diff --git a/.gitignore b/.gitignore
index 2c33734..f45b9e4 100644
--- a/.gitignore
+++ b/.gitignore
@@ -1,3 +1,5 @@
.secret/
sitemap.gz
sitemap.txt
+.wip/*
+blug.md
diff --git a/blog/2022/04/07/openphil-allocation/index.md b/blog/2022/04/07/openphil-allocation/index.md
index f759cff..f3e991d 100644
--- a/blog/2022/04/07/openphil-allocation/index.md
+++ b/blog/2022/04/07/openphil-allocation/index.md
@@ -1,4 +1,4 @@
-Open Philanthopy's allocation by cause area
+Open Philanthopy allocation by cause area
===========================================
Open Philanthropy's grants so far, roughly:
diff --git a/blog/2022/12/20/hacking-on-rose/.src/draft/.index.md.swp b/blog/2022/12/20/hacking-on-rose/.src/draft/.index.md.swp
new file mode 100644
index 0000000..b755abc
Binary files /dev/null and b/blog/2022/12/20/hacking-on-rose/.src/draft/.index.md.swp differ
diff --git a/blog/2022/12/20/hacking-on-rose/.src/draft/.src/libre_redirect.png b/blog/2022/12/20/hacking-on-rose/.src/draft/.src/libre_redirect.png
new file mode 100644
index 0000000..610cf89
Binary files /dev/null and b/blog/2022/12/20/hacking-on-rose/.src/draft/.src/libre_redirect.png differ
diff --git a/blog/2022/12/20/hacking-on-rose/.src/draft/.src/matt-levine-clean.png b/blog/2022/12/20/hacking-on-rose/.src/draft/.src/matt-levine-clean.png
new file mode 100644
index 0000000..730208c
Binary files /dev/null and b/blog/2022/12/20/hacking-on-rose/.src/draft/.src/matt-levine-clean.png differ
diff --git a/blog/2022/12/20/hacking-on-rose/.src/draft/.src/matt-levine-og.png b/blog/2022/12/20/hacking-on-rose/.src/draft/.src/matt-levine-og.png
new file mode 100644
index 0000000..6fd184d
Binary files /dev/null and b/blog/2022/12/20/hacking-on-rose/.src/draft/.src/matt-levine-og.png differ
diff --git a/blog/2022/12/20/hacking-on-rose/.src/index.md b/blog/2022/12/20/hacking-on-rose/.src/index.md
new file mode 100644
index 0000000..eb5f169
--- /dev/null
+++ b/blog/2022/12/20/hacking-on-rose/.src/index.md
@@ -0,0 +1,37 @@
+Hacking on the Rose browser
+===========================
+
+I recently spent a few days hacking on the [rose browser](https://github.com/mini-rose/rose). It's a minimalist browser for Linux machines which is easy to play with. I enjoyed the experience, and thought to leave some notes.
+
+Rose is written in C, and uses the Webkit and GTK libraries. Webkit, which drives Safari, started out as a fork of some previous open-source libraries, [KHTML](https://en.wikipedia.org/wiki/KHTML) and [KJS](https://en.wikipedia.org/wiki/KJS_%28software%29). [GTK](https://en.wikipedia.org/wiki/GTK) is a library for creating graphical interfaces. You can use the two together using [WebKitGTK](https://webkitgtk.org/).
+
+
+* Pictured: An earlier version of this blogpost in the rose browser.*
+
+Personally, I find the simplicity of the code soothing. Still, to improve usability and for my own enjoyment, I've been adding some code to it. You can have a look at my tweaks [here](https://github.com/NunoSempere/rose-browser). These implement two main changes:
+
+1. I've implemented a [reader mode](https://github.com/NunoSempere/rose-browser/tree/personal/plugins/readability)
+2. I've added [redirection functionality](https://github.com/NunoSempere/rose-browser/tree/personal/plugins/libre_redirect), which redirects users to open source frontends of otherwise annoying websites.
+
+The reader mode, for instance, returns this beautiful to read page:
+
+
+
+when fed the cluttered aesthetic abomination that is Matt Levine's newsletter:
+
+
+
+I lifted the code for doing this from [Firefox](https://github.com/mozilla/readability), figured out how to integrate it with WebKit, and then made a few modifications.
+
+As mentioned, I also wrote a version of [LibRedirect](https://libredirect.codeberg.page/) for the rose browser. It which direct annoying websites into better open-source versions:
+
+
+
+Lastly, I've added some code for installing the Rose browser on [Ubuntu 20.04](https://github.com/NunoSempere/rose-browser/raw/personal/rose), which is what I'm personally using.
+
+If you are a close friend using Linux feel free to reach out if you get stuck trying to install this.
+
+
+
+
+
diff --git a/blog/2022/12/20/hacking-on-rose/index.md b/blog/2022/12/20/hacking-on-rose/index.md
index eb5f169..79dedf2 100644
--- a/blog/2022/12/20/hacking-on-rose/index.md
+++ b/blog/2022/12/20/hacking-on-rose/index.md
@@ -1,37 +1,32 @@
-Hacking on the Rose browser
-===========================
+Hacking on rose
+===============
-I recently spent a few days hacking on the [rose browser](https://github.com/mini-rose/rose). It's a minimalist browser for Linux machines which is easy to play with. I enjoyed the experience, and thought to leave some notes.
+The [rose browser](https://github.com/mini-rose/rose) is a minimal browser for Linux machines. I've immensely enjoyed hacking on it this last week, so I thought I'd leave some notes.
-Rose is written in C, and uses the Webkit and GTK libraries. Webkit, which drives Safari, started out as a fork of some previous open-source libraries, [KHTML](https://en.wikipedia.org/wiki/KHTML) and [KJS](https://en.wikipedia.org/wiki/KJS_%28software%29). [GTK](https://en.wikipedia.org/wiki/GTK) is a library for creating graphical interfaces. You can use the two together using [WebKitGTK](https://webkitgtk.org/).
+Rose is written in C, and it's based on Webkit and GTK. Webkit is the engine that drives Safari, and a fork of some previous open-source libraries, [KHTML](https://en.wikipedia.org/wiki/KHTML) and [KJS](https://en.wikipedia.org/wiki/KJS_%28software%29). [GTK](https://en.wikipedia.org/wiki/GTK) is a library for creating graphical interfaces. You can conveniently use the two together using [WebKitGTK](https://webkitgtk.org/).
* Pictured: An earlier version of this blogpost in the rose browser.*
-Personally, I find the simplicity of the code soothing. Still, to improve usability and for my own enjoyment, I've been adding some code to it. You can have a look at my tweaks [here](https://github.com/NunoSempere/rose-browser). These implement two main changes:
+Personally, I find the simplicity of the code soothing. At the same time, though, I've been adding some code to it, to make the browser more usable and for the fun of it. You can see my tweaks [here](https://github.com/NunoSempere/rose-browser).
-1. I've implemented a [reader mode](https://github.com/NunoSempere/rose-browser/tree/personal/plugins/readability)
-2. I've added [redirection functionality](https://github.com/NunoSempere/rose-browser/tree/personal/plugins/libre_redirect), which redirects users to open source frontends of otherwise annoying websites.
-
-The reader mode, for instance, returns this beautiful to read page:
+My first tweak was to add a [reader mode](https://github.com/NunoSempere/rose-browser/tree/personal/plugins/readability). It, for instance, returns this beautiful page:
-when fed the cluttered aesthetic abomination that is Matt Levine's newsletter:
+when fed the aesthetic abomination that is Matt Levine's newsletter:
-I lifted the code for doing this from [Firefox](https://github.com/mozilla/readability), figured out how to integrate it with WebKit, and then made a few modifications.
+I lifted the code for doing this from [Firefox](https://github.com/mozilla/readability), figured out how to integrate it with WebKit, and then improved it a bit.
-As mentioned, I also wrote a version of [LibRedirect](https://libredirect.codeberg.page/) for the rose browser. It which direct annoying websites into better open-source versions:
+My second big tweak was to add [redirection functionality](https://github.com/NunoSempere/rose-browser/tree/personal/plugins/libre_redirect) from annoying websites to their open source frontends, in the style of [LibRedirect](https://libredirect.codeberg.page/):
-Lastly, I've added some code for installing the Rose browser on [Ubuntu 20.04](https://github.com/NunoSempere/rose-browser/raw/personal/rose), which is what I'm personally using.
-
-If you are a close friend using Linux feel free to reach out if you get stuck trying to install this.
-
+And I've added some code for installing this browser on [Ubuntu 20.04](https://github.com/NunoSempere/rose-browser/raw/personal/rose), which is what I'm personally using. If you are a dear friend using Linux feel free to ask for help if you get stuck trying to install this.
+
diff --git a/blog/2022/12/23/ai-risk-rohin-shah/index.md b/blog/2022/12/23/ai-risk-rohin-shah/index.md
new file mode 100644
index 0000000..8f44e2e
--- /dev/null
+++ b/blog/2022/12/23/ai-risk-rohin-shah/index.md
@@ -0,0 +1,43 @@
+A basic argument for AI risk
+============================
+
+Rohin Shah [writes](https://www.lesswrong.com/posts/bkpZHXMJx3dG5waA7/ways-to-buy-time?commentId=jropYhtAW72zfHRBr) (referenced [here](https://forum.effectivealtruism.org/posts/2S3CHPwaJBE5h8umW/read-the-sequences)):
+
+> Currently, I'd estimate there are ~50 people in the world who could make a case for working on AI alignment to me that I'd think wasn't clearly flawed. (I actually ran this experiment with ~20 people recently, 1 person succeeded. EDIT: I looked back and explicitly counted -- I ran it with at least 19 people, and 2 succeeded: one gave an argument for "AI risk is non-trivially likely", another gave an argument for "this is a speculative worry but worth investigating" which I wasn't previously counting but does meet my criterion above.)
+
+I thought this was surprising, so here is an attempt, time-capped at 45 mins.
+
+### 1. The concern is valid in the limit: An entity of arbitrary, God-like intelligence would be very scary.
+
+An entity of arbitrary, God-like intelligence would be very scary. We already see that the most intelligent humans, like von Neumann and the other Hungarian Martians, were able to perform feats of science that steered humanity's fate.
+
+If we had an artificial system that was as intelligent as von Neumann, this would be very scary, because it might be able to perform similar feats of science and engineering. Also, Moore's law is uncertain, but it seems likely that in a generation we could go from having just one artificial von Neuman to having many of it. This increases the stakes.
+
+### 2. Year and year we are approaching the limit: What comes after Comprehensive AI Systems?
+
+We can think about how the world would look like in 2030-2035 if many tasks are delegated to successors of GPT-3. In this world, interfaces between many services and AI systems exist, and these are used to affect real-world systems. For example, I might routinely use GPT-3 to edit my texts, translations might be performed in real time through speech recognition, or [automatic threat detection](https://www.lesswrong.com/posts/vDvKWdCCNo9moNcMr/us-military-global-information-dominance-experiments) and deployment might be used by the world's military.
+
+Then we can think about, well, how does the world look like the decade after that one, and the decade after, and so forth. And it looks like ML systems generally acquiring more and resposability around managing human systems.
+
+And in general, we might worry that as current systems become more and more capable, they might eventually exceed humans and start to manifest some of the dangers that a being of God-like intelligence would also display. Crucially, we think that the human brain is made out of human atoms and so in principle replicatable in silico.
+
+### 3. Alignment proposals are uncertain, shaky and untested
+
+In this situation, we would love to have some mathematical proof that these AI systems which might end up making important decisions are, in some sense, friendly to humans. We don't have that guarantee. We might hope that there might be a strong incentive to create systems that increase human flourishing rather than reduce it.
+
+But on the other hand, if we look at e.g., the algorithms driving social media and reducing attention spans, this doesn't really inspire much confidence. We can also look at the organizing structure of society and notice that it alredy falls prey to Goodhart's law in that maximizing profit does not maximize flourishing. Some examples might be the [Sumangali system](https://en.wikipedia.org/wiki/Sumangali_%28child_labour%29), Nestlé [stealing water to bottle in a drought](https://www.theguardian.com/us-news/2021/apr/27/california-nestle-water-san-bernardino-forest-drought), or lobbying from weapon manufacturers making war more likely. We might worry that similar dynamics might exist if more and more decisions are put in the hands of AI systems and these are programmed to maximize something that at first looks like human flourishing but ultimately doesn't.
+
+We could also worry about more speculative failure modes, in which very intelligent systems at first appear to be helpful but then stop being so as they become harder to stop. We see that these kinds of things happen with [reward hacking](https://docs.google.com/spreadsheets/d/e/2PACX-1vRPiprOaC3HsCf5Tuum8bRfzYUiKLRqJmbOoC-32JorNdfyTiRRsR7Ea5eWtvsWzuxo8bjOxCG84dAg/pubhtml) in curent models, and pop up in [models of very intelligent systems](https://www.lesswrong.com/posts/nyCHnY7T5PHPLjxmN/open-question-are-minimal-circuits-daemon-free).
+
+### 4. Expected value calculations lead the way
+
+Because we expect these systems to be tightly integrated with the human economy, we would like to have guarantees. But we don't have them. Given that we don't have guarantees that bad outcomes won't happen, they might in fact happen. This would be bad.
+
+If we multiply the number of people who we would expect to be affected by future AI systems, this would be a large number. This is similar to how e.g., the US Federal Reserve sets monetary policy in the US, and so improving its decisions would reverberate across the US. But because AI systems will have a large scale impact, steering that impact would also be valuable. If we multiply out that impact (e.g., [here](https://www.beguide.org/hostile-artificial-intelligence.html)), the expected value turns out to be higher than other oportunities.
+
+
+---
+
+
+
+
diff --git a/blog/2023/01/03/me-in-2022/index.md b/blog/2023/01/03/me-in-2022/index.md
new file mode 100644
index 0000000..fa02579
--- /dev/null
+++ b/blog/2023/01/03/me-in-2022/index.md
@@ -0,0 +1,85 @@
+Me in 2022
+==========
+
+
+
+The above *.gif* shows a snapshot taken by my computer at 12:00 for a large number of days of the year. The below *.gif* shows a snapshot taken by my computer every five minutes on the 22/02/2022:
+
+
+
+### To create such a *.gif* in Linux
+
+**a. Create a script to take a photo from your webcam:**
+
+```
+## Defs
+panopticonDir="~/.panopticon"
+
+selfiesDir="$panopticonDir/selfies"
+today="$(date +"%Y%m%d")"
+timeNow="$(date +"%H%M%S")"
+
+## Init
+mkdir -p "$selfiesDir/$today"
+
+## Selfies
+
+selfieFileName="webcam_$timeNow.jpg"
+/usr/bin/fswebcam -r 1280x720 "$selfiesDir/$today/$selfieFileName"
+# ^ works with pipewire
+
+```
+
+**b. Call the script from within crontab**
+
+Call
+
+```
+$ sudo crontab -e
+```
+
+and then paste:
+
+```
+*/5 * * * * /home/loki/.panopticon/shoot.sh
+```
+
+**c. Put it all together into a gif**
+
+One gif from snapshots taken at 12:00.
+
+```
+summary_dir="~/.panopticon/summary-12"
+mkdir summary_dir
+for dir in ~/.panopticon/*/ ; do
+ if [[ -d "$dir" && ! -L "$dir" ]]; then
+ echo "Entering $dir";
+ dirname_raw="$(echo "$dir" | sed 's#/##')"
+ temp_photo_name="$(echo $dirname_raw/webcam_1200*)"
+ echo "$temp_photo_name"
+ cp $temp_photo_name temp.jpg
+ mv temp.jpg $summary_dir/$dirname_raw.jpg
+ fi;
+done
+
+convert -delay 10 -loop 0 $summary_dir/*.jpg 12.gif
+
+```
+
+One for each day with something like:
+
+```
+for dir in */ ; do
+ if [[ -d "$dir" && ! -L "$dir" ]]; then
+ echo "Entering $dir";
+ dirname_raw="$(echo "$dir" | sed 's#/##')"
+ convert -delay 10 -loop 0 $dirname_raw/*.jpg daily-gifs/$dirname_raw.gif
+ echo "Created daily_gifs/$dirname_raw"
+ fi;
+done
+
+```
+
+
+
+
diff --git a/blog/2023/01/07/forecasting-newsletter-november-december-2022/index.md b/blog/2023/01/07/forecasting-newsletter-november-december-2022/index.md
index c24ca86..2c0e7d9 100644
--- a/blog/2023/01/07/forecasting-newsletter-november-december-2022/index.md
+++ b/blog/2023/01/07/forecasting-newsletter-november-december-2022/index.md
@@ -8,7 +8,7 @@ Forecasting Newsletter for November and December 2022
- Forecasting community member Avraham Eisenberg [arrested for $100M+ theft](https://www.coindesk.com/policy/2022/12/27/mango-markets-exploiter-eisenberg-arrested-in-puerto-rico/) ([a](https://web.archive.org/web/20230106173959/https://www.coindesk.com/policy/2022/12/27/mango-markets-exploiter-eisenberg-arrested-in-puerto-rico/)).
- Forecasting Research Institute [launches publicly](https://ea.greaterwrong.com/posts/kEd5qWwg8pZjWAeFS/announcing-the-forecasting-research-institute-we-re-hiring)
- Blogpost [suggests that GiveWell use uncertainty](https://www.metacausal.com/givewells-uncertainty-problem/), wins $20k
-- Contrarian [offers $500k bet](https://www.metacausal.com/givewells-uncertainty-problem/), then chickens out
+- Contrarian [offers $500k bet](https://stevekirsch.substack.com/p/1m-bet-rules), then chickens out
- Walter Frick writes a [resource to introduce journalists to prediction markets](https://www.niemanlab.org/2022/12/journalists-wake-up-to-the-power-of-prediction-markets/)
- Yearly feedback form
@@ -61,7 +61,7 @@ The man behind the thing is [Filip Pidot](https://twitter.com/FlipPidot) ([a](ht
Polymarket deploys an [adapter](https://github.com/Polymarket/uma-ctf-adapter) ([a](http://web.archive.org/web/20221013234314/https://github.com/Polymarket/uma-ctf-adapter)) between its Keynesian beauty contest resolution mechanism, [UMA](https://umaproject.org/) ([a](http://web.archive.org/web/20230104180400/https://umaproject.org/)), and actual resolutions. There is an "emergencyResolve" feature, which judging by the name could be used to override or bypass the UMA resolution. This is marginally concerning, because it could be used to siphon money away from markets, by betting on the less likely outcome and then resolving to it.
-
+
Ultimately, I don't think this matters very much, because if Polymarket was compromised, users' wallets could probably be drained through some change in the webpage. Still, this downgrades UMA from its previous status as a source of truth to merely a convenient resolution mechanism,
@@ -105,7 +105,7 @@ Metaculus held a [Million Predictions Hackathon](https://scribe.froth.zone/annou
Vincent Luczkow [analyzes](https://github.com/vluzko/manifoldpy) ([a](http://web.archive.org/web/20221228201434/https://github.com/vluzko/manifoldpy)) whether Manifold Markets are well calibrated, and finds that they are:
-
+
A Manifold Community member analyzed the performance of various prediction platforms on the [US elections](https://firstsigma.substack.com/p/midterm-elections-forecast-comparison) ([a](http://web.archive.org/web/20221215030200/https://firstsigma.substack.com/p/midterm-elections-forecast-comparison)). I find that these analyses don't really take into account that political outcomes are massively correlated. And so in years in which Democrats do well, platforms with a Democratic bias will do well, and vice-versa. Still, this isn't totally uninformative.
@@ -199,7 +199,7 @@ Ben West looks at the [rate of fraud in YCombinator startups](https://ea.greater
This [old Washington Post article](https://www.washingtonpost.com/news/wonk/wp/2016/03/18/how-the-likelihood-of-breaking-up-changes-as-time-goes-by/) ([a](http://web.archive.org/web/20211005190327/https://www.washingtonpost.com/news/wonk/wp/2016/03/18/how-the-likelihood-of-breaking-up-changes-as-time-goes-by/)) shows the annual risk of breakup for married and unmarried couples:
-
+
Cultivate Labs' CEO [reviews](https://www.cultivatelabs.com/posts/2022-in-the-rearview) ([a](http://web.archive.org/web/20221223123956/https://www.cultivatelabs.com/posts/2022-in-the-rearview)) his company's activities in 2022. In particular:
@@ -267,8 +267,8 @@ A smallish company [rises](https://www.campaignlive.co.uk/article/truth-comes-be
Except that this time, the [trick](https://taz.de/!5137745/)[succeeds](http://www.bertelsmannkritik.de/index.htm) ([a](http://web.archive.org/web/20230105003044/http://www.bertelsmannkritik.de/index.htm)), and after the death of the founder, it allows the family to keep control of assets while minimizing tax exposure. But the logic of the scheme requires that the company give out as few dividends as possible—because they would then be trapped inside the foundation and be difficult to get out. Which is why Bertelsmann has greatly expanded and is now a $30B behemoth which owns [Penguin Random House](https://en.wikipedia.org/wiki/Random_House) ([a](http://web.archive.org/web/20221224150408/https://en.wikipedia.org/wiki/Random_House)) as one of many items in an absurdly [long list of assets](https://en.wikipedia.org/wiki/List_of_assets_owned_by_Bertelsmann) ([a](http://web.archive.org/web/20221031225743/https://en.wikipedia.org/wiki/List_of_assets_owned_by_Bertelsmann)). Amusingly, they also [have the same person](https://en.wikipedia.org/wiki/Brigitte_Mohn#Career) ([a](https://web.archive.org/web/20221114210641/https://en.wikipedia.org/wiki/Brigitte_Mohn#Career)) represent the interest of the family and of the foundation.
-
-*Cover of the "Heimat, wider Heimat" book, which translates to something like "Motherland, oh Motherland again", published by Bertelsmann. Note dog whistle connotations of [Heimat](https://en.wikipedia.org/wiki/Heimat#Modern_history). The author, Gustav Schröer, is one of four authors that I could quickly find that were both published by Bertelsmann and signed the "[Declaration of Allegiance of German Writers](https://de.wikipedia.org/wiki/Gel%C3%B6bnis_treuester_Gefolgschaft) ([a](http://web.archive.org/web/20230105003103/https://de.wikipedia.org/wiki/Gel%C3%B6bnis_treuester_Gefolgschaft))" to Adolf Hitler. The others are Rudolf G. Binding, Otto Flake and Fritz Müller-Partenkirchen. The case of Otto Flake is particularly interesting, since Bertelsmann rescued him from poverty in 1958 by reprinting old books of his.*
+
+* Cover of the "Heimat, wider Heimat" book, which translates to something like "Motherland, oh Motherland again", published by Bertelsmann. Note dog whistle connotations of [Heimat](https://en.wikipedia.org/wiki/Heimat#Modern_history). The author, Gustav Schröer, is one of four authors that I could quickly find that were both published by Bertelsmann and signed the "[Declaration of Allegiance of German Writers](https://de.wikipedia.org/wiki/Gel%C3%B6bnis_treuester_Gefolgschaft) ([a](http://web.archive.org/web/20230105003103/https://de.wikipedia.org/wiki/Gel%C3%B6bnis_treuester_Gefolgschaft))" to Adolf Hitler. The others are Rudolf G. Binding, Otto Flake and Fritz Müller-Partenkirchen. The case of Otto Flake is particularly interesting, since Bertelsmann rescued him from poverty in 1958 by reprinting old books of his.*
Now, the Bertelsmann foundation has made a small contribution to the forecasting world by starting the [Range Forecasting](https://www.rangeforecasting.org/) ([a](http://web.archive.org/web/20230105003141/https://www.rangeforecasting.org/)) platform, which runs on Cultivate Labs software. Per a recent Cultivate Labs newsletter:
diff --git a/gossip/.subscribe/index.html b/gossip/.subscribe/index.html
deleted file mode 100644
index 97a760b..0000000
--- a/gossip/.subscribe/index.html
+++ /dev/null
@@ -1,15 +0,0 @@
-
diff --git a/index.md b/index.md
index 3b9a759..ed93822 100755
--- a/index.md
+++ b/index.md
@@ -1,4 +1,4 @@
-I'm Nuño Sempere. I do research, write software, and predict the future.
+I'm Nuño Sempere. I [do research](https://quantifieduncertainty.org/), [write software](https://github.com/NunoSempere/), and [predict the future](https://samotsvety.org/).
