Best of Twitter - Week of July 20, 2020
I often look at people's achievements and think: I wish I'd done that.
More rarely, I see the work that went into those achievements and think: I wish I were doing that.
Chase the latter.
Y Combinator has lost its soul: A YC founder's perspective...
The batches and checks are too big and the team is too small making it marginally useful unless youre a b2b saas or dev tool company, and it's filled with tech climbers.
4. YC has become a tech climber MBA/sabbatical.
The bio group has changed things a bit, but there is a huge cohort of YC that is just PMs with 2-3 years of experience in FANG who want a break. Smart, capable nerds, but often lacking what it takes to really challenge anything.
It feels like the partners are attached to a late 2000s view of the world where tech is new, risky, and innovative.
Its not. You're funding the people who would have gone into consulting or finance 15 years ago. RPM at Facebook is the new IB @ goldman (and actually pays more)...
^ hmmmmmmmmm thread
More bad news for #growth_mindset: Two near-replication studies find zero or small negative influence of growth mindset beliefs about intelligence across the challenging transition to university. new paper @ISIRonline authors.elsevier.com/c/1bPOgaSXLuHBA
Building a biotech company in Silicon Valley is amazing, but I quickly learned to not take common startup-isms at face value & to validate for myself that they translate over to building in bio. Many don't. Here's some differences I've noticed so far:
I've got a brewing theory that COVID, for a certain cohort of the affluent Zoom class, has actually been a quality-of-life boon, and many would love the status quo to continue.
But I want to see data beyond my anec-data.
^ thread
I perceive opinions on twitter increasingly from an ecological perspective. If an opinion is possible, it will find someone to champion it. If it aligns with opinion having incentives, it's going to be abundant. I mostly don't think: oh, you're wrong! but ah, there you are!
This is especially fun with the subvariant "Try as you might, no amount of engineering or operational effort can cram that 0.1% number down, so it scales directly with the size of the business, and eventually you need an organization housing the teams housing specialists in X."
Thank you, future homesteaders! @MorlockP
I made a fully functioning search engine on top of GPT3.
For any arbitrary query, it returns the exact answer AND the corresponding URL.
Look at the entire video. It's MIND BLOWINGLY good.
cc: @gdb @npew @gwern
@zackkanter @pavelow_pr "We sort of had to collectively admit we were wrong on the premise that you will be happiest if you work on something you personally want to work on the most," perhaps valve is not the best example after all arstechnica.com/gaming/2020/07… (discussion: news.ycombinator.com/item?id=237880…)
This is an absolutely wild collection of reddit posts by someone through the ~decade long journey of trying heroin for first time to finally getting clean
It's like Boyhood but in all the most terrifying ways
A few months ago, I volunteered to put a $3,000 bounty on my own research.
107 issue reports later, I reflect upon what I’ve learned as part of a three-part blog series at @the100ci.
the100.ci/2020/06/29/red…
Nicholas Coles, PhD @coles_nicholas_
Discuss here: https://forum.guzey.com/t/best-of-twitter-week-of-july-20-2020/178