I’m a developer on the EA Forum (the website you are currently on). You can contact me about forum stuff at will.howard@centreforeffectivealtruism.org or about anything else at w.howard256@gmail.com
Will Howard
Karma: 876
Happy to see python holding its own against the real languages 😌
Hi Devin, it should work if you refresh the page. There is a bug where it fails the permissions check the first time you visit the page but then it passes the second time, this will be fixed properly later today
lmk if you have any more issues!
We’re working on adding a daily notification for new subforum comments, would you all prefer that this was opt in by default or opt out? You will be able to opt in/out either way, probably with a little bell at the top of the page. (upvote the replies below to vote)
Upvote this if you would prefer the default to be that you do get a daily notification if there are unread comments in the subforum
Upvote this if you would prefer the default to be that you do not get a daily notification if there are unread comments in the subforum
It looks like he is already a member, welcome Nir! Invites are a good idea though, we will probably add that at some point in the future, for now just dm-ing someone the link is pretty much the best way
Not currently but there will be soon
Overall I disagree with the general thrust of this post on the grounds that the risk of death from nuclear war (conditional on nuclear use in Ukraine) seems so much higher than from covid for the typical EA demographic, but I would like to +1 the point that people may be able to eliminate most of their risk without moving to completely the middle of nowhere.
Russia has enough active nuclear weapons to completely destroy something like 1-500,000 km^2*. The UK alone has a land area of 250,000 km^2, the US is around 10 million km^2 and NATO as a whole is 25 million. So even in a total nuclear war scenario it’s likely that just going to a relatively small town will eliminate most of the risk, and there are many places like this where you can have a good work from home setup and commute to the big city occasionally.
*rough estimate, where “completely destroy” means kill most people in the area:
- Russia has 1588 active warheads, with a total yield of ~800MT, so ~500 kilotons each
—This tool says that a 500kt warhead can destroy around 100-250 km^2
**another way of arriving at the same conclusion is that the US has 20,000 towns, which is far higher than 1588
My overall feeling is that:
Some of those things (slack, discord, facebook) are sub-optimal because they are private, things being public is good for acculturation of new people, and so it’s good to make a public place that competes with these
The discussion section on the other topics is intended for discussing what should be in topic description (although this isn’t that clear at the moment, maybe we should change the blurb at the top)
I’m also just like 🤷♂️ eh, how bad is it to have competing standards anyway? Competition makes it more likely that the best place for discussion will win out in the end, it’s kind of bad because it splits the discussion between multiple places. I have a weak opinion that the first thing is more important
Some examples from my last company:
(product thing) We got into a trap of mainly developing features for our biggest customer, which made the product less generally useful to other customers, which made us even more reliant on our biggest customer in a sort of downwards spiral. About half the stuff I did was pretty low value for this reason. This is a problem I’ve heard a lot of companies get into, it’s hard to avoid because from a short term perspective keeping your main source of revenue happy makes sense
(eng thing) Our api got really slow and we launched a big project (~3 engineers for like 3 months, including me) to fix it, basically by manually optimising individual endpoints. About 2 months in I worked out that a big part of the problem was due to our load balancer sending too many requests to the same instance in a row (for a reason I still don’t understand) and fixing this (which took ~1 day) solved like 70% of the problem, and counterfactually I think we wouldn’t have done a big performance project if this had been fixed beforehand. There were lots of ~valuable lessons~ I learned from this, mostly very object level stuff:
“Work expands to consume the resources allocated to it”—having a big project with lots of engineers biased me towards assuming that the solution would require lots of engineering work, I think it would have been fixed faster if I was told I only had 1 week or something
Keep things simple, especially things that are fiddly, rarely changed, and a single point of failure (servers) - we had kubernetes set up with load balancing, autoscaling, pods with multiple docker containers, fine tuned health checks etc. All of this stuff interacted in accursed ways which obscured the problem. We didn’t really need any of it, and if we had just had 1 Big Server the problem would have been obvious, and probably wouldn’t have happened in the first place
Reproduce bugs in exactly same environment in which they occurred—creating a clone of the production environment with no traffic would have also made it obvious that the problem was due to servers getting overloaded
Anecdotally: I did not donate for a while because of the ugh around actually making the donations (e.g. having to research different donation platforms, trying to avoid taxes), I just siphoned the money into a different pot and left it there. I guess I still donated the same amount in the end but I think in having a common knowledge best donation platform (this niche is now filled by GWWC/EA Funds) that makes it super easy would cause me to donate more
Question from a friend who is doing a startup: is Flutter a good mobile framework/what is the best cross-platform mobile framework? (this is for a b2b SaaS app, but I’m also interested in general)
Is CEARCH pronounced “search”?
It’s for a sports betting app 🤷♂️, not what I initially thought. My friend is doing a health tracking app but apparently he’s trying to convince his friend not to use Flutter for their app lol. They (the second order friend) have an engineer who is dead set on Flutter
From the little info I have I am leaning towards suggesting that he use a simple “wrapper around a website” type solution, as it shouldn’t require much native functionality, and they’ll probably need a web version anyway
I think the big problem with the suffering/day of life estimate is that it assumes suffering can’t go negative. If you think suffering can go as low as 0.015 suffering-units it doesn’t seem too much of a stretch to think their lives could be net positive.
(this is a general problem with reducing-suffering derived estimates imo)
Could you explain the “Number of days of life equivalent to pain of suffering” parameter in your guesstimate model? I don’t get that one
lmao
There is already a topic for the FTX discussion which you can add a filter for on the frontpage to reduce how much you see it:
Thank you for this, it was great! I find ‘trying to understand the successes and failures of altruistic movements’ really interesting in general