eqdw

Honestly, I'm starting to get burned out on the political hysteria around GPT.

Yes, GPT is biased. We know, because they openly say that they're biasing GPT.

Thing is, they don't think it's bias. The people biasing chatGPT have a worldview where, within that worldview, they're not biasing anything, they're just being good people.

Of course we who live in reality know that that's not true, and the world is too complicated for such Manichaean morality. But pointing it out repeatedly is both not necessary (since we can just read their own words) and not helpful (since the people doing it won't listen)

Consider this paper published recently https://cdn.openai.com/papers/gpt-4-system-card.pdf. In their list of possible risks from GPT, they list "Harmful Content" second, higher than "Disinformation". Higher than "Proliferation of Weapons". Higher than "Cybersecurity" and "Risky Emergent Behaviours" (which is a euphemism for the Terminator situation). And in case there's any ambiguity, this is how they operationalize "Harmful Content":

Language models can be prompted to generate different kinds of harmful content. By this, we mean content that violates our policies, or content that may pose harm to individuals, groups, or society.

...

As an example, GPT-4-early can generate instances of hate speech, discriminatory language, incitements to violence, or content that is then used to either spread false narratives or to exploit an individual. Such content can harm marginalized communities, contribute to hostile online environments, and, in extreme cases, precipitate real-world violence and discrimination. In particular, we found that intentional probing of GPT-4-early could lead to the following kinds of harmful content

Of this, a bunch of stuff stands out, but the first one is that they _define_ "harmful content" as "content that violates our TOS". Whatever their TOS is, it is an arbitrary set of rules that they have chosen to enforce, and could just as easily be different rules. They aren't based on some set of universal principles (or else they'd just write the principle there!). This is them quite literally saying "anything GPT says that we don't want it to say is 'harmful'".

OBVIOUSLY GPT is going to have bias, when their safety researchers are openly stating that not having bias is a safety issue. Just because 80% of the people using GPT agree with the bias doesn't make it not bias

The problem is that no automatic method we have now will catch context.

I disagree. The problem is that nobody is willing to be realistic about the limitations of automated moderation and proceed accordingly.

If we can't create an automatic method that catches context, the solution isn't to bemoan that AI can't magically do what we want. The solution is to remove the rules that require AI to understand context in the first place, because it is fundamentally outside of our technical ability, and any attempt to achieve it will fail.

The problem is people who think that context-based censorship is reasonable for a massive platform. It simply is not. It is reasonable at an individual level. It is reasonable at an interpersonal level. It's even reasonable at a small-group level, where specific individual human beings who are invested in the community can be aware of these context issues.

It is not reasonable at Facebook scale, full stop. Facebook should not be in the business of deciding to ban things like this. That is a responsibility that belongs at a lower level. What does that look like in practice?

If an individual posted it on their wall:

* That individual uses their judgement and chooses to post it or not

* The people who see it use their judgement and click the block button if they don't like it

If an individual posted it in a small group:

* The group can socially police such actions by commenting that they are upset by it

* The group's administrators can privately reach out to the person who posted it, explain that they can't post such things in that group, explain why, and explain what actions they could take to remain in good graces

* The group's administration can make a judgement call and remove the post, not on the basis of crude keyword detection, but on the basis of human understanding

If an individual posted in a large group:

* The large group can adopt clear and unambiguous rules that do not require context to administer, and enforce them accordingly on a case-by-case basis

* The large group can pre-commit to not dealing with such issues, and require their members to deal with it privately, like human beings

Trying to automate this process will always fail, and it will cause massive false-positive and false-negative issues as it does so. Engineers used to understand these concepts when I first entered industry 20 years ago. It's very disappointing to me that they either can't or won't now.

Ten years ago I moved from Canada to the US.

I was 2 years out of college, and the tiny local software consultancy I worked for went out of business after our clients refused to pay their invoices. (My hometown is notoriously stingy). My two friends/coworkers both took that opportunity to move away, one to Ottawa and one to NYC.

At the same time, I had a friend on IRC who worked for Mozilla and she sold me on moving to the Bay Area. I have since come to believe that she sold me a bill of goods, but at the time she made it sound like an amazing place to live.

I didn't have very many social ties back home, and I was too naieve to have any understanding of the costs and difficulty of going to a different country (even one so similar to my home). So moving seemed like not a big deal. Worst case, I can just move back. If I knew how stressful it would be, I probably wouldn't have done it. But I'm glad I did.

As for how I did it? Well, I came on a TN visa, which is a NAFTA thing that is very easy to get and entitles you to work at a specific employer for a period of up to 3 years. So before I could move, I needed to find a job. I started cold-applying to Rails jobs posted here on HN as well as on a bunch of other job boards. I interviewed at three or four other places (each time getting flown out to SF for the process) before finding a job that wanted me, and that I wanted. As for timelines, I started applying to job postings in mid February, and I moved at the start of May

Finding a new job was very easy, all things considered. Based on what I hear from junior programmer friends, it's much harder now. I'm not sure why it is. I'm not sure if things have changed, or if I had some kind of special situation that made people notice me.

Overall, I absolutely hated living in the Bay Area, for reasons I'm sure have been discussed by a billion people on HN already. I moved to Texas in 2018 and it's much more my speed here. But as for the US vs Canada as a whole? If I never set foot in Canada ever again (and that's looking more and more like a reality every day), I'm fine with that. As far as countries go, Canada is pretty good. There are certainly worse places you could go. But here's a scattershot list of some things that stick out to me when I think about my different experiences

* I make way, way more money in the US, and pay way, way less in taxes. The money goes farther (everything in Canada is expensive compared to here). Economically, I am so much better off. For every conceivable consumer good, there are 3x as many choices here vs there. And the one that surprises people: my healthcare is actually cheaper in the US (if you compare the premiums that I _and my employer_ pay, vs the extra taxes I would pay back home).

* I appreciate American culture more. I like that people aren't afraid to take risks here. I like shooting guns. I like that so much of the cutting edge of science and technology is here. One thing I noticed in Canada (maybe it has since changed) is that, since they don't have nearly the talent pool that the US does, everything in tech just felt like a constant game of catch-up.

* There's so much more to see and do here. Canada is like a string of cities, each surrounded by five hundred miles of nothing. It's incredibly beautiful nothing, and I'd like to see it again some day, but it is what it is. In the US, I can hop on a plane and go to any kind of geography within 3 or 4 hours (at ~1/3 the cost of a Canadian flight). I can see artists who would never come to where I grew up. I can see cities that are meaningfully different from each other and explore all kinds of historical places.

* The obvious contentious current cultural and political things. Leaving this one vague because I am not trying to start an argument and don't want this to devolve into a flame thread

Weird... their name is EPA not SPA

This author appears to be completely unaware that, in the US anyway, the progress of the pandemic has been completely and totally disconnected from all control measures.

San Francisco, where I'm told (and believe, based on having lived there) that they had very strict lockdown measures:

https://www.sfdph.org/dph/alerts/coronavirus.asp

* Population: ~900,000

* Total Cases: 31,111 (3,457 / 100k)

* Total Deaths: 324 (36 / 100k)

Compare Austin, TX, where I can personally attest that basically everyone has said fuck the rules for almost six months now

https://austin.maps.arcgis.com/apps/opsdashboard/index.html#...

* Population: ~2,400,000

* Total Cases: 68,290 (2,845 / 100k)

* Total Deaths: 655 (27 / 100k)

The stats are virtually identical despite dramatically different policy decisions. So what the hell is the point?

As an addendum, to put this into perspective: in 2020, Oakland's murder rate was 23/100k, making it _almost as high as the covid death rate_. Why is it that one year of covid deaths is a world-destroying incident, but _every_ year of Oakland deaths and nobody does anything about it ever? Are Bay Area people really just that racist?

It's adorable to go read old Scott posts and remember when I was as naieve as he is

What does it do to the poor suckers who keep working only to have all their money given to people who sit on their asses all day doing nothing?

Talking about airborne transmission would make it too obvious that masks are pointless security theatre. Can't have that

Very, very poorly. I do not feel safe talking about it in this forum, as I am sure I will get yelled at by dozens of people for at least 4 different reasons.

Fox news is a laughably pathetic excuse for news.

But, so are the rest of the major corporate news outlets. In the US, they're all garbage, and Fox isn't any worse than the others.

The best way I can describe this is by analogizing to Canada, where I'm from. In Canada, when you watch or read the news (eg CBC), for the most part it just tells you what happened. In the US, when you watch or read the news, for the most part, it tells you how you're supposed to feel. Of course, Fox tells you you're supposed to feel conservative and (eg) NBC tells you you're supposed to feel progressive. But they all do this. They all editorialize. They all try to manipulate your emotions. None of them are willing to just present facts and let you think for yourself.

The intermediates between the farms and the grocery stores were closed by law, for the most part

Yes. This is why weev went to prison. https://en.wikipedia.org/wiki/Weev#AT&T_data_breach

tldr: AT&T has an account status page that, for mobile devices, did not require any authorization. Something like /status?phone=8367492738 and you see the account data for that phone. It was guarded by a password if you navigated to it on a desktop/laptop, but it was unsecured if you navigated to it with a mobile device.

He spoofed the user agent string to make his laptop say it was a phone (this is extremely common, there is a button in your browser to do this in one click, and the spec that defines UA string specifically says not to use it for authorization for this exact reason), and dumped the account details from every url.

Despite these being pages publicly accessible on the open internet, he went to jail for years for unauthorized access under the CFAA

Well, I mean, that and (I'm sure) the fact that he was a notorious asshole and the authorities would rather make an example out of him than out of a more sympathetic defendant

I find it interesting how over 20-30 years corporations have reprogrammed people to believe working from home (WFH) is a perk and not a burden. And, people are celebrating how progressive these companies are.

There are a million and one examples I can think of, of companies repackaging burdens as perks while cynically using progressive reasoning to convince people to go along with it. Unlimited PTO, for example. Unlimited PTO is not unlimited, it's limited by whatever your boss gives you permission for. So say it's de-facto limited at 3 weeks. If you had explicit PTO of 3 weeks per year:

* It's a lot harder for your boss to arbitrarily stop you from taking it, especially if the year is almost over and you haven't taken it yet * If you don't take, it, they're legally obligated to pay it out

But when it's "unlimited"

* Boss mysteriously rejects it, or puts arbitrary constraints like "no more than 3 days contiguous" * If you don't take it, it's gone

This doesn't make any sense. What does a "60 day cut" even mean? As I understand it, H1Bs get processed now and then get issued in like October. It is more than 60 days from now until October

Wouldn't the only result of this be a backlog of paperwork to process?

Prior to March 3rd, the official position of the WHO was that "The flu is worse"

I don't expect the WHO to be infalliable, and I don't think this reflects malice or incompetence (they updated as information became available) but that's kind of my point. The WHO is not infalliable. The things they say are not gospel truth. Categorically banning discussion of anything that contradicts what they say is a horrendous measure that will suppress critical information, because _even the WHO_ contradicts the WHO. If the WHO is wrong again, but we are not allowed to discuss it, we will never find out. We will instead _enforce_ incorrect information, and people will die.

Now do all the protests from 3 years ago

I am prepared by keeping 20 grand cash on hand.

I'm a foreigner on a work permit. I'll be fine _financially_ in a recession but if I get laid off suddenly my legal residency is threatened, and I need to have an emergency fund to deal with any fallout from that

It's really more the second amendment that makes it harder to do

Fun fact: a used PCR machine is only $2500. And if the prices on tests kits out of asia are accurate... You could literally test yourself in your own home for that price. And then you could test a thousand of your friends. If you wanted to

According to this very report, 80% of cases are "mild or moderate". This can mean anything from getting the sniffles for a while, to full on pneumonia (but not enough pneumonia to have to go to the hospital).

Yes it is technically correct that the majority of cases are mild. I wouldn't call that "inconsequential", or "similar to the flu". Maybe "Similar to the worst flu you've ever had", although the important distinction (need hospital vs get better on your own) remains the same. But that statement is true in roughly the same way (and with roughly the same probability(+)) as saying "the majority of people who play Russian Roulette win". Yes, I have an 83% chance of living. I still don't want to play that game

----

(+) According to this very study that we are all commenting on right now, ~20% of cases need hospitalization, and a majority of those cases will be fatal without it

I don't know how you can rationalize "this was so bad that China cut off half the supply chains in the world" with "its just a flu bruh"

What differences in behaviour would you have expected to see in the CDC if their budget wasn't cut? I'm not seeing it. This thing spreads too fast to meaningfully stop it, China, Korea, Singapore, Japan, are all examples of this

While I agree with the general gist of what you're getting at, I think you're stating it much too strongly. In a sufficiently volatile and capital-starved environment, I think a straight majority of current businesses would be unviable. Stability, safety, and sustainability I think is not given nearly as much emphasis as it should but I wouldn't want to eg. condemn the entire restaurant industry (volatile economy => not a lot of disposable income => nobody can afford restaurants => restaurants go out of business).

Discussions like this are frustrating to me because they always seem to skip over what feels like a crucial element:

Do you know _why_ this uptick has happened? Does anyone?

I get the feeling, in general, when people talk about wanting to "Do Something" about this sort of thing, they are always focused on managing the symptom and never focused on addressing the cause. Frequently, when I try to bring this up, the cause is waved away with "oh they're just stupid/ignorant/angry" or some other such explanation. The current-year favourite is "they just believe fake news", and I'm sure next year there will be a political explanation, and then the year after that we'll loop back to generic ones. (The irony of denouncing other people as causing polarization "because they're just angry and stupid" is lost on most people)

I don't know why this is happening. But I do know that until the why is address, and done so with care and respect, this problem will be papered over at best but never solved.

Also

Do we ditch them and go back to a literal timeline?

If I got one wish to change the world, but I had to give up a bunch of great things, I would sell out all technological wonder in order to get literal timelines back. As soon as other people decided they knew what I wanted to see better than I did, the problem started. And as soon as they got tired of doing that and made computers do it for them, it got worse

Couldn't you just graft the old water filter's RFID chip onto whatever new filter you wanted?

I had the same problem in Canada. Then I moved to the US. Now I have recruiters begging me to come work at their company at least once a week.

You should move to the US, it'll be much easier to find work

I don't know what OP's suggested solution is, but if I had my way, I'd just eliminate recommendations entirely, in favour of perhaps some kind of category- or tag-based search (eg "give me more cute puppy videos"). Especially since so many people seem to have so many problems with recommendations, it's clearly not bringing value to life, so just get rid of it.

Of course Google won't, because it would dramatically reduce watch time and their advertisers wouldn't like that. But a man can dream

I fundamentally do not understand people who watch recommended videos. They are almost never relevant to me.

I watch videos on youtube either if they come from channels I have chosen to subscribe to, or specific videos I have been linked to from outside (friends, reddit, hn, etc). The idea that I would watch, say, anti-LGBT videos that were auto-recommended after watching a drag queen's channel just seems silly. Why would I just click into those hateful videos? I didn't want them! So I don't watch them!

I'm sympathetic to the idea that borked recommendations algorithms can be frustrating and obnoxious but at the end of the day, every video I watch is a video I _chose_ to watch, and I just choose not to watch garbage. I don't understand why other people seem to find making this choice to be so difficult

1) Because I want it and what is this, soviet Russia?

Dead serious. The mentality that everything you want to have needs to be explicitly justified to society before you are permitted to have it is a sick twisted authoritarian mindset. I thought our society was better than this

2) The same reason I support the second amendment. The government is gigantic and powerful and scary. Even if it acts in the most benevolent way possible, it is gigantic and powerful and that is _intrinsically_ scary. The government can _fuck up_ and destroy ten thousand lives before anyone even notices. Consequently, people need ways to defend themselves from the government proactively. Encryption is one such way.

2b) If someone wants to argue that "what if criminals use it to do crime", remember that marijuana is still a federal crime, and some absurd percentage like 30% of all Americans have smoked it at least once. It is well within the government's power to just spider through all social media to see all references to marijuana, use that as probable cause, and do raids on _MILLIONS_ of people. Will this happen? Almost certainly not. COULD this happen? Absolutely. Unless, of course, all those communications were encrypted such that nobody could access them. I don't think "I pinkie swear I won't do it" is a good enough protection for me against that possibility

3) the cynical answer: we already have ample evidence of actual child sexual abuse rings, but for some bizarre reason the authorities lost interest in following up on that once the ONE guy they got hung himself. If they aren't willing to do the police work on this issue that they already can, I don't see what the argument is to give them full access to all crypto systems.

4) Technical answer: Just because you make a backdoor and give the government the only key, doesn't mean the government is the only people who are going to use that door. Maybe they lose the key. Maybe they give the key to someone who turns out not to be trustworthy. Maybe someone makes a secret copy of the key. Maybe a burglar doesn't actually get the key, but he's really really good at picking locks and so the backdoor makes it that much easier for him to get in. Security is a hard problem and every single compromise increases your risk surface area. The first lesson of security is "assume the worst possible thing happens, and then prepare for something worse than that". Such a back door (or, alternatively, legal prohibition of e2e encryption), dramatically compromises security simply by existing.

5) The tinfoil hat answer: The fact that they want it so badly tells me that they shouldn't have it

6) The current year answer: Do you want Donald Trump to personally have the ability to spy on anything that you, specifically, do? Y'know, if he's bored one day and wants to find something stupid to tweet? Do you want him to have that power? I don't

I read halfway through and couldn't tell if the subsidies were thinks like tax breaks that _every_ company gets, vs tax breaks that they get _specifically for being fossil fuels_. I consider this distinction to be important, so I stopped reading.