2022-23 New Year review

This is an annual post reviewing the last year and setting goals for next year. Overall, this was a reasonably good year with some challenges (the invasion of Ukraine and being sick a lot). Some highlights in this review are improving digital habits, reviewing sleep data from the Oura ring since 2019 and calibration of predictions since 2014, an updated set of Lights habits, the unreasonable effectiveness of nasal spray against colds, and of course baby pictures.

  1. 2022 review
    1. Life updates
    2. Work
    3. Health
    4. Parenting
    5. Effectiveness
    6. Travel
    7. Fun stuff
  2. 2022 prediction outcomes
  3. 2023 goals and predictions

2022 review

Life updates

I am very grateful that my immediate family is in the West, and my relatives both in Ukraine and Russia managed to stay safe and avoid being drawn into the war on either side. In retrospect, it was probably good that my dad died in late 2021 and not a few months later when Kyiv was under attack, so we didn’t have to figure out how to get a bedridden cancer patient out of a war zone. It was quite surreal that the city that I had visited just a few months back was now under fire, and the people I had met there were now in danger. The whole thing was pretty disorienting and made it hard to focus on work for a while. I eventually mostly stopped checking the news and got back to normal life with some background guilt about not keeping up with what’s going on in the homeland.

Work

My work focused on threat models and inner alignment this year:

Health

Physical health. I’ve been sick a lot this year – 6 colds and one bronchitis since Daniel started nursery in June, plus one cold earlier in the year. Had covid in April, thankfully a mild case with no obvious long-term effects. I also had two short bouts of covid-like symptoms (fever, muscle aches and fatigue) in May and October that lasted about 2 days each. I tested negative for covid both times, and recovered too quickly for flu, so I’m pretty confused about what this was, maybe a bizarre form of long covid?

Being frequently sick was pretty depressing and demotivating, and I put some effort into decreasing the rate of catching colds from Daniel. I tried improving hand hygiene and not sharing food with Daniel, which had a lot of overhead and didn’t seem to do much. I also experimented with various supplements, starting with vitamin C and zinc, which didn’t seem to help much, and then added beta glucans and broncho-vaxom, which possibly helped but I’m not sure. The only thing that seemed clearly effective was a nasal spray called “dual defense”, which seemed to make any symptoms go away whenever I applied it. This made the last (probably) cold I had mild enough to be barely perceptible (not included in the number of colds above).

Sleep. Similarly to last year, I consistently slept for 7 hours at night on average, with a standard deviation of 1 hour. The rate of insomnia was 10% of nights, better than last year (20% of nights). I was awake for an average of 0.6 hours (40 minutes) each night.  (As usual, all the sleep metrics are excluding jetlag.)

I have now added some Oura ring data to my life tracking database as well. The ring provides a score for my sleep each night and “readiness” for the day. These scores are on a scale from 0 to 100, where presumably a score of 100 means you’re completely refreshed and ready to move mountains and 0 means you’re about to drop dead (on both of these dimensions, the highest score I’ve ever had was around 90 and the lowest was around 30). These scores take into account the amount of sleep, frequency of waking up, heart rate and body temperature at night, and activity levels. The ring usually detects when I’m sick, assigns a low readiness score and suggests to take a rest day. I didn’t wear the ring during the day between March 2020 and October 2021, which resulted in much lower activity scores, but I’m not sure how this impacted the sleep and readiness scores.

One interesting thing I noticed is that while the amount of sleep per night has stayed level at 7 hours in the past few years, my sleep score has been trending upward. I switched to the Generation 3 Oura ring in Jan 2021, which is supposed to measure sleep quality more accurately, so this could also be an artifact of the change in measurement rather than an actual improvement in sleep.

The readiness score shows no upward trend – it’s averaged around 70 the whole time.

Mental health. Better than last year, but not as much better as I hoped. There was a definite improvement in the first half of the year. In January, I shifted my meditation practice to self-love meditation, which was helpful for a while but seems to be wearing off (maybe I need to find some new recordings on Insight Timer…).

There were 7 episodes of particularly bad mental states, mostly in the second half of the year. Being sick a lot in the fall was a major factor – I often found myself judging my body as weak, being angry at my immune system, or judging myself for not protecting myself enough when Daniel was sick. I think the self-judgment also led to a hopeless mindset where I felt like I tried everything feasible to avoid getting sick when I actually had not, e.g. I later tried the nasal spray and it seemed to help a lot.

One improvement in mental health this year was a decreasing rate of night terrors (waking up startled soon after falling asleep) – I had 13 recorded this year, and 37 recorded the previous year. This might have something to do with Daniel getting older and me having less subconscious worry about him falling or getting trapped under the blanket or whatever. However, I developed a new anxiety symptom after he started walking and bumping into things and making lots of mess – I often noticed myself holding my breath when taking care of him. I try to get back to normal breathing when I notice it, but it tends to come back when I’m not paying attention. It’s been a bit better lately, but still not a solved problem.

Parenting

Breastfeeding. I continued to breastfeed Daniel this year, with decreasing frequency as he asks for it less often. I think at this point there is no more milk, and he is just looking for comfort when he asks for a feed. I never really figured out a plan for how to stop breastfeeding, and I’m still not sure what the endgame for this looks like.

Potty training. We transitioned Daniel out of diapers using the “oh crap” method over a long weekend in May, which went pretty well. He is good at using the potty when prompted, but took him a while to learn to ask to go to the potty – he’s getting better at this now but we still need to prompt him a lot. He usually has a few accidents a week, which seems ok. These days he doesn’t wear a diaper during the day except for naps and long travels.

Sleep. Daniel usually sleeps from around 9-10pm until around 6-7am, with an average of 0.3 wakeups per night (excluding jetlag). He had a sleep regression in November (which seems to be common around 2 years of age), so he started waking up more and being more difficult to put back to sleep. It’s interesting to compare the data on wakeups and night feeds (12-6am) – I often managed to put him back to sleep without breastfeeding during most of the year but it didn’t work anymore during the regression.

Childcare. Daniel started full-time nursery in June (it’s open until 6pm, which works great with our 10-6 work hours). He also spends Sunday afternoons and evenings with his nanny (who used to care for him full time after nursery), which gives us some time together to do our check-in with each other, go climbing or relax in the sauna, though often a bunch of this time block gets eaten by logistics.

Taking turns. Janos and I alternate taking care of Daniel in the mornings, since neither of us is a morning person (though we’ve shifted towards an earlier sleep schedule since having a kid). Starting in June, we also introduced a schedule where each of us gets one evening at the office per week while the other one takes care of Daniel. These arrangements were quite helpful for giving me more sleep, productive time and alone time, and setting up regular time blocks for Janos to be alone with Daniel.

Languages. Daniel is pretty talkative in English and Russian (still working on the Hungarian though). He knows to address me and my relatives in Russian and other people in English. He is starting to say long words and short sentences, and recently got into the habit of reciting his favorite songs and stories from memory in both languages. It’s not always clear which language he is speaking, which is a bit confusing.

Effectiveness

Lights. I continued using the Lights spreadsheet for tracking daily habits that I started using in 2021. I’ve stopped tracking a few habits and started some new ones, but overall the set of habits mostly stayed the same – here are the habits that I kept from last year:

  • Life tracking
  • Make a list of intentions (in the todo notebook)
  • Ask myself what I want today
  • Meditation
  • Exercise (changed to “today or yesterday”)
  • Leg & shoulder stretches
  • At least 2 hours of deep work (if working)
  • Braindump / journaling (at least 1 sentence)
  • Reading
  • Appreciate a thing I did today
  • Exchange appreciation with Janos
  • Go to bed by 11:30pm

Habits from last year that I dropped:

  • Emotional release practice (mostly superceded by self-love meditation)
  • Avoid processed sugar (doing this anyway, don’t need to track)
  • Use work cycles (doing this anyway in the form of time tracking)
  • Check internal dashboard (didn’t resonate / wasn’t useful)
  • Go outside (more automatic post-pandemic)
  • Notice when I am picking my nose (didn’t work well)

New habits I added this year:

  • Fill out lights (makes it easier to see which days were filled out retroactively, was intended to motivate me to fill out lights every day but that didn’t work)
  • Negative visualization on making mistakes (helps with self-judgment)
  • Practice effective rest (breaks during the day where I pay attention to what I want)
  • Take supplements to avoid / mitigate colds
  • Use eye drops (to address dryness from using contact lenses)

I did about 70% of the lights on an average week. The most difficult lights were deep work, going to bed and reading. The main failure mode with Lights was not filling them in on some days (usually weekends), which resulted in doing fewer of the habits on those days. I have a solid habit of filling out the lights at the office, but I need to have a more reliable time block to do this on weekends (probably after lunch during Daniel’s nap).

Time tracking. In June, I switched from using work cycles to doing time tracking during work hours. I realized that I wasn’t doing much of the built-in reflection in work cycles, and was mostly using them as a less systematic time tracking setup. The time tracking shows that in an average work week since June, I spent 27 hours on work activities: 9 hours in (non-research) meetings, 7 hours on research, 4 hours on reading, 3 hours on comms (giving feedback on docs, giving talks, etc), 2 hours on planning and 2 hours on admin. I also spent 10 hours on non-work activities: 6 hours on self-care (exercise classes, therapy, meditation, naps), 1 hour on parenting, and 3 hours on random stuff.

The easiest way to improve on this is to increase work hours – I can add another office night (every 2 weeks), and experiment with going in to work early on mornings when Janos takes care of Daniel. I also hope to spend less time being sick next year, with the more effective supplements and Daniel hopefully bringing home fewer germs after the first half-year at nursery.

Deep work. I did 363 hours of deep work (1.7 per work day), compared to 311 hours of deep work in 2021 (1.78 per work day). This was more than last year (and resulted in a lot more output), but still short of the pre-parenthood baseline.

This was the first full year (with no leave) since 2019. The number of workdays was a bit lower than the expected 225 workdays in a normal year (260 weekdays minus 10 holidays and 25 vacation days), which was mostly due to sick days for myself or Daniel. The rate of deep work per work day was lower than 2020-2021, mostly due to going to conferences again (which are usually work days with no deep work).

For the purpose of this summary, work days include weekends where I did at least 2 hours of deep work. There were 14 weekend workdays in 2019 and around 3 on each subsequent year (unsurprisingly, having a kid decreased my ability to work on weekends).

Digital habits. In the spring, I read the Digital Minimalism book and felt inspired to set up better systems for intentional use of technology. The book recommends doing a digital declutter, where you stop all technology use that is not absolutely necessary for a month and then add some of it back in a limited capacity. This seemed a bit extreme and I couldn’t get myself to do it, but I made a list of necessary and optional technologies and how I would like to use them (which was useful in itself). I implemented various measures to cut down unnecessary technology use, which were generally effective:

  • Using grayscale on my phone by default to make it less visually stimulating. After I got used to grayscale, the regular full-color mode started to seem too bright and overwhelming, so I don’t really want to use it unless I’m looking at photos or watching videos with my kid.
  • Using an app (Actuflow) that asks to enter my intention whenever I unlock the phone. Together with grayscale, this reduced the number of unlocks from 70-100 per day to 20-40 per day. This includes using the phone to watch videos with Daniel, so the number of phone unlocks just for myself is lower than that.
  • I got into more of a habit of going to places without my phone sometimes, e.g. going for walks, getting lunch, or working in the library. I got a wristwatch to be able to check time without looking at my phone, which made going without it a lot easier.
  • I muted all channels on work slack except those specific to my team and projects.
  • I installed News Feed Eradicator for Facebook and Twitter on my computer browser (which are the only social media I use). I still check the news feeds on my phone sometimes, but not that often since it’s less convenient in grayscale and in the phone browser (I don’t have the apps installed). It would be great to restrict the Facebook feed only to life updates like someone finishing their thesis or having a baby, but sadly this option doesn’t seem to exist (probably by design).
  • I also wanted to use the Inbox When Ready extension that hides the inbox by default, but it is unfortunately not allowed on my work devices, so I compromised by defaulting to the (usually empty) Starred view of my inbox.

Travel

In March, we went to the Bahamas for an AI safety workshop. To enable both of us to attend the workshop, we imported Janos’s dad and his partner from Toronto to hang out with Daniel on the beach in the meantime.

We spent two weeks in August in a cottage on Manitoulin Island with family. Daniel was eager to hike with the big hiking stick (or better yet, two of them). He was less into swimming in the lake than last year, except when jumping off a paddleboard. He was also really scared of the resident squirrel at the cottage for some reason, and ran inside whenever he heard it chirping.

In September, we caught the tail end of the hiking season in the Dolomites – it was chilly but pleasantly uncrowded. Daniel was a trooper as usual and learned a lot of new Russian words hanging out with his grandma. We were hoping to do some via ferrata climbing but the weather was too wet for that.

We went to Toronto for the winter holidays and got a couple of days of skiing before all the snow melted. Daniel took some time to get used to the cold weather, but he really enjoyed throwing snowballs.

Fun stuff

  • I read 10 books this year: Three Body Problem, The Gifts of Imperfection, How to Talk so (Little) Kids will Listen (and Listen so Kids will Talk), Oh Crap Potty Training, Hunt Gather Parent, Bacteria to Bach and Back, Messy, Digital Minimalism, and Decisive.
  • Janos and I took Daniel to the “In with the spiders” exhibit at the London zoo, where you can see big (and harmless) spiders up close. Daniel loves spiders so this was a treat for him.
  • We did some aerial silks in the park, for the first time since having Daniel (I still remembered a few tricks). It took a surprisingly long time to find a good tree to rig the silks on (with a big sturdy branch at the right height).
  • We visited the Bay for the first time since before the pandemic, leaving Daniel in Toronto with his grandma to get a break and spare him the jetlag.
  • We had a photoshoot with a professional photographer – turns out, Daniel has a very photogenic smile :).

2022 prediction outcomes

Resolutions

  • Author or coauthor 4 or more AI safety writeups (70%) – yes (8 writeups)
  • Meditate on at least 230 days (70%) – yes (272 days)
  • At least 450 deep work hours (70%) – no (363 hours)
  • Do 3 consecutive chinups (60%) – no (2 chinups)
  • Avoid processed sugar at least 6 months of the year (60%) – yes (the whole year except a few days)

Predictions

  • I will not catch covid this year (60%) – no (got it once)
  • I will write at least 3 blog posts (2 last year) (60%) – yes (3 posts)
  • I will read at least 5 books (70%) – yes (10 books)
  • Daniel will be potty-trained by the end of the year (out of diapers when awake) (70%) – yes (since May)

Calibration

This year was pretty good:

  • 60%: 2/4 correct
  • 70%: 4/5 correct

Calibration over all annual reviews:

  • Overall my predictions tend to be overconfident (the green line is below the blue line, which represents perfect calibration).
  • I was overconfident in 2014-16, underconfident in 2017-19 (probably to compensate), and went to being overconfident in 2020-22.

2023 goals and predictions

Goals

  • Meditate on at least 250 days (272 last year) (80%)
  • At least 400 deep work hours (363 last year) (60%)
  • Write at least 4 blog posts (3 last year) (70%)

Predictions

  • I will avoid processed sugar for at least 10 months of the year (80%)
  • I will read at least 7 books (80%)
  • I will catch at most 4 colds (60%)
  • Daniel will be potty-trained for the night by August (70%)

Past new year reviews: 2021-22, 2020-21, 2019-20, 2018-19, 2017-18, 2016-17, 2015-16, 2014-15.

2 thoughts on “2022-23 New Year review

    1. Victoria Krakovna Post author

      Happy to hear that you enjoyed the post and that it helped to inspire your investigation of cold treatments, I enjoyed reading your review of those! I hope you will have fewer colds in your house soon :).

      Like

      Reply

Leave a comment