Antidepressants and Medical Uncertainty

I’ve occasionally talked about how great my experience with antidepressants was. First one (2015) worked great and reduced my trigeminal neuralgia to boot. But it wasn’t enough so I started a second one (2017), which was also great and also helped my trigeminal neuralgia with no other side effects. I knew this experience wasn’t universal, but I would occasionally share it so people would have data on the best case scenario.


I tried lowering both meds last summer, but each time my neuralgia got worse so I resumed my regular dose. Then in January USPS delivered my pills two weeks late, and I was forced to go off them (luckily I saw this coming as was able to taper with my remaining pills, so it wasn’t cold turkey). Like over the summer my neuralgia got worse, as did my anxiety, but the depression definitely did not come back.

A bunch of stuff happened here. First I was fine, then I was sleeping an awful lot, then I was sleeping quite a bit less (and felt fine about it). Every time I felt bad I would wonder “is this a transient reaction to an external stimulus, a sign of returning depression, or both?” When I had my checkup with my psychiatrist and mentioned that I no longer needed a daily nap, she immediately said “Oh that’s [medication 2]”.

I’ve talked to multiple doctors about my previously quite extreme sleep needs, including her. They all tried a bunch of complicated stuff, and they all had access to my list of medications, but no one ever said “hey, it might be this medication you’re on, let’s try adjusting that”. It’s possible the medication was the right choice for a period of time even when it was eating two hours of of my day, it’s possible it wasn’t eating two hours of my day and it’s a coincidence things have improved. But when my doctor is so sure it was the problem now and nobody even mentioned it before, something has gone deeply wrong somewhere.

There was a time this would have really freaked me out and possibly triggered a panic attack, but TBH it’s about what I expect from medicine at this point. So the point of this post is mainly to be an accurate data point for people assessing antidepressants for themselves, because God knows the medical community isn’t going to give them to you.

I still think medications were mostly a success for me. I really needed them when I started them, I might really need them again in the future and would consider taking them even with the risk of nap. But I would pay a lot for a doctor who even raised this possibility.

PS. Given what I do for a living, “why didn’t you research this yourself?” is a reasonable question. The answer is:
1. I was not in a good place to do this when I started them
2. The fatigue did not kick in immediately- I was actually hypomanic for a bit, and the mandatory naps kicked in at least a year later
3. I did in fact know that fatigue was a possible side effect of the medication, but I was taking it off label at a drastically lower dosage than is typically prescribed (<1/20th), so it didn’t seem very applicable. I don’t think better scientific studies are the solution here, humans are too variable. What’s needed is a system that is responsive to individual feedback, including doing experiments to get that feedback.

What can a covid test tell you?

The following is extracted from research I did for a client (which I’m sharing with their permission). An important thing to keep in mind while reading this is that the information was gathered to answer particular questions (Primarily “quarantine procedures for very risk sensitive, cost insensitive people”), and that biases what I looked at in weird ways. But I think it’s still much more useful for some set of you to have this information than not, so here it is.

The Tests


Of young, healthy people caught by prophylactic screening (typically contact tracing), 30-50% who test PCR+ will never report symptoms (in some cases these subjects are proactively screened for temperature, in others it’s not clear. Additionally, other studies have found that some percentage of people who report no symptoms will go on to develop CT lung anomalies). Additionally, most people who do develop symptoms spend several days being contagious before they develop symptoms.  So symptoms are not a very good metric at all for determining if someone is infectious.

 [Full notes and sources]


PCR tests look for specific viral RNA sequences and amplify them to make them easier to detect. They can vary in a number of ways:

  • Specific sequence searched for
  • Number of amplifications done
  • Collection site and mechanism of sample

PCR false positives are quite rare, and typically have to do with sample contamination. False negatives are more common, and can come from a variety of sources:

  • Covid was present but did not have the particular sequence searched for
  • Did not do enough amplifications to notice effect
  • Patient has covid but the particular site sampled does not.

How common are false negatives? That’s hard to define, because currently a nasopharyngeal PCR is what gets you diagnosed as having covid- if you fail that, you’re assumed to have one of the many things that produces similar symptoms. However, we can make some guesses.

The following graphs show the calculated chance of testing PCR+ on a nasopharyngeal test on a given day after exposure, given frequent testing and an eventual positive result.

But that’s just for nasopharyngeal (NP) tests. Data on sicker patients (the only ones studied) show it’s possible to have a negative NP test while another area of the body (lower respiratory or gut) tests positive.

Figure thumbnail gr1

But these are all people who tested PCR positive eventually. What about people who get sick without ever testing positive? A very small study found that of 24 health care workers who developed positive antibodies over a 3 month period, 10 never had a positive PCR test despite being tested twice weekly. The paper offers multiple explanations for this, and I’m very reluctant to draw conclusions from such a small study, but it is concerning.

[Notes on NP tests]

[Notes on stool tests]

[Notes on spit tests]

[Notes on lung tests]

Antibody tests

Antibody tests look for an immune reaction to viral proteins. They can be negative when a person is contagious (because they haven’t formed a large enough reaction yet) and positive when they are not (because they successfully fought off an infection). 

A reasonable question I have not investigated is “does an antibody+ test mean I’m immune?” 

[Full notes and sources]

LAMP tests

Like PCR, LAMP tests look for specific viral RNA sequences and amplify them to make them easier to detect. LAMP tests go through many fewer amplification cycles, making them less sensitive but much faster: 30 minutes vs 2-3 days for PCR, plus the transit time for the PCR samples. 

A LAMP test every three days is much more valuable than one PCR test, because catching the peak is much more important than the sensitivity. I’m not sure if a LAMP test taken today is more or less useful than a PCR test taken to give results today. I stopped digging into this because there were not yet any LAMP tests on the home market.

[Full notes and sources]

Antigen tests

Antigen tests look for specific viral proteins in a sample. There’s no amplification, which makes them less sensitive, but manufacturers report catching 88% of cases caught by PCR, so maybe this is fine, especially since you get results days faster? 

The 88% sensitivity number is much, much higher than you see in literature (which is something like 20%-50%), I assume because they used more abundant samples. There’s some controversy over whether that’s because many PCR positives are driven by dead virus, in which case the antigen test returning a negative result is a feature, not a bug. This may well be true, however at least one study was able to culture live virus from a PCR+/antigen- sample, so it’s not foolproof.

Of the studies looking at antigen tests, all that I found either started post-symptoms, or took a random sample of people showing up at a sampling site. None were in a position to determine how good antigen tests are as an alarm system for catching an early infection (as opposed to diagnosing symptoms or determining when someone has ceased being contagious)

However antigen tests return results in minutes, while PCR tests take 2-3 days to complete not including shipping time. Given how quickly covid multiplies, it’s possible that an antigen test now is more sensitive than a PCR test from four days ago, especially if the exposure occurred in the intervening time.

[Full notes and sources]

What do any of these mean for contagiousness?

That’s a good question, which with the exception of asymptomatic spread can’t really be answered without human challenge trials, which the entire world declined to do. I would be surprised if any of these tests’ thresholds lined up perfectly with the threshold of infectiousness, because there’s no reason to expect they would. I expect culture tests to be much closer, but still not necessarily exactly on line (plus AFAIK they’re not available for diagnostics, even through a doctor). Given that I asked…

How long do I need to stay in quarantine after potential exposure?

The best data on this is from New Zealand, which has a very strict isolation policy for entrants: 14 days isolation, with PCR tests on days 3 and 12 of isolation (a pre-departure test is required only if you’re coming from the US or the UK). By my earlier estimations, day 3 of isolation is already likely to be past the peak of detectability, unless travellers were exposed right before they left. Nonetheless, New Zealand reports only a handful of import-related covid cases after this policy kicked in, and a model attributes those to people who caught covid in isolation (e.g., a couple who shares a hotel room and one partner gives it to other on day 4), rather than people who entered New Zealand infected.

Transmission type of COVID-19 cases

I did look for similar data from other countries, but NZ had both the best quarantine and the best data.

So despite the tests’ low sensitivity, 14 days + a test at the end really does seem to be a long-enough isolation period.

[Full notes and sources]

Thanks to my original clients for funding this research, and my Patreon patrons for contributing to the cost of preparing this for publication

Follow Up: Predictions as a Substitute For Reviews

In August I wrote about using PredictionBook as a substitute for weekly reviews. Several months later, I am rather predictably not using it, for the same reason weekly reviews stop working: it reminded me of something I endorsed doing but didn’t finish, and that put an ugh field around the whole thing.

Weirdly, I re-picked-up that thing a week ago. This was completely independent: I haven’t looked at PredictionBook since September. I don’t think this has any bearing on the viability of predictions substituting for reviews, it’s just a cute coincidence.

Interview with former cult member Duncan H.

As my twitter followers already know, I started a deep dive on cults back in September. To the extent I had a goal for this, it was “figure out behaviors that differentially push you away from actual cults, without throwing the baby out with the bathwater”.

I mentioned this to a friend, who suggested I talk to a friend of hers, Duncan. Apparently Duncan gets this request a lot, because his condition for talking to me was that we record the conversation so he can share it with the next person who asks. We had mixed success at this- the audio quality was horrible for the first section of our conversation, and the transcriptionists had limited ability to save it.

But the second half got quite interesting. Some highlights:

  • Duncan got a lot from the cult and would make the same decision again (and then leave again).
  • If power is creepy, only creeps will have power.
  • How society interacts with strong belief.

Part 1 video (low quality) and transcript
Part 2 video (much better quality) and transcript

These calls lasted 2 hours and I still had more I wanted to talk about, so I suspect you do too. Duncan has indicated a strong willingness to answer more questions, as long as they are interesting. So if there’s anything you’re dying to know after listening/reading these, comment here, @ me on Twitter, email me at, etc.

Thanks to Patreon patrons and Duncan for helping to defray the cost of the transcripts.

Looking For Work

Three and a half years ago I announced I was leaving software to try to work on longtermist projects instead. A year ago I announced I was focusing specifically on research on the meta level. Today, I’m sad to announce that neither of these have worked out like I hoped; I’ve failed to find a way I’m confident has a large positive impact on the future. Three years seems like long enough to give this project, especially given that I don’t know what I’d try next if I wanted to. In light of that, I’m stepping back from big Save The World projects to focus on finances for a while.

What this means for the blog: I still like reading stuff and talking about it, so that will probably happen sometimes just like it did before I took this on ~full time. Maybe even more, now that the pressure to do the Most Impactful Research is gone. I don’t plan on putting more effort than is fun into refining knowledge bootstrapping or learning-and-trauma studies for the foreseeable future.

What this means for me: I’m now looking for employment that isn’t “the most impactful thing I can do”, and would appreciate any leads you can give me. My primary skills are as follows:

  • Research
  • Was once a programmer. I didn’t like it then but ironically the learning-and-trauma stuff is making it much easier to do and I may hit enjoying it again on a reasonable timeline.
  • Was once an SDET. I’m going to be even pickier about SDET work than dev work, but it is a skill I have.

The research thing is tricky. There’s not a lot of call for uncredentialed laymen who can read books about the history of childhood really thoroughly. But those skills are applicable to a wider range of problems, if people know to conceptualize the problems the right way and know that they have the option to turn them over to an expert. I spent one of the last three years at a company that *did* know how to use those skills, and was fantastically useful as I twitter-stalked investors, found the unwritten rules for getting on exchanges, and forecast the economy of Argentina. But of course I would claim that.  Instead of taking my word for it, here’s a description from my former boss at Reserve, a for-profit company where I worked as Head of Research for a year:

From managing Elizabeth at Reserve, I would strongly recommend her as a researcher to other companies, and even more so to startups, that need help making difficult highly complex decisions.

Early on at Reserve we faced many very complex challenges – e.g. navigating a labyrinth of legal issues, identifying product-market fit, and making major strategic decisions about how to fundraise. A major recurring bottleneck to make decisions was that we lacked information on various key unknowns, and so to find this critical information we tried out hiring Elizabeth as a researcher. Initially her scope of research at Reserve was pretty limited, but we quickly realized that her skill and autonomy meant she could do alot more for us and so we kept giving her larger and larger research tasks and she quickly became a pivotal member of the team.

Elizabeth has high agency at learning about an extremely wide range of topics and enjoys research work, and so she happily took on any research challenge that we gave her. She’s very competent – I remember we needed background info on someone, and she found a lawsuit against them hidden away on the internet. She will also research to any depth that you want – she can do 1 hours worth of research on something and then report back, or 5, or 10 hours. Best of all she’s highly autonomous and needs very little oversight, so once she knows what you need she can independently make decisions about what’s worth researching, and at what point is it best to stop and report back, which makes her especially well suited to a startup.

[Said manager didn’t want his contact info plastered all over the internet but it is available on request]

[If you’re wondering why I left a job I enjoyed and was good at and appreciated me that much, the answer is a combination of “changing business needs” and “crypto winter”.]

If there’s somebody you think this could be useful to, I’ve also included this in my Hire Me page and LinkedIn, so you can share it with them without the rest of this post.

Some other traits that make things a good fit for me:

  • Not make the world worse on net
  • Not be a bullshit job, in the Graeber sense of the word. One of the hardest things about the last three years has been the lack of feedback loops, and one of the things I’m most looking forward to is having concrete contact with reality.
  • < 25 hours/week on average (large week-to-week fluctuations are fine)
  • Working with other people who can give me feedback. One of the other hardest things about the last three years has been doing it more or less alone, and worrying about what I missed is just exhausting.

This would have been a really devastating announcement if I was making it any earlier than I am, but after my last grant rejection, I came to terms with it pretty rapidly. I tried something, I failed, and I’m pretty fortunate to have had both the option to try something unlucrative and a lucrative thing to return to. I don’t think my current orientation towards money and mental space is permanent- I just don’t currently have anything compelling enough to justify further burning of runway.

If you’d like to contact me directly, my e-mail address is

Thanks to everyone for all their support and encouragement over the years.


Epistemic Spot Check: This Isn’t Sparta


Despite not normally being a fan of military history, I’ve really enjoyed Bret Devereaux’s blog Acoup, in which he uses pop culture representations as a starting point to teach the subject. Unfortunately I got stuck in a loop where I didn’t want to read Acoup without fact checking at least one post, and I didn’t want to fact check a post because military history is boring when taught by anyone besides him and Dan Carlin (and I was really disappointed in Carlin’s book. Why, you might ask? Yes, I too wish I’d written that down).  

But I really wanted to start reading Acoup again. My compromise is to:

  1. Use a sequence that involved a lot of non-combat facts (as good military history usually does, and by “good” I mean “enjoyable for me to read”)
  2. Limit myself to some pretty bare bones fact checking. This does not protect me against sophisticated forgeries, but is surprisingly good at catching people with a sincerely believed agenda.

And now, a good old fashioned epistemic spot check of Bret Devereaux’s This Is Not Sparta.


Devereaux is not so much sharing historical facts as comprehensive models, informed by both local facts about the specific area of interest and general knowledge of humans and historical trends. This is great and I wish more people would do it. Along the way he sometimes presents facts as more certain/less controversial among historians than they are. He’s not hiding that he does this, but on first read through I did walk away with the impression that certain things were more settled than they are, and unprepared to argue for their truth.

E.g., Acoup describes helots not only as slaves, but as slaves that were especially poorly treated even by the standards of the time. 

First, let us dispense with the argument, sometimes offered, that the helots were more like medieval serfs than slaves as we understand the ideas and thus not really slaves – this is nonsense. Helots seem to have been able to own moveable property (money, clothing etc), but in fact this is true of many ancient slaves, including Roman ones (the Roman’s called this quasi-property peculium, which also applied to the property of children and even many women who were under the legal power (potestas) of another). Owning small amounts of moveable property was not rare among ancient non-free individuals (or, for that matter, other forms of slavery).

As noted below, this is not the consensus view. The writing leaves me with a good sense of why Devereaux believes what he believes, but not prepared to teach the controversy. I think if I started an argument with a helots-are-serfs partisan based purely on this blog post, I would look stupid and unprepared. Which is fine. The goal of the post is not to help me look smart at parties with classicists, it’s to leave me with better models of militant societies as a group. This is a restatement of the truism that if you really care about something you should probably read more than one source.

A broader example is that Devereaux fills in the paucity of written records about helot life with patterns known from other slave societies.  E.g. there isn’t actually a written reference to Spartiate men (the Spartan nobles) raping helot women. But we can take a look around at better documented slave societies, and at the legal code around the existence of children with Spartiate fathers and helot mothers, and make some educated guesses. This is a good and reasonable thing to do if what you want to do is picture ancient Sparta more accurately, but should not be double counted as evidence for the trends and patterns used to fill in the gaps.

I feel dumb ending with the conclusion “this is good but also never rely on one source.” You knew that already, and so did I, and yet doing this spot check made me more cautious in relying on Acoup than I previously was. Humans, man.

The Actual Spot Check

Full notes are available in my Roam graph

All claims are taken from posts 1 and 2 in the Sparta series. Claims were selected for being easy to verify and do not in the slightest constitute a random sample.

Claim: “Sparta was not a city-state for the simple reason that it didn’t have a city – it had five villages instead”
Verdict: Plausible, subject to arguments on definitions. It’s easy to find tertiary sources describing Sparta as a city, but none saying “I have considered the possibility that Sparta was in fact five villages in a trenchcoat and rejected it for these reasons”. I couldn’t actually find any references to Sparta being five villages- sources that described it as an amalgamation always said four. Bu when I reached out to Mr. Devereaux’s Twitter he explained that he was counting a fifth village that had joined later (he also acknowledged there was some controversy in the description, which he didn’t in the original post). I still wish this had come with exact definitions of what constitutes a city vs a village, and what the implications of the difference are.

Claim: “Spartan boys were, at age seven, removed from their families and  instead grouped into herds (agelai) under the supervision of a single adult male Spartan”… “While the heirs of Sparta’s two hereditary kings were exempt from the agoge – perhaps because the state couldn’t afford to risk their lives so callously – Leonidas was a younger brother and thus was not exempt”
Verdict: confirmed in secondary source (although Devereaux goes to pains to point out the biases of the secondary sources)

Claim: “The boys were intentionally underfed. They were thus encouraged to steal in order to make up the difference, but severely beaten if caught”
Verdict: confirmed in secondary source

Claim: “Not even the exemplary boys escaped the violence, since the Spartan youths were annually whipped at the Altar of Artemis Orthia”
Verdict: confirmed in tertiary source

Claim: “Then there is the issue of relationships. At age **twelve** (Plut. Lyc. 17.1) boys in the agoge would enter a relationship with an older man – Plutarch’s language is quite clear that this is a sexual relationship (note also Aelian VH. 3.10, similarly blunt).”
Verdict: Well supported inference, not actually proven. My literal reading of the sources cited leaves a lot of ambiguity over whether the relationships were sexual or not. This could be a lack of experience reading between the lines of ancient Greek sources, but other historians who have presumably read them also dispute the claim. Given what we know of violent all-male institutions in general I think it’s an extremely reasonable inference that the relationships were sexual (and by our standards, extremely coercive at a minimum), in fact it would be surprising if they weren’t, but this isn’t a smoking gun.

Claim: “These Spartan boys will have to apply to be part of a mess-group (syssitia – a concept we’ll return to later) when they are twenty”
Verdict: confirmed in tertiary source.

Claim: Some boys in the agoge were selected for the krypteia, which patrolled farms at night to murder slaves.
Verdict: confirmed in tertiary source.

Claim: “we actually know that individual Spartans painted their shields with a variety of individual devices.”
Verdict: Multiple tertiary sources report this being true at the time 300 would have taken place, but that the army kit was eventually standardized.

Claim: “While there were supposedly 8,000 male spartiates in 480 there seem to have only been 3,500 by 418  just 2,500 in 394 and just 1,500 in 371.”
Verdict: tertiary sources gave slightly different numbers at slightly different times but the trend was confirmed repeatedly.

Claim: The perioikoi were poor farmers on marginal land on the outskirts of sparta. They were free save for mandatory participation in the army, but had no say in government.
Verdict: gist confirmed in tertiary sources.

Claim: “The hypomeiones seem to consist of the men (and their descendants) who had been spartiates, but had been stripped of citizen status for some reason, usually poverty (but sometimes cowardice)”
Verdict: gist confirmed in multiple tertiary sources, sometimes using a narrower definition

Claim: “The mothakes (singular: mothax) seem to have been the bastard off-spring of spartiate men and helot women”
Verdict: gist confirmed in tertiary source, although they seemed to consider hypomeiones a subset of mothakes.

Claim: Neodamodes were freed spartan slaves
Verdict: gist confirmed in tertiary sources, although they refer only to slaves freed for military service

Claim: Sparta’s population distribution was roughly

Verdict: agrees with cited source.

Claim: Helots were owned by the Spartan state, who assigned them to work land owned by the Spartiates (method of assignment unknown)
Verdict: confirmed by tertiary source.

Claim:  Helots were poorly treated slaves, not serfs or a stage between slave and free
Verdict: certainly seems justified, but officially controversial

Thanks to my Patreon patrons for helping to fund this work.

Update 11/25: Bret Devereaux has a few comments about this on Twitter.

Update 11/27: A friend checks the Spartan military record and finds it meh.

Looking For Test Subjects: Breaking Questions Down

As long time readers are well aware, my main focus for the last year has been developing a method for Knowledge Bootstrapping– going from 0 to 1 in an unfamiliar field without undue deference to credentialism. I’m at the stage where I have a system that works well for me, and I’ve gotten feedback from a few other people about what works and doesn’t work for them, but there’s a long way to go. A lot of my knowledge is implicit and not explained on the page, plus I am only one person; what works for me will not translate perfectly for every human. So I’m looking for test subjects.

One particular part of my method is breaking down one large question into many smaller questions. This has several purposes: it forces you to clarify what you actually care about, and makes it more obvious what information is relevant. I describe this process and the reasoning behind it here, but not very well.  I’m looking for test subjects that have a research question, and would like to practice breaking it down into smaller questions, with the goal of refining the technique and my teaching of it.

What This Looks Like

  1. Come up with a question you might like to research.
  2. You book a phone call with me via calendly, or email me at elizabeth -at- to set up a time. 
  3. We discuss your question in an attempt to break it down into smaller parts.
  4. I sure hope some people actually go off and research the new questions but there’s no commitment required to do so.

What Are the Expected Outcomes?

  • You will have a better understanding of what you actually want to know and will be better positioned to find answers.
  • You will be better able to break down your next research question, without me.
  • I will make some of my metis on breaking down questions more explicit.
  • I will become better at teaching the technique of breaking down questions.
  • I learn techniques from you I couldn’t have learned on my own.

No Notes Research

I started watching The Vow with a friend, and got inspired to do a bunch of reading on cults in general and Vow’s cult, NXIVM, in particular. I didn’t originally take notes because this was coded in my head as a leisure activity, not real research. Eventually it became clear it was a real research project, but it seemed unfair to introduce real notes halfway through, so I decided to use it as an experiment in research without detailed notes instead (I did end up writing a few, but a far cry from The Algorithm). This turned out to be the right situation for that experiment, because my friend was a check on how much I actually remembered, especially on things we disagreed on, which was a lot.


  • Memory is in fact hard. 
    • When I went to share what I learned with my friend, I often had to look back at my (sparse) notes to remember things I wanted to talk to him about. This is true even when I was talking to him the morning of the day after I read the book. 
    • Often he would ask me questions and the answer wasn’t in my notes- sometimes it was firmly in memory and I’d just forgotten to bring it up, sometimes I knew the book had the answer but I had lost it.
  • I mixed up sources a lot.
    • I would frequently assume my friend knew something, only to find out it wasn’t from our shared source. 
  • I’d compress specifics into patterns. This hindered me arguing with my friend.
  • In combination with the source mix-up, this often meant I couldn’t tell apart the following situations:
    • Books A and B repeating the same story with the same source (almost equivalent to one source)
    • Books A and B tell the same story from their own perspectives (stronger evidence the thing actually happened, but not evidence of a pattern)
    • Books A and B tell stories about similar things happening to different people (evidence of a pattern).
  • Losing the specifics that demonstrated a pattern also made it much harder to change my mind in response to new evidence. Is this more credible or a stronger signal than the data my current view is based on? Who’s to say, if I can’t remember the original evidence?
  • This feels way easier with something as emotionally salient as cults than it did with my more distant historical research. And when I did shift to a more history-style book (Mystics and Messiahs), I suddenly had to take real notes.
  • I did go into this with a question, but I didn’t know what it was until I’d read a few books and seen what felt live and what didn’t. 
    • The question was: how do we cultivate instincts/responses that seamlessly antagonize the unhealthy parts of cults while allowing for communities and new ideas.
    • Also: Interpersonal power: How does it work?

I feel like there’s more to learn from this experiment, but I really needed to write *something* today and a draft post I had on a beautiful theory of mine took a hit from an ugly gang of facts, so this is it.

Coping with Cats

I love my two cats dearly and they bring a lot of joy to my life, but they also bring other things, like poop, and fur.  When I tried to research mitigations for any of these it was 100% SEO listicles, so I’m sharing my raw data in the hopes that the next person can find something real-if-incomplete instead of a list of every automated litter box that has ever been manufactured.

I wrote the bulk of this in February, but then one of the cats developed dermatitis and I put it on pause until I could make sure it wasn’t caused by anything on this list. Alas, it wasn’t, she’s just allergic to my new apartment. But then I wanted to test something else, and suddenly six months had passed. So if the tenses seem weird, that’s why.

(Links are affiliate links when possible)


I screwed up choosing my new apartment (this was written before the allergy thing). I remembered to check for space to put the litter box in the bathroom, but in my joy at finding a place that allowed that at all after so many that did not,  I forgot to check if that space was also where you stood to wash your hands. So I ended up having to put the box in the living room. I’ve done that before and didn’t want to go back to having to keep the window open year round and still smelling poop. I also didn’t love the aesthetic.  So I set up something kind of elaborate.

KittyVent Litter Box Ventilator

This is a long piece of hosing with a gentle fan in it, and a plastic thingy to block off a window that has a hole for the hose to connect to (similar to what you’d use for a portable air conditioner)

The window-blocking thing works vertically but won’t stay up on its own, I had to tape it. The fan is extremely quiet, you can’t tell it’s on from more than six inches away. I’m very sure it works because the first time I noticed a smell, I found the fan had been turned off.

There are instructions for how to DIY this available online (e.g. here), but I just bought it and it worked fine.

Litterbox Holding End Table

To make it less hideous, I got an end table designed to hold a litter box. Amazon is full of these, but most have only a single entrance and no ventilation. I assume these are aimed at people who haven’t thought things through. I wanted something I could connect a vent to and actually get rid of the smell. You might be able to cut a hole in one of the ventless ones to attach the hose to but I didn’t feel like bothering, so I got one with ventilation.

I chose ecoFlex (affiliate link)  because matched my it existing furniture and was the only option with both ventilation holes and a second shelf. For this, I accepted it being really much larger than it needed to be. Also the copytext said something about liquid-proof, non-smell-absorbant material, although who knows how true that is. Then I packing taped the ventilation hose to a hole in the back that wasn’t shown in the pictures, apparently intended for electric cords to automated litter boxes.

My cats adapted stupidly well to this- when I set up the litter box with the door open, they both used it within 10 minutes, despite their other litter box still being available. They did just as well when I closed the door.

Some litter does escape the table, so have a plan for that.

Dog Poop Bags

My weird conservation instincts push me not to empty the litter box until I can fill a bag. Otherwise I’m wasting valuable bag. So I bought a bunch of very small, very cheap bags and stored them in that shelf in the litter box end table, which is why I wanted it so much. I chose Amazon Basics Dog Poop Bags (affiliate link) based on WireCutter’s recommendation and so far they are resoundingly fine.

Dr Esley Cat Litter (affiliate link)

I got this because Sweethome recommended it. I didn’t appreciate how much better it was working until a delivery screw up made me use Arm and Hammer and there started to be a smell where there hadn’t been before.

Smell- Things That Didn’t Work

Litter Robot 3

This is so close to being a truly great product, but isn’t quite reliable enough to be the only litter box. It had a tendency to get stuck in an unusable position, which eventually forces the cats to go elsewhere. The drawer the poop stays in doesn’t completely contain the smell. Also the range of litter between “excess will get dumped” and “too low” is uncomfortably thin. I’d still recommend it as a second litter box to cut down on scooping, or in a populous household where there will always be someone to turn it off and on again.

Two Layer Litter Mat (affiliate link)

This actually worked great at its stated purpose of catching litter the cats track out of the box, but when the cats pooped on it (say, because the litter box had been upside down for six hours), it was almost impossible to clean, especially if I caught it less than instantly. Plus it is ugly.

Open Window with a Fan

This was definitely better than not having an open window with a fan, but there’s no comparison between it and the vent-with-hose.

Cat Hair

Silicone Grooming Gloves (affiliate link)

These collect a lot of hair, plus their squishiness means you can brush bony areas that would be a problem for a metal comb. The cats start purring the instant I pet them with these.

Downside: kind of a pain to get the hair off the glove, does not pick up 100% of the fur it loosens, so be sure to use it somewhere easy to vacuum or at least not your own lap.

The Dryer

Turns out dryers are reasonably good at removing cat hair if you wait long enough. Remember to clean the lint trap frequently.

Reusable Lint Roller

I tried two made of the same material- a paddle shaped one (affiliate link)

and a more traditional roller shaped one (affiliate link)

Of these, I found the wand shaped one easier to use, and the roller easier to clean. I prioritized use, but your mileage may vary. Also, you need to actually read the instructions on how to use it, running it back and forth like disposable lint roller just redistributes the hair.

Here is a picture of a pillow after I ran the paddle shaped one over part of it several times. I think this gives a good idea of what it’s capable of, i.e. it’s a significant improvement but doesn’t get everything.

Note: inbetween writing this and publishing, the paddle brush people offered me actual money to publish a review, which I plan on taking them up on, assuming I can find the form.

Shark Pet-Perfect II Handheld Vacuum (affiliate link)

Sweethome’s top handheld vacuum pick was sold out everywhere and their second is specifically called out as being bad at fur, so I selected based on Amazon reviews. As a vacuum cleaner it is Not Fucking Around- it is super powerful and sucks anything off my floors. It advertises a special whirling brush aimed specifically at pet hair, but I found it choked on soft surface I tried it on, and the bristles were better. The biggest complain on Amazon was the battery life, but I got it for spot cleaning not marathons, so it’s been fine and I appreciate not paying for a feature I don’t use.

All of the anti-cat-hair devices work better if I stay very on top of them. Given time, the fur works itself deeper into the fabric and nothing will work except picking it out with actual human fingers.

Cat Hair- Mixed Results and Stuff that Didn’t Work

The Furminator (affiliate link)

This is another one that was great at its official objective but failed a secret objective. It was absolutely amazing at removing fur from the cats (on areas broad and unbony enough to support the comb), and they liked it, but it left them less soft. I did not get these cats to have them at less than their maximum softness.

Rubber Thingies You Throw in the Washer and Dryer 

I ran an experiment and as far as I can tell these are no better than just running an article through the dryer. Before/after photos (which I unfortunately didn’t save) didn’t show more hair being removed and the thingies are completely free of hair after many uses, which suggests to me they’re not helping much.

I tried wetting them and running them over a pillow case and they indeed picked up something, so I’m leaving them in my washer just in case.

I used LEADTEAM (affiliate link) and didn’t test any others, they looked identical.

Wool Dryer Balls (affiliate link)

Smart Sheep Reusable Wool Dryer Balls, 3-Pack

Testing the dryer inserts is difficult because it’s really hard to get a control group. I can’t put in half a blanket and see what the other half does. I can’t even use my two identical blankets, because they’re not guaranteed to have identical amounts of hair on them. I tried with socks but was kind of out of patience by that point. I was pretty confident the rubber ones were doing nothing, but despite a really silly amount of rigor involving photos and test swabs with the lint roller, I just can’t tell with these.

Disposable Lint Rollers

These were better than nothing but not as good as the reusable roller.


My cats were never great at eating but had gotten to the point where between the two of them someone was vomiting almost every day without cause (cause = “I went too long without eating”, “I ate too fast because I went too long without eating” and hairballs), and one had lost a noticeable amount of weight (her current weight would be fine if she’d started there, but losing almost 20% of her body weight for no known reason is concerning). There was also behavior that made me think they didn’t like their food- things like persistently begging and then when I put new food down, sniffing it, looking upset and begging more. So I talked to the vet.

Prescription Food and/or B12 vitamins (affiliate link)

I started both of these at the same time and now the cats are down to vomiting once a week, tops, some of which is for cause. They don’t love the prescription food by they will eat it. They’re fine with the vitamins.

In terms of specific brands:

  • They hate Royal Canin Duck and Pea protein. When I included this in their diet they ate only two cans of food per day (one Royal Canin, one of their old food, Friskies), which is little enough that the already-underweight one was losing more weight. Also it is even more expensive than regular prescription food. I tried it because it’s supposed to be hypoallergenic, but it turns out the cat’s problem was not food allergies.
  • They will tolerate Hills Digestive Care, although they eat it slower than the Friskies (but I think they eat the Friskies faster now that they’re eating Hills at all).
  • Purina Pro Plan Veterinary Diets EN Gastroenteric Formula is perhaps liked better than the Hills.

In terms of where to buy:

  • There’s a rebel pet store in Oakland that sells prescription pet food without a prescription (thanks federal government, for taking time out of your busy schedule to prevent me from paying more money for food my cats like less without doctor’s approval), but not any cheaper than stores that do require a prescription.
  • For delivery I use, which is set up to track prescriptions so you don’t abuse that sweet, sweet easy-digestion cat food. Be warned they do not warn you when they’re about to send an autoship order, so you need to stay on top of that yourself.

Small Radius Carpet Cleaner (affiliate link)

This carpet cleaner deep cleans the 9 inch (?) circle you put it down on. This is the perfect size for cat vomit. I only used this for 9 months until I moved somewhere with hardwood, but I did get the security deposit back on the carpeted place.

Downsides: noisy; downstairs neighbor will complain if you use it too late.

Couch Cover

Between the vomiting and the hair, it was really essential my couch have a removable, washable, dryer-able cover. I got a couch that came with removable, replaceable covers, but

  1. the covers are almost 1/3 the cost of the couch
  2. they were extraordinarily difficult to get off and on

At which point the covers might as well be attached to the cushions after all. I selected a more universal couch cover from amazon with the qualifications that

  1. It handle the fact that my back cushions are removable and sit on top of the seat cushions
  2. It be able to go in the dryer, which is critical to the “don’t live life covered in cat hair” plan.

That left me a single choice, the Chun-Yi one piece (affiliate link).  Elsewhere in this article I’ve stolen Amazon’s pictures to showcase an object: in this case I’m deliberately not doing so, because this cover does not look much like the picture. It fits fine, but not nearly as crispy as implied, and the color is off.

Having a loose cover like this is also a pain in the ass for spot-hair-removals with the vacuum cleaner, because as mentioned above the vacuum cleaner’s fancy twirling brush just sucks the cover off the couch. The with bristles works well enough but still requires two hands to do well with.

Predictions As A Substitute For Reviews

Update: You can see how this worked out after 6 months here. Short version: it did not escape the weekly review trap.


There is a lot of praise among my friends for weekly reviews. There is noticeably less doing of weekly reviews, even among the people doing the praising. It’s a very hard habit to keep up. We could spend a lot of time diving into why that is and how to fix it in a systematic way… or I could tell you how I use PredictionBook to get many of the promised benefits of weekly reviews without any willpower.

In a nutshell, when I find myself making a choice about how to spend time or money that’s dependent on some expectation, I write out the expectation as a prediction (with % likelihood) in PredictionBook, which then automatically prompts me to evaluate the prediction at a date I set. This has so many benefits I’m struggling to figure out where to start. If I had to sum them up in a phrase, it would be “more contact with reality”. But to expand on that…

What contact with reality may look like


Making the prediction forces me to assess my anticipated outcomes and do an expected value calculation. This in turn forces me to be explicit about what I value, and how I will know if I got it. It also makes it explicit when I’m doing something because I expect the median or modal outcome to be good, vs. when I’m doing it for the long tail of unlikely but super good outcomes. It also gives me a chance to say “huh, that EV is not competitive” and do something else.

Evaluating the prompt gives me more information on how my plans are working out. Andy Matuschak talks about the dilemma of knowledge work, where you can’t make a plan for success but leaving things open ended often leads nowhere. His solution to this is to give himself unstructured time, but then look back and see if it worked out the way he wanted, and if it didn’t, do something different next time. Predictions provide a really natural, lightweight prompt for this reflection.

Sometimes I withdraw a question for being poorly phrased- often when my answer makes the decision sound “bad” but I feel like it was actually “good”, or vice versa. This is an easy way to notice I was wrong about what I valued and look for what is actually driving my decisions.

Then there are the framing benefits. It’s easier to view something as an experiment if you’re explicitly writing down “5% chance of success”, which makes failure feel less bad. I didn’t fail to make something work, I executed the correct algorithm and it failed to produce results this particular time.

Making an explicit prediction and writing it down closes open loops in my head, freeing RAM for other things.

It’s helpful to notice when the habit doesn’t trigger. For example, I have a meeting planned today (when I’m writing this) that I just “didn’t feel inspired” to make a prediction for. When I looked closer I realized that’s because I was kind of eh on this meeting but felt obliged to take it, and was avoiding that fact. I don’t know how this is going to work out because I’m choosing to work on this blog post before tackling that, but I still credit the prediction habit with noticing, which is the first step towards better choices.

Also it seems like this might make my predictions more accurate, which has all kinds of applications. But to be honest that’s not what’s reinforcing this.


So we’re on the same page, here are some sample predictions, and the decisions that rested on them:

  • Reading Pavlov and his School will lead to behavior change in the area of learning or sleep (20%)

    • Should I buy a physical copy of Pavlov and his School and spend the time to read it?

  • Talking with my friend Jane on 8/1 will be energizing (95%)

    • Should I take a call with Jane?

  • Bob the recruiter will describe a job that I am at all capable of and interested in doing (3%)

    • Should I take a call with Bob?

  • I will judge the seminar on 7/28 will be worth the interruption of flow (40%)

    • Should I spend money and time on this seminar?

  • Sam will finish task Y by 7/31 (20%)

    • No immediate decision riding on this one, but it sure seems useful to calibrate on how well I can predict a project partner’s productivity.

  • I will play with an Oculus Quest at least an hour a week (98%)

    • Should I buy a Quest?

  • California will catch fire to the point I need to keep my windows closed (95%)

    • Should I buy an air conditioner?

  • Company Z will offer me at least an hour of work (15%)

    • No immediate decision

  • Supplement S will improve my sleep to the point I don’t need a nap (3%)

    • Should I buy and use supplement S?


If I were starting from scratch, here are the instructions I would want to receive (many of which I did receive, from Raemon)

  1. Create an account on

  2. Go to Settings to correct your time zone and set your prediction default to “visible to creator” rather than “visible to public” (unless you’d like them to be public by default).

  3. Create a handful of predictions that will resolve over the next week (PredictionBook will e-mail you when they should be resolved).

    1. The goal is twofold:

      1. Quickly get feedback about how this feels for you.

      2. Get yourself in the habit of making and resolving predictions.

    2. When making predictions, try to hone in on things that are decision-relevant to you; that you would do something different based on if the prediction was true or not.

  4. Create a link to in your browser so you can access it easily.

  5. As you take more actions based on predictions (implicit or explicit), notice that you are doing so, and register the predictions in PredictionBook

  6. Resolve predictions as you are prompted to do so.

  7. After a week or two, check in with yourself and how you feel about the project. For calibration: I found creating the initial predictions kind of a pain but by a week in the project was naturally rewarding and required no will power on my part.

This should take < 30 minutes, and it should take < 45 minutes total to get a good sense of whether this is working for you.

Why Does This Work?

I’ve been thinking a lot about flow and distraction recently. One thing I’ve noticed is that it takes an awful long time, measured in hours, to get into the mindset for certain tasks. Those tasks then feel amazing, unless I’m pulled out of them prematurely, which hurts (and makes it take even longer to get into the next time). There’s a lot of implications of this that I’ll hopefully get to in some other post; the relevance here is that I think weekly reviews might be one of those tasks that requires a lot of time to get into the required headspace and hurts to leave prematurely. This makes them costly- much more costly then I would have estimated before I learned to bill tasks for their prep time. It’s also something of an all or nothing task- doing 50% of your weekly review does not get you nearly 50% of the benefits of a full weekly review.

But predictions, both making and evaluating, integrate into my life pretty naturally most of the time, and scale gracefully. Writing them up can take as little as 15 seconds, and when it takes longer, it’s because I’ve discovered something important I need to work out. Doing this fits in naturally with the process of making plans, so I don’t need to spend time getting into the right head space. Evaluating them is usually trivial- and if it’s not, it’s highlighting a problem with my models.

Tips and Tricks

“I will enjoy X” is very rarely the right prediction for me to make, because enjoyment is a tricky thing for me. I don’t like admitting I didn’t enjoy something, and sometimes I will burn a lot of energy trying to make something enjoyable, which can be a bad decision even if it works. My equivalent is “X will be energizing”

When trying to create predictions, think about what you’ll be spending money and time on in the next week. How do you think those things are going to go? What outcomes would make you change your decision?


My original inspiration for this experiment, Raemon, hasn’t gotten the benefits I describe. He’s more focused on improving his prediction calibration and accuracy, and so far hasn’t done as much decision-relevant predictions. I’ve encouraged him to try it my way and hopefully he’ll comment after he’s spent a few weeks on that.

PredictionBook is pleasantly lightweight but it’s already a little cluttered with predictions. If I really do solidify this habit I may need to find or create a better system.

I’ve been doing this for about a month. People have kept up weekly reviews for longer before letting them fall by the wayside. I’m writing this up now because the act of solidifying a habit makes me worse at writing about it. I’ve set a reminder for 12/1 to write an update.

Thanks to Raemon for pushing me to a better version this post and for initiating the experiment in the first place.