hours spent on anything project-related this week).

Derailments

• None

New goals

• forum_journal - to make sure I remember to post here, of course! Set to 1/week.

Changes to goals

• A funny coincidence, exactly on the day when I was reminded in the conversation with @poisson above that I could make a script to gather the data for my meta goal calculating the total number of safe days across a list of goals - exactly on that day, AltBee interface that I was using to calculate that manually stopped working for a couple of days. So, I did hack a little Python script that gets the current info and sends a datapoint with the API.
  After sending in the datapoint, it also provides me with a hint on how much is left to work on the goal for the day and what other goals can be helpful in reaching that. Like this:
  

• I have been playing around with the autodialer and enabled it for several goals to try it out (setting my previous rate as the min, so that it doesn’t fluctuate below that). I am still not sure how I feel about that; very likely that I will turn that off soon.

• Have been making good progress on cleaning my mailbox, so I ratcheted both of those goals, so as not to be encouraged to leave emails to pile again. BTW, apparently it is better not to use the ratchet function on whittle-down goals but to adjust the road manually or with the graph editor. The ratchet thing doesn’t understand that an inbox cannot be cleared down to a negative number

[poisson]

As insane as it is, I still always print papers to read them.

The script looks great! I have some scripts for adding data to meta goals, but some kind of data summaries would probably be worth thinking about as well…

Is there a particular reason you think you will end up turning off the autodialer? I’ve been thinking about, well, if not autodialing, at least having a systematic procedure for adjusting the slope of my goals, so I’m curious if something about it seems bad to you.

[scarabaea]

@poisson, thanks for the encouragement to think about coding a script to do that!

@poisson

Is there a particular reason you think you will end up turning off the autodialer?

A couple: 1) I like my goals (other than explicitly short-term ones) to run for years, and having a lot of road changes makes graphs heavy and slow to load. Once I realized the autodialer changes the road every single day, I understood I might be creating myself a problem long-term.
2) I realized there’s a value to me in knowing clearly what my rate is and comparing how I am doing with regard to this ideal rate by gaining or losing days of buffer. I found the daily fluctuation by decimal points annoying. Maybe it works differently for those who prefer to have all or most of the goals in the red almost every day but this is my experience with my setup where I always have buffer on most goals.
I actually went ahead and turned off the autodial since I wrote that update earlier today. Instead, I decided to introduce short-term changes using graph editor. And I will try to remember to critically reconsider my goals in that light often. I think it is about the balance between the long-term and the short-term impact of the goals. For example, I want to beeminder to nudge me to do yoga 3 times a week on average in the long term, but if for a specific period of time my routine is such that I should better be doing it 5-6 times a week (for the lack of any other physical activity), it is better to introduce it explicitly through break interface or graph editor than use the autodialer which will creep up the rate every day, as it sees that I have been doing a bit better recently. I.e. datapoints should follow the rate, not vice versa.
The most useful insight from the week of playing with autodialer is the 30-day average. Because it was surprising for me that for some goals my average is below my rate. That’s probably due to some past breaks or having gained buffer prior to the 30-day window and then using it up within the window… but that was a surprise. I would love to have that recent average visible somewhere for all my goals. Now that I have been experimenting with API, might as well try that.

[scarabaea]

General impressions

• This week didn’t feel as a super-productive one either. But I am working through my to-do lists and I am still keeping on top of my goals, so I guess it’s fine. I think this is one of the most important effects of beeminding for me: to mitigate the self-scolding for not doing enough. All these numbers are giving me a solid ground to persuade myself that, yes, more could have been done, but if I am still on top of my goals, it’s fine, even if I had to use up some of the previously earned buffer. (Helps against spiralling into “I am useless” and not doing anything at all)
• Exactly half a year after I started my focused_work_early goal (my 2024 challenge, the goal which prioritizes the time spent on my postdoc earlier in the day) - it feels like the gamification effect of it is wearing off and the goal just stopped motivating me by its pure existence. I am ploughing through the previously earned buffer these days, so I still have several days to regroup and find some solution for this new feeling. Part of that effect is that my primary task right now within the general post-doc project is to write an article draft. And it has always been harder to put myself to sit and write. Of course it is easier to just read more (primary sources, others’ scholarship).

Derailments

• None

New goals

• I have revamped the setup related to reading. In addition to whitte-down goals that count owned books and ARCs, I have created a goal that measures a total of those categories + books checked out from library. (Caveat: requested ARCs count, requested library books don’t, as I still have an option in Libby not to check out the book if my turn came unexpectedly. So I don’t have to count them here just on the basis of adding myself to the queue.)
  
  Maybe that’ll prove excessive at some point, and I will archive some of those.
• Another related metric to ARCs - netgalley_feedb_r measuring the feedback ratio at NetGalley. For those not familiar with the idea: that’s a website where advanced copies of new/upcoming books can be requested in exchange for a review, and it is suggested that the feedback ratio should be at least ~80%. That is, of those ever approved for me to access, I have reviewed 80%. I am still not sure if that’s a valid thing to strive to, because the purpose of that is that it’s easier to get approved for next ARCs, and as soon as I am approved, the feedback ratio plummets again… but here we have this goal for now.
• Finally, I started migrating (or archiving) my reviews from Goodreads to a separate blog, so I created a goal for that: bookblog_wordpress. Once I am done migrating and the new posts are only the new reviews on recently finished books, this goal will supercede the reviews_goodreads.

Changes to goals

• As I mentioned in a comment after the last week’s update, I have disabled the autodialer after experimenting with it for a week, and instead set up increased rates for some of the goals that I saw thanks to the autodialer that I am capable of doing more of currently. Not indefinitely, but just for the period of time while I am in my current location, until my next journey and change in routine. I am planning to include temporary changes of that kind in the routine now.
• Speaking of that upcoming travel, I have just introduced short breaks for Sep 26 to 28 for the most of my goal, as I will be travelling for almost 40 hours on those days. I would normally do that closer to the dates but I realized these weekly reviews are a good point in time to consider all upcoming changes for which it will be already too late by the time of the new weekly review. (Most goals = almost all that don’t already have buffer through that moment in time, minus the focused_work_early that has its own rules for allowed breaks).