3 ms·

Ignoring the targeting aspect for a second, our comments here on HN aren't vetted. If I make a medical claim or a legal claim or a financial claim, @dang doesn'

by prawn 5y ago

Ignoring the targeting aspect for a second, our comments here on HN aren't vetted. If I make a medical claim or a legal claim or a financial claim, @dang doesn't call in a doctor, lawyer or CPA to vet it before it gets live. That would be absolutely cost-prohibitive. And I suspect that vet-before-publish would assign more legal responsibility to the publisher which is partly why this doesn't happen on any major social network or forum. How could a human review an ad with a name, without having contact details for that named individual? Or with a photo if they didn't recognise that person? Or if it was an ad about a person such as a news item? Is there a way with digital keys so that any media claiming endorsement by a person can be signed off? Or requiring an equivalent of talent releases (I get these as photographer/videographer)?

BLKNSLVR 5y ago
To eschew responsibility for the vetting of the advertisement content, there would need to be better vetting, like KYC / AML requirements, for those submitting the ads in order to ensure there's some real person / company at the end of the chain to be brought into court.
- prawn 5y ago
  I can't see how that is possible without blocking countries from the advertising/participation pool. It's the Wild West otherwise. Pointless having a name if your pursuit doesn't have legal teeth.
  - BLKNSLVR 5y ago
    How does it work in pre-internet media advertising? > It's the Wild West otherwise. That's the "do nothing" case that we're in now though isn't it? > Pointless having a name if your pursuit doesn't have legal teeth. Can't advertise into country X if you don't have a legal presence in country X. (Simple? Pending the years of legislative navigation, by which time alternatives will have filtered into the ecosystem: decentralisation and non-censorability, etc.). Which is basically what you've said "blocking countries from the advertising/participation pool". So, yes, I think I agree :)
    - prawn 5y ago
      Pre-internet, there weren't infinite scammers hammering magazines and newspapers from overseas with questionable ads. There was still some junk, but no where near as relentless. And most products were physical, so even if you were advertising miracle skin repair, you were manufacturing and shipping a physical product so it was harder to appear and disappear. Ease of money transfer hasn't helped prevent scams either. The issue with locking down by country is, say you have an upstanding SaaS startup and you want to advertise to the world. Can you realistically justify registering your business in 100+ countries? Even something like Stripe Atlas is arguably too heavy-handed for side-project stage. I could see there being a verification engine whereby an advertiser posted a bond before that verification service endorsed them to publishers. But this would require countries to require publishers to use something like this, because they won't do it off their own back.
      - Nextgrid 5y ago
        > The issue with locking down by country is, say you have an upstanding SaaS startup and you want to advertise to the world It depends on what the ad is about. The verification requirements should be much more stringent for a financial or medical service than some random SaaS (the latter's customers can already be made whole by just disputing the transaction with their card issuer). I don't believe it's that far-fetched to require financial services to have a business presence in the local country and abide by local laws before being able to advertise.
  - Nextgrid 5y ago
    > I can't see how that is possible without blocking countries from the advertising/participation pool. "Nigerian price" scams were basically non-existent before Internet access became easy, cheap & widespread. It isn't that something fundamentally changed with people becoming more gullible or suddenly having more money to spend, it's just that Internet access allowed people in foreign countries with law/corrupt law enforcement to suddenly be able to make big bucks while being out of reach from the target country's law enforcement. Back in the day this would've been theoretically possible, but would've most likely required being on the ground (putting you in reach of LE) and having some initial credibility and funds to convince a media network to broadcast your scam. Maybe some barriers to entry are a necessary evil?
    - dragonwriter 5y ago
      > "Nigerian price" scams were basically non-existent before Internet access became easy, cheap & widespread. False; the model of scam (“advance-fee scam”) dates back to at least the 19th Century and became particularly popular in the 1980s, long before widespread internet access.
      - Nextgrid 5y ago
        I'm sure the model was there, but were they so widespread? I'd expect that without the internet and requiring a business presence in the country, it would raise the risk of law enforcement involvement considerably as opposed to just being able to spam email from a foreign country.
        dragonwriter 5y ago
        > I'm sure the model was there, but were they so widespread? I'd expect that without the internet and requiring a business presence in the country, it would raise the risk of law enforcement involvement considerably as opposed to just being able to spam email from a foreign country. They were less commonly addressed to the kind of people who would never even consider going along with them, because selectivity in presentation was a survival skill for scam artists, but, yes, advance-fee scams were rather notoriously common scams before the internet became popular.
  - toss1 5y ago
    >>without blocking countries from the advertising/participation pool Fine. If your country doesn't have sufficient commercial and legal infrastructure to have solid basic KYC and legal responsibility, you don't get to participate. Perhaps might provide motivation for countries to step up to the requirements of a civilized world — or not, and they can enjoy the benefits of a 'wild west' world. Pick a lane, but you don't get both
    - prawn 5y ago
      That's what I've been getting at, but it would require voluntary industry participation (hasn't and won't happen), or regulation from a set of countries able to drag everyone else along with them. EU might be a shot.
chillfox 5y ago
If you are charging for publishing the content (ads) then you should charge enough to cover the vetting. If you are not charging (comments) then different rules apply and you must take it down if reported for certain things (copyright infringement). If you are not charging (comments) AND you create your own content (news) then different rules apply and you must vet the content for certain things (libel, hate speech). If you are charging for content then you should probably also ask for whatever supporting evidence is required to do a reasonable job of vetting. If someone lies and provides a false release and you have a reasonable vetting process then you will likely be able to use that process as a defence.
Nextgrid 5y ago
HN comments don't scale by design. They are limited to the thread they're posted and to the effort it takes to manually post (most of the time - I am aware automation can briefly exceed this), and there is still moderation and banning not to mention some hidden anti-abuse mechanisms that could very well be culling a lot of spam we don't even realize was being posted. Both HN itself and our interaction with it takes into account the fact that the damage by a malicious user will be limited. We'd most likely reconsider our use of HN if everyone got a button "reply to all threads" that would auto-spam their reply to every thread and make the website unusable. Facebook "organic" content is the same. In most cases, organic content is limited in reach so that the potential damage caused by a malicious user is acceptable, although misaligned incentives sometimes prompt Facebook's algorithms to artificially boost the reach of organic content to increase engagement. Facebook paid ads are different - it's basically an officially-endorsed spam machine; you pay money and you are able to bypass the normal reach limits that organic content would get. This can cause a lot of damage and harm if used to spread malicious content, even more so when you can control targeting parameters to only show it to the most vulnerable and fly under everyone else's radar. > How could a human review an ad with a name, without having contact details for that named individual? If you're advertising a financial investment scheme it's reasonable to expect the company to have a real address, business registration, potentially some kind of license if the local laws require it, etc. If a financial investment company doesn't want to provide contact details, it's not an investment company, it's a scam. > Or with a photo if they didn't recognise that person? Or if it was an ad about a person such as a news item? Having a verified audit trail behind the ad would at least leave a paper trail for the copyright holder to go after who originally posted it. The idea isn't to be bulletproof, it's to make it both harder and riskier (legally) for malicious actors to operate. I believe it should be considered the "right thing" to do for a platform that allows ads to be pushed to potentially vulnerable people, but since it's Facebook and they have no concept of "the right thing", regulatory action is necessary.
- prawn 5y ago
  How are copyright holders going to chase spun-up shell companies in some random country abroad where there is zero chance of pursuing them? That'd be the main reason why Facebook is being pursued here, because getting to the advertiser is hopeless. OK, I'm a spammer. I say I'm not advertising financial investments but then my ad either more or less is a financial scam, or is some grey area where people are lured to a site (which gets changed after approval) or it names some new crypto thing which the vetter has never heard of before. Or it uses the likeness of a celebrity specific to the target country and that the vetter doesn't recognise. Does the celebrity want to be contacted for every submitted ad to answer "Is this you? Did you approve this?" Do you need to have vetters in every country and with knowledge of each industry - is this medical ad a scam? What about this one about naturopathy? Facebook might have the means of dealing with something like this, but for anyone smaller hosting ads that wouldn't be the case. So then you'd need a way of segmenting legislation by scale. Then it's whack a mole with new scammers reappearing if you smack one down. I can't see it happening without regulatory action either.

by prawn 5y ago

BLKNSLVR 5y ago

To eschew responsibility for the vetting of the advertisement content, there would need to be better vetting, like KYC / AML requirements, for those submitting the ads in order to ensure there's some real person / company at the end of the chain to be brought into court.

prawn 5y ago
I can't see how that is possible without blocking countries from the advertising/participation pool. It's the Wild West otherwise. Pointless having a name if your pursuit doesn't have legal teeth.
- BLKNSLVR 5y ago
  How does it work in pre-internet media advertising? > It's the Wild West otherwise. That's the "do nothing" case that we're in now though isn't it? > Pointless having a name if your pursuit doesn't have legal teeth. Can't advertise into country X if you don't have a legal presence in country X. (Simple? Pending the years of legislative navigation, by which time alternatives will have filtered into the ecosystem: decentralisation and non-censorability, etc.). Which is basically what you've said "blocking countries from the advertising/participation pool". So, yes, I think I agree :)
  - prawn 5y ago
    Pre-internet, there weren't infinite scammers hammering magazines and newspapers from overseas with questionable ads. There was still some junk, but no where near as relentless. And most products were physical, so even if you were advertising miracle skin repair, you were manufacturing and shipping a physical product so it was harder to appear and disappear. Ease of money transfer hasn't helped prevent scams either. The issue with locking down by country is, say you have an upstanding SaaS startup and you want to advertise to the world. Can you realistically justify registering your business in 100+ countries? Even something like Stripe Atlas is arguably too heavy-handed for side-project stage. I could see there being a verification engine whereby an advertiser posted a bond before that verification service endorsed them to publishers. But this would require countries to require publishers to use something like this, because they won't do it off their own back.
    - Nextgrid 5y ago
      > The issue with locking down by country is, say you have an upstanding SaaS startup and you want to advertise to the world It depends on what the ad is about. The verification requirements should be much more stringent for a financial or medical service than some random SaaS (the latter's customers can already be made whole by just disputing the transaction with their card issuer). I don't believe it's that far-fetched to require financial services to have a business presence in the local country and abide by local laws before being able to advertise.
- Nextgrid 5y ago
  > I can't see how that is possible without blocking countries from the advertising/participation pool. "Nigerian price" scams were basically non-existent before Internet access became easy, cheap & widespread. It isn't that something fundamentally changed with people becoming more gullible or suddenly having more money to spend, it's just that Internet access allowed people in foreign countries with law/corrupt law enforcement to suddenly be able to make big bucks while being out of reach from the target country's law enforcement. Back in the day this would've been theoretically possible, but would've most likely required being on the ground (putting you in reach of LE) and having some initial credibility and funds to convince a media network to broadcast your scam. Maybe some barriers to entry are a necessary evil?
  - dragonwriter 5y ago
    > "Nigerian price" scams were basically non-existent before Internet access became easy, cheap & widespread. False; the model of scam (“advance-fee scam”) dates back to at least the 19th Century and became particularly popular in the 1980s, long before widespread internet access.
    - Nextgrid 5y ago
      I'm sure the model was there, but were they so widespread? I'd expect that without the internet and requiring a business presence in the country, it would raise the risk of law enforcement involvement considerably as opposed to just being able to spam email from a foreign country.
      - dragonwriter 5y ago
        > I'm sure the model was there, but were they so widespread? I'd expect that without the internet and requiring a business presence in the country, it would raise the risk of law enforcement involvement considerably as opposed to just being able to spam email from a foreign country. They were less commonly addressed to the kind of people who would never even consider going along with them, because selectivity in presentation was a survival skill for scam artists, but, yes, advance-fee scams were rather notoriously common scams before the internet became popular.
- toss1 5y ago
  >>without blocking countries from the advertising/participation pool Fine. If your country doesn't have sufficient commercial and legal infrastructure to have solid basic KYC and legal responsibility, you don't get to participate. Perhaps might provide motivation for countries to step up to the requirements of a civilized world — or not, and they can enjoy the benefits of a 'wild west' world. Pick a lane, but you don't get both
  - prawn 5y ago
    That's what I've been getting at, but it would require voluntary industry participation (hasn't and won't happen), or regulation from a set of countries able to drag everyone else along with them. EU might be a shot.

chillfox 5y ago

If you are charging for publishing the content (ads) then you should charge enough to cover the vetting. If you are not charging (comments) then different rules apply and you must take it down if reported for certain things (copyright infringement). If you are not charging (comments) AND you create your own content (news) then different rules apply and you must vet the content for certain things (libel, hate speech). If you are charging for content then you should probably also ask for whatever supporting evidence is required to do a reasonable job of vetting. If someone lies and provides a false release and you have a reasonable vetting process then you will likely be able to use that process as a defence.

Nextgrid 5y ago

HN comments don't scale by design. They are limited to the thread they're posted and to the effort it takes to manually post (most of the time - I am aware automation can briefly exceed this), and there is still moderation and banning not to mention some hidden anti-abuse mechanisms that could very well be culling a lot of spam we don't even realize was being posted. Both HN itself and our interaction with it takes into account the fact that the damage by a malicious user will be limited. We'd most likely reconsider our use of HN if everyone got a button "reply to all threads" that would auto-spam their reply to every thread and make the website unusable. Facebook "organic" content is the same. In most cases, organic content is limited in reach so that the potential damage caused by a malicious user is acceptable, although misaligned incentives sometimes prompt Facebook's algorithms to artificially boost the reach of organic content to increase engagement. Facebook paid ads are different - it's basically an officially-endorsed spam machine; you pay money and you are able to bypass the normal reach limits that organic content would get. This can cause a lot of damage and harm if used to spread malicious content, even more so when you can control targeting parameters to only show it to the most vulnerable and fly under everyone else's radar. > How could a human review an ad with a name, without having contact details for that named individual? If you're advertising a financial investment scheme it's reasonable to expect the company to have a real address, business registration, potentially some kind of license if the local laws require it, etc. If a financial investment company doesn't want to provide contact details, it's not an investment company, it's a scam. > Or with a photo if they didn't recognise that person? Or if it was an ad about a person such as a news item? Having a verified audit trail behind the ad would at least leave a paper trail for the copyright holder to go after who originally posted it. The idea isn't to be bulletproof, it's to make it both harder and riskier (legally) for malicious actors to operate. I believe it should be considered the "right thing" to do for a platform that allows ads to be pushed to potentially vulnerable people, but since it's Facebook and they have no concept of "the right thing", regulatory action is necessary.

prawn 5y ago
How are copyright holders going to chase spun-up shell companies in some random country abroad where there is zero chance of pursuing them? That'd be the main reason why Facebook is being pursued here, because getting to the advertiser is hopeless. OK, I'm a spammer. I say I'm not advertising financial investments but then my ad either more or less is a financial scam, or is some grey area where people are lured to a site (which gets changed after approval) or it names some new crypto thing which the vetter has never heard of before. Or it uses the likeness of a celebrity specific to the target country and that the vetter doesn't recognise. Does the celebrity want to be contacted for every submitted ad to answer "Is this you? Did you approve this?" Do you need to have vetters in every country and with knowledge of each industry - is this medical ad a scam? What about this one about naturopathy? Facebook might have the means of dealing with something like this, but for anyone smaller hosting ads that wouldn't be the case. So then you'd need a way of segmenting legislation by scale. Then it's whack a mole with new scammers reappearing if you smack one down. I can't see it happening without regulatory action either.