Matt Goodman

Karma: 806

Matt Goodman Mar 16, 2023, 12:04 PM
9 points
2 ∶ 1
on: Reminding myself just how awful pain can get (plus, an experiment on myself)
I went through these experiences voluntarily and with the knowledge that I have the freedom to stop whenever I want. People suffering from painful disease, children dying of hunger, chickens being electrocuted to death, fish being asphyxiated to death—for these individuals, such experiences are a horrific reality, not an experiment
I think this is a very important distinction that should be given more emphasis. When I’ve experienced severe pain, the no.1 thought in my mind was “oh god make it stop”. This makes complete sense if you think of pain as your body’s way of saying, “ok, whatever it is you’re doing, you need to stop doing it now.” And I think a lot of the psychological suffering I experienced was due to the stress of not being able to stop the thing that was causing pain, and not knowing how long the pain would go on for. I add the word ‘psychological’ for clarity here, but in reality I don’t think there’s a clear difference between ‘psychological’ and ‘physical’ sources of pain. All pain in a sense is psychological—all of it happens ‘in your mind’, and factors such as knowing the pain will end soon can have a big effect on the experience of pain.
This distinction could also have a big effect on how people rate their pain on the pain-track framework. The framework seems to define pain a lot in terms of ‘how long could a person endure this?’ And that answer probably varies a lot depending on whether you know the pain will go away soon, or not. For ‘disabling’ pain, it could literally be less disabling, if you knows it’s going to end soon. You might think something like, “ok, I know this will end in 5 minutes, for now I’m going to do this other job to distract myself”. And looking back at the experience, and your behaviour at the time, you might read the scale, and think “ok it’s wasn’t that disabling, I could still do stuff”.

Matt Goodman Mar 16, 2023, 10:16 AM
3 points
1 ∶ 0
on: Reminding myself just how awful pain can get (plus, an experiment on myself)
Hey Ren, this is a great post!
I share your intuition that reducing extreme suffering is the no.1 moral imperative for humankind.
What charities do you recommend, if that’s what you value most? GiveWell recommended charities based on their own moral weights, which I don’t think weight as reducing extreme suffering as highly as me.
Then there’s many animal welfare charities. And there’s OPIS, which is the only charity I know that explicitly targets extreme human suffering. Are there any others that I’m missing?

Matt Goodman Mar 14, 2023, 9:22 PM
6 points
1 ∶ 0
in reply to: Ben Millwood🔸’s comment on: What is the formal definition of a ‘Factory Farm’?
My guess is that it wouldn’t change much
Maybe not for most people reading the people reading the EA forum. I think if you take a serious look at the issues of animal suffering and farmed animal conditions, you’ll probably arrive at a number similar to existing statistics on numbers of factory farmed animals.
But I think there’s plenty of people who have motivated reasoning to doubt those statistics, or minimise the badness/factory-ness of a farm, or farming practice. For example, my extended family run a dairy farm. I remember when first reading about factory farms thinking ‘well, the family farm isn’t like these factory farms… right? ’
I also think it’s possible animal agriculturists will seize on uncertainty around the term ‘Factory Farm’ to sow confusion and whitewash animal welfare issues. Suppose that in the future, the concept of ‘Factory Farms’ gains widespread public vilification, in the same way that ‘Fossil Fuels’ does now. Now imagine a pan-European animal agriculture lobby group seizes on the looseness of the term ‘Factory Farm’ to ensure European farms aren’t associated with it:
European farms aren’t Factory Farms! We have better animal welfare standards here. There are cage-free policies here! Animal welfare laws! Standards and checks! It’s only farms outside of Europe that are factory farms, those are the ones that should be counted in the statistics, not European farms!
I don’t see this as “economic or moral incentive to sit on the borderline” but rather ‘if forced to adhere to higher welfare standards, there’s an incentive to maximise the reputational gain from this’.
edit: added last paragraph

Matt Goodman Mar 6, 2023, 11:54 PM
16 points
4 ∶ 1
on: Matt Goodman’s Shortform
Why aren’t we protesting AI acceleration in the street?
I’m not super up to date with the latest EA thinking on current AI capabilities. The takes I read on social media from Yudkowsky and the like are something along the lines of ‘We’re at a really dangerous time, various companies are engaged in arms race to make more and more powerful AIs with little regard to safety, and this will directly lead to humanity being wiped out by AGI in the near future’. For people really believe this to be true (especially if you live in San Fransico) - why aren’t you protesting on the street?
Some reasons this might work:
- There’s lots of precedents of public pressure leading to laws being passed or procedures changed, that have increased safety standards across many industries
- The companies working on AI alignment are based in San Francisco. There’s a big EA and rationalist community in SF. Protests could happen outside the HQ of AI companies.
- Stories about silicon valley tech companies get lots of press coverage in mainstream media
- There’s a prevailing anti—big tech companies feeling in parts of society that could be tapped into it
- Specifically, there’s criticisms of the newest AIs for things like ‘training AI models on artists’ work, then putting artists out of a job’ (Dalle) or ‘making it much easier to cheat at university’ (ChatGPT). Whilst this isn’t directly related to AGI safety, it’s the kind of feeling that could be tapped into for the purpose of this protest
- If an AI safety researcher could be interviewed on camera at the march it adds credibility to the march, that experts are concerned
- It adds credibility to the voices of experts warning about AI risk, if they’re so worried they’re willing to get out on the street to protest about it

Matt Goodman Feb 21, 2023, 10:46 PM
5 points
6 ∶ 3
in reply to: Holly Elmore ⏸️ 🔸’s comment on: EA, Sexual Harassment, and Abuse
I feel uncomfortable with this kind of public character judgement of an alleged victim. Especially when it’s presented without a source or evidence backing up the claim she’s ‘hella scary’

Matt Goodman Feb 17, 2023, 8:29 AM
3 points
1 ∶ 0
in reply to: Matt Goodman’s comment on: Dissatisfied with the state of EA? You can give me a call.
maybe ‘social-justice-caring left’ is a better term

Matt Goodman Feb 17, 2023, 12:06 AM
4 points
2 ∶ 0
on: Dissatisfied with the state of EA? You can give me a call.
I think using the term ‘woke left’ will be counter-productive to your aim of reaching out to politically left people. While ‘woke’ started as a term used by the left, I now see it being used almost exclusively by the right as a pejorative term for the left, and most politically left people I know would be annoyed at being called ‘woke’.

Matt Goodman Feb 4, 2023, 1:55 PM
1 point
0 ∶ 0
in reply to: Ozzie Gooen’s comment on: Forecasting Our World in Data: The Next 100 Years
What would that add? I think that would add speculation on to what is already speculation, and I’d think only the passing of time would be able to give feedback on whether the predictions turn out to be true.
I guess it could give more information, if you sought out different people for the meta-predictions, than had made the original predictions. But then I’m not sure why you wouldn’t just have these new people do the original prediction questions directly.

Matt Goodman Feb 4, 2023, 1:50 PM
4 points
2 ∶ 1
in reply to: None of them along the line Know what any of it is worth’s comment on: EA, Sexual Harassment, and Abuse
I think this might be partly due to the complex structure (and subsequent re-structure) of CEA. ‘CEA’ used to be a dual name for both a legal entity and the community building organisation.
I think this led me in the past to having a vague idea of what ‘CEA’ was, and thinking that the public-facing Community Health Team was representing all of it and responsible for more than they were.
This is kind of a separate issue though, here I’d just like to say I’m grateful for the work the Community Health Team does, and don’t want to distract from the discussion of the accusations made here.

Matt Goodman Feb 2, 2023, 2:46 PM
4 points
1 ∶ 0
in reply to: Jason’s comment on: Questions about OP grant to Helena
Can you expand on what you dislike about the marketing? When looking at their website I was just dazzled at all the animations, my developer brain was trying to figure out how they worked ;)

Matt Goodman Feb 1, 2023, 11:32 PM
1 point
0 ∶ 0
in reply to: Rasool’s comment on: EA Forum feature suggestion thread
Thanks, I didn’t know that one!

Matt Goodman Jan 31, 2023, 1:21 PM
1 point
0 ∶ 0
in reply to: Eigengender’s comment on: Time-stamping: An urgent, neglected AI safety measure
Assuming that this is both useful and time or funding constrained, you could be selective in how you roll it out. Images of world leaders and high profile public figures seem most likely to be manipulated and would have the highest negative impact if many people were fooled. You could start there

Matt Goodman Jan 30, 2023, 3:46 PM
3 points
0 ∶ 0
in reply to: dan.pandori’s comment on: EA Forum feature suggestion thread
Props for taking the time to explain, even though you don’t like it!

Matt Goodman Jan 30, 2023, 3:44 PM
1 point
0 ∶ 0
on: EA Forum feature suggestion thread
I’d like to be able to hide the amount of karma and agreement points a comment or post has. I think seeing how many people have upvoted a statement affects how likely I am to agree with or upvote that statement. I think it makes me more likely to vote in accordance with social agreement, rather than whether or not I think a statement is true or well written. I’d like to be able to turn this off from time to time. Strongly downvoted comments should probably still be hidden.

Matt Goodman Jan 30, 2023, 3:31 PM
3 points
0 ∶ 0
on: EA Forum feature suggestion thread
I think the UI for voting could be improved in the following ways:
- The arrows for voting on Karma point sideways, not up and down. It’s not immediately clear which one is upvote and which one is downvote.
- The explanation text about voting (the one that explains Karma, agree/disagree and strong votes) only appears when you hover your mouse over the arrows. This means you never see it on mobile, where there’s no mouse.
- the hit boxes could be bigger for arrows on mobile.

Matt Goodman Jan 30, 2023, 3:14 PM
1 point
0 ∶ 0
on: EA Forum feature suggestion thread
The formatting toolbar doesn’t appear until after you highlight text. This means you can only format text after you’ve written it—you can’t for example, select bold and have your text appear in bold as you write it. This is something I find unintuitive. It took me a a few minutes of looking for the toolbar and googling how to do it before I realised the toolbar only appears when you highlight text. I’d like the formatting toolbar to always be on the page when I’m writing.

Matt Goodman Jan 30, 2023, 3:10 PM
1 point
0 ∶ 0
on: EA Forum feature suggestion thread
I’d like to be able to highlight a word or phrase in text I’m writing and Ctrl-V a URL link directly into that phrase. This is something that other platforms, like Slack do.
Yes, you can highlight a phrase and bring up the toolbar to add a link, but being able to do it immediately through a well known keyboard shortcut is easier.

Matt Goodman Jan 30, 2023, 3:04 PM
40 points
17 ∶ 1
on: Announcing Interim CEOs of EVF
Thanks for writing this up. It’d be nice to have a paragraph of bio for each of the board members on ev’s website. Google search didn’t give me much for some of the board members.
What links here?
- Matt Goodman's comment on Updates to the Effective Ventures US board by Zachary Robinson🔸 (Apr 18, 2023, 2:07 PM; 75 points)
- Matt Goodman's comment on Short bios of 17 “senior figures” in EA by Kirsten (Jun 29, 2023, 10:19 PM; -13 points)

Matt Goodman Jan 30, 2023, 2:53 PM
1 point
0 ∶ 0
on: Time-stamping: An urgent, neglected AI safety measure
Are you planning to back-date each piece of content that you timestamp to the time it was created? If so, how hard is it to find the time of creation of pieces at the moment? This seems to be the very problem this initiative is planning to tackle (so I’m guessing it’s at least somewhat hard) although I think the argument here is ‘it will get harder in the future’.
The only alternative I can see is to add the timestamp of the time that the content is processed as part of this initiative. This might be easier than finding the creation date but it would probably contain less useful information, because a bunch of content would all be timestamped around the time this initiative got started. And it might even be misleading, if people saw the timestamp and thought this was the time the content was actually created.

Matt Goodman Jan 21, 2023, 9:35 PM
8 points
0 ∶ 0
in reply to: Matt Goodman’s comment on: Does EA understand how to apologize for things?
Update: FLI FAQ on the rejected grant proposal controversy.
Although I still think the original statement was not good, reading the FAQ and comments in the linked post have helped me have more empathy for the difficulties of releasing a PR when under public pressure to say something urgently.
I think my tone here was too confrontational and demanding, and I’m sorry if that caused additional stress for FLI.
Thankyou to FLI for updating both the initial statement, and putting out the FAQ, which clears things up.