Sign in

Andrew Conner

@andrewconner.com
482 followers 164 following 278 posts

andrewconner.com

PostsRepliesMedia
Andrew Conner @andrewconner.com · 15/11/2024
Do you think the issue is one of over-inclusivity, or that a sort of Yarvin-friendly theme existed prior?
120
Andrew Conner @andrewconner.com · 14/06/2024
Cold plunge / saunas masochists. Breathwork tribes.
130
Andrew Conner @andrewconner.com · 14/06/2024
Supplement people, nootropics, peptides. Ayurveda, TCM.
220
Andrew Conner @andrewconner.com · 13/06/2024
and postrat before that. for the best, illegibility is a virtue for a reason
020
Andrew Conner @andrewconner.com · 13/06/2024
I was explaining to my wife how the red light people and woo people are totally different, and she didn’t believe me.
130
Andrew Conner @andrewconner.com · 13/06/2024
I want to make a map of all the wellness clusters. There’s western science people. Then, we have the grounding/red light/nasal breathing people. But these are different from the traditional woo people (crystals, etc). Peaters, keto, anti-BCAA. Who else?
220
Andrew Conner @andrewconner.com · 21/05/2023
030
Andrew Conner @andrewconner.com · 04/05/2023
Yeah this works. Though, this is kind of what Twitter does. But they care about engagement, so boost popular tweets more than I would choose to.
000
Andrew Conner @andrewconner.com · 04/05/2023
What you actually want is a "For You" algo that has good incentives. Instead of maximize engagement, maximize something else. Follows, likes, etc. The algo needs to be careful about accidentally optimizing for bubbles *and* outrage (ex: show an alt-right meme to a socialist, and they hate reply).
040
Andrew Conner @andrewconner.com · 04/05/2023
What typically happens: you find a couple people in each community that are interesting (from the platform, or from other sources). Those people like/repost in their community actively, thus your second-degree graph starts to include more of the communities you care about.
230
Andrew Conner @andrewconner.com · 04/05/2023
Examples: you like 3 (disjoint) groups: math, generative art, and finance. Suppose you don't know of the finance community yet. Finding it is the same as randomly sampling from every similar group, and is *very* low signal. What's worse is that it provides intermittent reward — bad dopamine loops.
120
Andrew Conner @andrewconner.com · 04/05/2023
Conclusion from playing with potential What's Hot algorithms: people say they want to see cool smaller communities, but sampling randomly from smaller network clusters ends up being random and not useful.i.e., the useful pockets you'd like eventually get reflected in the follows/likes/replies graph.
150
Andrew Conner @andrewconner.com · 01/05/2023
"Awareness month"s exist to sell you things
000
Andrew Conner @andrewconner.com · 01/05/2023
Teacher: "You are a wonderful student, already perfect in every way. You do not need to earn anything. You are already insightful, charming, and cunning. Your wit knows no bounds. Complete this 100 question multiplication speed test as fast as you can."
010
Andrew Conner @andrewconner.com · 29/04/2023
I suggest random loading spinners, delays, and !important's in your CSS to re-align expectations.
030
Andrew Conner @andrewconner.com · 25/04/2023
Arc'teryx seriously misunderstands my mom.
020
Andrew Conner @andrewconner.com · 25/04/2023
Not sure, but this was years ago and from my understanding, the project mostly failed. They had a few videos demoing what a modern OS could be, which was exciting. But turns out that's a hard problem, and Google's appetite for 10+ year projects diminished.
010
Andrew Conner @andrewconner.com · 25/04/2023
Snazzy! This was sorta a demo years ago for Google's Fuchsia operating system. A bummer it never went anywhere. The demo I saw allowed for sandboxed apps that could easily move between devices. Have an app on your laptop, instantly loaded on your desktop. Or phone. State was synced real time.
110
Andrew Conner @andrewconner.com · 25/04/2023
If you reload the feed, it's in the exact same order, and new posts are added to the top. Thoughts about this? Does this break in any predictable manner?
330
Andrew Conner @andrewconner.com · 25/04/2023
Example: there's 1000 potential posts to show you, and algo wants to show 100. It ranks the 1000 (based on distance to poster, engagement, etc), picks the top 100, and displays those in a feed. Then, importantly, they're *there* statically —
110
Andrew Conner @andrewconner.com · 25/04/2023
(more here: andrewconner.com/information-overlo…). A key part is being able to consume feeds linearly to curate content, and be able to "catch up". What if social feeds ranked content, but then only displayed it linearly?
120
Andrew Conner @andrewconner.com · 25/04/2023
Social networks do this because they *want* infinite feeds that intermix new content with old. It provides an intermittent reward, even if you've read most of the existing posts. I've designed a lot of my content consumption around the concept of separating curation from consumption
140
Andrew Conner @andrewconner.com · 25/04/2023
Every time you load the homepage you see the "best" posts. Reload, and it's re-computed. New posts emerge through the ranking. Overall, I think this is bad. You can never "catch up" on your feed. It's infinitely long, and if you wait an hour and come back, you'll see new posts mixed with old ones.
140
Andrew Conner @andrewconner.com · 25/04/2023
Algorithms do two things: filter and rank. Ranking can, in effect, filter by lowering posts down a feed, but practically, low-ranked posts are never shown. Digg was the first time I remember seeing dynamic algorithmic ranking, though I'm sure it existed earlier. Reddit did it as well...
160
Andrew Conner @andrewconner.com · 25/04/2023
Been thinking about timeline algorithm design. Some thoughts, would love feedback. Algorithms are useful/needed bc the amount of content is too much even if you consider just your follows. ie, do you want every reply? every liked post? Content from people you don't follow can be good as well!
150
Andrew Conner @andrewconner.com · 23/04/2023
If anyone wants to play with the data set powering this, it's available here: huggingface.co/datasets/andrewconne…
020
Reposted by Andrew Conner
David Gasquez @davidgasquez.com · 23/04/2023
Behold is the Bluesky World! Got "The Bluesky Dataset" from @andrewconner.com and used Graphext (from @victoriano.bsky.social) to visualize all the connections. It looks beautiful! 😍 The best thing is that you can explore it too! 👇 public.graphext.com/fe28f76fe67518a…
3247
Andrew Conner @andrewconner.com · 22/04/2023
Yeah, I scraped it. My current scraper is pretty tuned to my own use case, but let me see what I can do. My only real concern is a ton of people hitting their API really hard.
040
Andrew Conner @andrewconner.com · 22/04/2023
Yeah. It’s around 32k accounts.
130
Reposted by Andrew Conner
Andrew Conner @andrewconner.com · 21/04/2023
There's already a critical mass of people here, so if you're interested in chatting with other people hacking around on the Bluesky API, go here:
0142
Andrew Conner @andrewconner.com · 21/04/2023
There's already a critical mass of people here, so if you're interested in chatting with other people hacking around on the Bluesky API, go here:
0142
Andrew Conner @andrewconner.com · 21/04/2023
The issue seems to be CORS — loading a remote page cross-domain only works if they have a very open policy (or none at all). Solution is to proxy the request through the backend.
100
Andrew Conner @andrewconner.com · 21/04/2023
I saw this message and almost sent it to my friend Noah, before seeing who posted it.
000
Andrew Conner @andrewconner.com · 21/04/2023
You may like this essay I wrote: andrewconner.com/value-creation-vs-…
andrewconner.com
120
Andrew Conner @andrewconner.com · 21/04/2023
Solve Bloom's 2 sigma problem. I have ideas, but it's a nebulous and open problem how to do it at scale. If pulled off, impact in terms of net new value for recipients would be $trillions. en.wikipedia.org/wiki/Bloom%27s_2_s…
en.wikipedia.org
110
Andrew Conner @andrewconner.com · 21/04/2023
I have several. DM me on Twitter (@connerdelights) and I can give them to you — would love more PKM people here!
010
Andrew Conner @andrewconner.com · 21/04/2023
Let me sit with this. Valuable, want to think of the best way to approach it.
010
Andrew Conner @andrewconner.com · 21/04/2023
Started it but then felt I didn't have much to add/synthesize beyond the documentation. Any specifics or gaps you're interested in? What's most useful to you: bsky itself, or atproto?
110
Andrew Conner @andrewconner.com · 21/04/2023
I got you. andrewconner.com/things
andrewconner.com
Recommended Things
Having great things can meaningfully increase the quality of your life. However, in general, most people aren’t intentional about finding things that make day to day life better, don’t take inventory of the things that they benefit from the most, and recommend specific products that work well to
010
Andrew Conner @andrewconner.com · 20/04/2023
The database is 300mb, so too big for Github, but I'll share with anyone who's interested.
200
Andrew Conner @andrewconner.com · 20/04/2023
I'm calculating it. Establishing trust in a decentralized network is similar to knowing what web pages are trustworthy, since spam is near free and hard to stop. So, you can find someone's overall PageRank (global trust), but also run it just for your own network neighborhood (personal trust).
200
Andrew Conner @andrewconner.com · 20/04/2023
database access ↓
080
Andrew Conner @andrewconner.com · 20/04/2023
Updated my Bsky profile/follower sqlite database. Makes it pretty easy to see overall growth and what's happening in the network. Will post about follows cluster analysis later, but for now: how many unique people follow someone, per day? (Database is available if you want a copy, see below)
1131
Andrew Conner @andrewconner.com · 20/04/2023
I found it interesting that getting follows via app.bsky doesn't seem to give the creation date (yet you do get the profile view, which is handy), but com.atproto.repo.listRecords gives created at dates. So, to accurately get follower history, you need the latter.
020
Andrew Conner @andrewconner.com · 20/04/2023
I was curious how big the SpaceX Starship is. Answer is 390 ft with a 30ft diameter.
020
Andrew Conner @andrewconner.com · 20/04/2023
One character does it for me. 😬
150
Andrew Conner @andrewconner.com · 20/04/2023
Seeing your reflection from those closest to you in life.
010
Reposted by Andrew Conner
Andrew Conner @andrewconner.com · 20/04/2023
Get hyped
youtube.com
Watch SpaceX TEST Starship, the biggest rocket ever, LIVE
This is the first fully integrated full stack test flight of Starship and the mighty Super Heavy booster. At lift off, it will become the largest and most most powerful rocket to ever fly producing over twice as much thrust as the Saturn V that took humans to the moon. The goal of the test is to get as far along in the mission as possible with a handful of important goals such as; clearing the launch pad, reaching max Q, getting to stage separation, ignition of Starship, burn Starship's engines for 7 minutes and 20 seconds which would get Starship up to nearly orbital velocities and would place Starship on a suborbital trajectory that will cause it to reenter just north of Hawaii. This would allow the teams to test the reentry profile and heat shields for the first time from orbital velocities. Want more information? Check out our Prelaunch Preview written by Austin Desisto - https://everydayastronaut.com/starship-superheavy-orbital-flight-test/ Want to know where to watch this live? I made a video on how to visit Starbase and where to watch a launch from - https://youtu.be/aWvHrih-Juk -------------------------- Want to support what I do? Consider becoming a Patreon supporter for access to exclusive livestreams, our discord channel! - http://patreon.com/everydayastronaut Or become a YouTube member for some bonus perks as well! - https://www.youtube.com/channel/UC6uKrU_WqJ1R2HMTY3LIx5Q/join The best place for all your space merch needs! https://everydayastronaut.com/shop/ All music is original! Check out my album \
021
Andrew Conner @andrewconner.com · 20/04/2023
Get hyped
youtube.com
Watch SpaceX TEST Starship, the biggest rocket ever, LIVE
This is the first fully integrated full stack test flight of Starship and the mighty Super Heavy booster. At lift off, it will become the largest and most most powerful rocket to ever fly producing over twice as much thrust as the Saturn V that took humans to the moon. The goal of the test is to get as far along in the mission as possible with a handful of important goals such as; clearing the launch pad, reaching max Q, getting to stage separation, ignition of Starship, burn Starship's engines for 7 minutes and 20 seconds which would get Starship up to nearly orbital velocities and would place Starship on a suborbital trajectory that will cause it to reenter just north of Hawaii. This would allow the teams to test the reentry profile and heat shields for the first time from orbital velocities. Want more information? Check out our Prelaunch Preview written by Austin Desisto - https://everydayastronaut.com/starship-superheavy-orbital-flight-test/ Want to know where to watch this live? I made a video on how to visit Starbase and where to watch a launch from - https://youtu.be/aWvHrih-Juk -------------------------- Want to support what I do? Consider becoming a Patreon supporter for access to exclusive livestreams, our discord channel! - http://patreon.com/everydayastronaut Or become a YouTube member for some bonus perks as well! - https://www.youtube.com/channel/UC6uKrU_WqJ1R2HMTY3LIx5Q/join The best place for all your space merch needs! https://everydayastronaut.com/shop/ All music is original! Check out my album \
021
Andrew Conner @andrewconner.com · 19/04/2023
you're loved ❤️
050