See also: [[_02022 Amble 5 – plan]] *5 months from July to November.* My original plan for this period was: a week on [Comment Helper](https://comment-helper), then 2-3 weeks exploring project ideas for 2023 (c.f. [[Next big project project]]). I extended this period from 4 weeks to 8, then 8 weeks to 12, then 12 weeks to 18, then 18 weeks to 22. My "exploration of project ideas" took on a lot of unexpected traction, and then [FTX happened](https://photos.app.goo.gl/P1S8fk9eJLsHQbac7). Before I get into all that: the most important news. In July, I asked Chloé to marry me, and she said "yes". 5 months later, there's not been a day where I've doubted this decision. I did not believe this was possible. ## Highlights - Did I mention that Chloé agreed to marry me? - I began exploring project ideas for 2023. I began by writing out a project sheet ([[Next big project project]]), tidying up my [[_Next big project longlist]], updating my CV and reflecting on that. - I noticed several small projects, on the order of 1-4 weeks FTE work, which I have long wanted to try. I decided to explore a couple of them, with a view to testing some working relationships, getting some positive attention within EA, practicing some relevant skills, and, of course, doing something reasonably useful. - The first project was: get more people to read more academic papers by Nick Bostrom. I brainstormed about 20 ways to do this, and went with [Radio Bostrom](https://radiobostrom). Highlights: - I raised funding and hired a team within 2 weeks. - We narrated 27 papers in 70 days, roughly one every three days. - We released [An Introduction to Nick Bostrom](http://radiobostrom.com/introduction), to complement the main feed. - Nick's personal web site (nickbostrom.com) now links to our episodes. - Buzzsprout reports >10,000 episode downloads, and we didn't start the marketing campaign yet. - A BBC TV & Radio star has agreed to narrate a couple of "special episodes" to help with marketing. - Stakeholders were delighted and approved additional funding so that we can be more ambitious on marketing. - Next steps: another round of marketing, a brief impact evaluation. - [Parfit Archive](https://parfitarchive.org): I listened to Tyler Cowen interviewing Will MacAskill, and yet again, I found myself wanting to read the correspondence between [[=Derek Parfit]] and [[Bernard Williams]]. With Tyler's encouragement, I wrote to Parfit's widow, Janet Radcliffe-Richards. Four days later, I had met Jeff McMahan (literary executor) and raised funds to support the creation of the Parfit Archive. Summary: - I will take care of all the non-academic work. - The physical papers will be scanned by the British Library Digitisation Team during December. - Plan for the hard drives is to be confirmed. - First version of the digital achive will be available online to selected scholars by early 2023. - I am looking forward to (finally!) having a look at the correspondence between these two great philosophers. - I met [David Edmunds](https://twitter.com/davidedmonds100), co-creator of Philosophy Bites and author of Wittgenstein's Poker. We got on rather well and had a couple of great conversations about Parfit. I also setup a visit to the 80,000 Hours office, where he had 1-1s key staff, and also me. David has a Parfit biography coming out in 2023—I'll help a bit with the promo. - Writing: my final 1-4 week project was: write up more thoughts from project 32, and try a few months of being more active on [Twitter](https://twitter.com/peterhartree) and the [Effective Altruism Forum](https://forum.effectivealtruism.org/users/peterhartree). Again, partly for its own sake, partly to attract collaborators. This went really well! Highlights: - Twitter: [My best tweets](https://tweets.pjh.is) are quite good. My follower count grew from ~400 to >1K, and my tweets received more than 1 million impressions. - A [thread about Peter Thiel](https://twitter.com/peterhartree/status/1594429794044694528) went viral, but I fear that most people who read it probably ended up with worse beliefs. This was part of my multi-year mini-campaign to get some important figures to [stop underrating Thiel](https://twitter.com/peterhartree/status/1583509982099607552). - EA Forum: [my least popular post so far, and also my best](https://forum.effectivealtruism.org/posts/G6EWTrArPDf74sr3S/bernard-williams-ethics-and-the-limits-of-impartiality) Also, [my most popular post so far](https://forum.effectivealtruism.org/posts/av7MiEhi983SjoXTe/some-core-assumptions-of-effective-altruism-according-to-me), which was featured in Marginal Revolution ["Assorted Links"](https://marginalrevolution.com/marginalrevolution/2022/07/saturday-assorted-links-369.html). My karma increased from ~400 to around 2K. - LessWrong: [Wanted: notation to express credal resillience](https://www.lesswrong.com/posts/uGqdJCrqznzLBDXcr/wanted-notation-for-credal-resilience). - [TYPE III AUDIO](https://type3.audio): Radio Bostrom led to a lot of people asking me to arrange the narration of other writing relevant to the effective altruism and longtermism community. - I reflected a bunch, spoke to about 20 people, and decided to give it a shot. - TYPE III AUDIO launched on October 1st, with a plan to release at least 500 hours of narration within a year. - Prior to launch, we made a $300K first sale, to the Effective Altruism Infrastructure Fund. - 8 weeks in, demand is overwhelming and [we've done quite a lot](https://blog.type3,audio/hello-listeners/). - And a fantastic team is emerging (currently ~5 FTE). We've also hired [And-Now](https://and-now.co.uk) for branding and visual identity (and strategy advice by stealth, as the best agencies usually do). - It turns out that AI narration is better than I realised, so we're now on track to release thousands of hours of AI narration in year 1, as well as ~500 hours of human narration. - My vision for this project is evolving quickly—to include original content and a substantial curation layer, to complement the core narration service. I think it could be big. - I continued with my advisory role at [80,000 Hours](https://80000hours.org). - I ramped up mentoring efforts with notable successes, that I should keep private. - I identified Conor Barnes as candidate job board hire, then helped get him hired. - I decided not to revisit idea for another full time role there until at least winter 2022. - I joined the team retreat in early October. - I made big progress on persuading some key figures to start a rebranding project. We have an initial meeting scheduled with And-Now. - I asked for a huge pay rise on the morning of Tuesday November 8th. The FTX blow up became clear in the afternoon. - I flew to Oxford at short notice to help CEA and 80K with the FTX response. I did some valuable work and learnt some very important lessons, but didn't have as much influence on the overall strategy as I hoped. I expect to write more on this, for now see: [Samo on FTX & EA](https://forum.effectivealtruism.org/posts/ctiLEKuC23oTzsDtm/samo-burja-what-the-collapse-of-ftx-means-for-effective), and [Byrne and Dwarkesh](https://forum.effectivealtruism.org/posts/YFYgtZzfwjW7cEnkf/byrne-hobart-and-dwarkesh-patel-on-hardcore-believers). - Funding helps: I suggested and helped Joe Walker, Andy Matuschak and Perrin Walker write succesful funding applications to speed up their work. I called with Gus Docker, who I helped get a grant last summer. He just started work as podcast host for FLI, thanking me with the words "it definitely wouldn't have happened without you". I'm very happy about this, he has already released some great interviews. I helped one other person write an unsuccessful application, and sent them a $5K micro-grant from my personal donations pot. - Networking: several Oxford visits, many 1-1s with new contacts. - [Comment Helper](https://comment-helper): released an update, got 5 paying customers ($150 MRR) and a bunch of feedback. Hired a contractor and we have another major update nearly ready for release. But I deprioritised this against other projects, so while I'm happy with progress, we several months behind the roadmap. - [Inbox When Ready](https://inboxwhenready.org): I hired a marketing firm to conceive and run a B2B sales campaign. I invested a day of my time and $7.5K, and we achieved a spectacular $0 return. I'm glad I tried the experiment, but I should have screened the agency harder and made the initial commitment smaller. (They turned out to be fairly incompetent.) - Personal: moved from Penne d'Agenais to Clermont-Ferrand for two months. Hosted another Bridge Club Retreat and had a great summer. Won my second go karting trophy (endurance race). Two week holiday with Chloé and family in Cornwall. And then we moved to Reykjavík, for my regular autumn visit. Chloé landed a wonderful job, so we extended our stay until end February. Overall I rate this period: 12/10. I've not felt this sharp, and consistently high energy, since autumn 2010. Not sure, but I think it's probably the new medication I started in spring, plus Chloé. My biggest concern is that I drifted into a big time commitment to TYPE III AUDIO before I seriously explored a bunch of other promising project ideas on my longlist. Noticing this, I resolved to spend at least 1/4 of my time exploring other things between October-December, then review plans for 2023 in January. I am still aiming for this, but in practice the FTX disaster mostly derailed this. There's also small chance the Parfit Archive spirals into a much more time consuming project than I expect (academics!), but I'm not too worried for now. In the worst case, I can make a graceful exit. What's next? In December & January, TYPE III AUDIO will be the main focus. I'll also do a _bit_ of exploring of other candidates for main focus in 2023. But I'm now at >2/3 that TYPE III AUDIO will be the Thing. C.f. [[02022 Amble 6 – Plan]].