Notescreated: ; modified:
2022-04-03 Centralization vs Decentralization
- pro-centralization: network effects
- pro-centralization: lower transaction costs (see Coase’s Theory of the Firm)
- pro-centralization: full specialization of agents by being able to serve just one “customer” instead of being in the open market
- pro-centralization: vision alignment to work on big problems
- pro-centralization: speed of innovation of the stack with vertical integration is higher
- lower barriers for entry for individuals: you can be specialized and don’t need to be a self-directed agent, thus costs for becoming useful and finding something useful to do are lower
- pro-decentralization: censorship resistance
although see My first impressions of web3 by Moxie:
After a few days, without warning or explanation, the NFT I made was removed from OpenSea (an NFT marketplace):
The takedown suggests that I violated some Term Of Service, but after reading the terms, I don’t see any that prohibit an NFT which changes based on where it is being looked at from, and I was openly describing it that way.
What I found most interesting, though, is that after OpenSea removed my NFT, it also no longer appeared in any crypto wallet on my device. This is web3, though, how is that possible?
A crypto wallet like MetaMask, Rainbow, etc is “non-custodial” (the keys are kept client side), but it has the same problem as my dApps above: a wallet has to run on a mobile device or in your browser. Meanwhile, ethereum and other blockchains have been designed with the idea that it’s a network of peers, but not designed such that it’s really possible for your mobile device or your browser to be one of those peers.
A wallet like MetaMask needs to do basic things like display your balance, your recent transactions, and your NFTs, as well as more complex things like constructing transactions, interacting with smart contracts, etc. In short, MetaMask needs to interact with the blockchain, but the blockchain has been built such that clients like MetaMask can’t interact with it. So like my dApp, MetaMask accomplishes this by making API calls to three companies that have consolidated in this space.
2021-08-02 Social gestures
- if you don’t reply for a while - apologize.
- if you forget about a meeting - apologize
- if you have to cancel a meeting right in advance - apologize profusely
2021-03-19 Google is plenty innovative
- Google Research employs more than 2k people
- Google X works on cool shit that doesn’t work (this is the point)
- Waymo is leading self-driving efforts
- Gmail is becoming better and better, fixing my typos, grammar, and helping to compose messages
- Calico spends hundreds of millions of dollars (I think?) a year on basic biological research
- Deepmind spends hundreds of millions of dollars a year on basic ML/neuroscience research
- Android is better than iOS (I used both recently and extensively and Android is miles ahead). Google Pixel is one of all-around nicest phones out there
Why resisting temptation is probably easier than you think
Resisting temptations supposed to be difficult. I think it’s often much easier than we are used to believe. 3 reasons:
- our preferences are very, very dumb and desires tend to fall off extremely quickly.
- also we often literally forget about the temptation even if we do something else for like 1 minute
- also brain interprets any amount of waiting as a credible signal that we’re prepared to wait and decreases temptation’s strength
- people who invite me to things but don’t send calendar invites
- people who do not let me know how the meeting went after I introduced them to someone
Listening to music in 2020
One way in which 2020 did not disappoint is in listening to music. I open a “My Mix” randomly generated playlist on Youtube and start listening to an almost perfect mix of my favorite+new songs. Love this.
How long does it take to write a post/essay?
- How Life Sciences Actually Work: 2 weeks non-stop, so around 100-150 hours of combined thinking and writing
- How to make friends over the internet: 12 hours over 2 days
- A single Quarterly links post: usually 15 hours over a few days
- Doing Good Better post: >100 hours over 6 weeks
- Why You Should Start a Blog Right Now: 5 hours at once initially + several hours over next several months
- Research Ideas: 10 hours over 1 day
- Every productivity thought I’ve ever had, as concisely as possible: 50 hours over 6 months + 5 hours in edits over the next year
- Where Does Talent Come From: 15 hours over 2 days
- On Friendship: 2 hours
Many of my posts start as a bunch of bullets that are accumulating in some text file, before reaching the critical mass and making me realize that I should finally write the post. Many of them are me just being hit by an idea (usually upon reading something on the internet) and starting to write immediately.
I also have many unpublished drafts which took anywhere from 1 minute to over 30 hours.
So, you want to write something…
Almost all the writing I’m most proud of was started early in the afternoon and finished at whatever the natural stopping point was, generally between 6 PM and 2 AM.
I have more than 100 text files in my “drafts” folder. Some of them consist only of a few lines; some are hundreds and thousands of words long. How many of them will I finish? 10 would be an optimistic prediction.
Behind every draft is an idea and all of them were created when that idea was burning in my mind. I had to take it out and put it into writing. However, a month, a week, a day later, the idea is no longer burning. Its home is now a text file that will live on my hard drive but to which I will probably never return. Perhaps I will open it, take a look at whatever I wrote and think “man, I should’ve just written it up then”.
If you have an idea burning in your mind – write now; everything else can wait.
(if something seems like to could start burning at some point but right now it’s too vague or uncertain, feel free to dump notes in the text file. But never start writing a coherent text and abandon it halfway through.)