Hacker News Reader

Comment by klyrs | Hacker News Reader

My biggest technical issue with this "common carrier" movement is spam. The author mentions it, but seems to suggest that users can simply block or ignore spammers. As somebody who witnessed the commercialization of the internet, I find this extremely naive. We generally think of email as a "common carrier," but even there, spam is blocked. Blocking spam is censorship. Failure to automatically detect 1% of spam will still overwhelm non-spam content.

If I lose common carrier status because I'm blocking spam, and thus section 230 protections, am I on the hook if I fail to block a fraudulent spam comment that leads to a user being harmed?

View on HN

> We generally think of email as a "common carrier," but even there, spam is blocked. Blocking spam is censorship. Failure to automatically detect 1% of spam will still overwhelm non-spam content.

These aren't quite the same. The paper recommends common carrier protections only for publishing that people explicitly seek out or to whom they subscribe. If you're subscribed to spam, the solution is obvious and technically trivial (unsubscribe).

The spam problem for email is not trivial because email is, by design, a public addressing system which permits unsolicited messaging.

Well, the common carrier approach would not necessarily mean that every user can publish what they want.

Common carrier treatment arises when you control a unique piece of infrastructure, or, more loosely defined, a dominant one. As a result you need to provide non-discriminating access for competitors.

In my interpretation, this requirement would also be satisfied by Facebook allowing publishers and companies to create pages. A minimal threshold of importance that's equal for all page owners might be valid.

And the paper touches on this, mentioning that there's a valid role in moderating comments:

> There might thus be reason to leave platforms free to moderate comments (rather than just authorizing users to do that for their own pages), even if one wants to stop platforms from deleting authors’ pages or authors’ posts from their pages

Simple, outsource the moderation. Provide a common access API for third parties to provide moderation. Whether their home feed content resemble 4chan or the New Yorker is entirely up to the user.

I don't think this is a major issue at all. All the common carriers (phone line, cell phone, public square in shopping mall, etc.) designated by law have (or can have) spam problems. This does not change a thing. No serious common carrier will challenge the court with the argument: I can ban spammers so I can ban everyone I want to.

Phone companies are regulated as common carriers, but there's an exception allowing them "to block by default illegal or unwanted calls based on reasonable call analytics before the calls reach consumers."[1]

[1]: https://www.fcc.gov/consumers/guides/stop-unwanted-robocalls...

But that puts us back where we started:

FB, et. al., will argue that they are blocking unwanted messages based on reasonable analytics. So we will continue to see exactly the same type of feed curation as today.

The only difference (maybe) is that a platform (probably) can't just wholesale ban a particular person... but I bet that even the phone company is allowed to ban a person is deemed to be sufficiently abusive or harassing.

The fundamental problem is that there is a vocal minority who really, really don't want to believe that they are the vocal minority who everyone else wants to ignore. There isn't some magic set of laws and regulations you can create to force the silent majority to listen to you when they think you're dumb and annoying.

The first amendment grants you the right to speak and the right for people (who wish to) to listen, but it doesn't grant you the right to an audience for your speech.

If you really believe that applying common carrier laws to social media companies, with the same exception for spam that phone companies have, wouldn't make a difference, you have no reason to oppose that, right?

> There isn't some magic set of laws and regulations you can create to force the silent majority to listen to you when they think you're dumb and annoying.

You're not applying the rule correctly. Twitter and FB are preventing, say Trump, from posting at all, even to those who want to receive his messages, so they aren't just blocking his messages to recipients who don't want to receive them. That goes beyond simply "blocking unwanted messages".

In a decentralized system, users could have various "vouches" for the authenticity of their messages.

A trusted user could vouch for a new user, and/or the user could go through various types of third party verification, such as phone number, ID verification, or other verification, and then the third party could vouch for them on that basis.

You could even have monetary vouches, where a third party vouched that the user paid a certain amount of money, solved a captcha, or solved some computational problem.

The user could be in control of what measures would be required to reach their screen. As it stands, social media essentially incorporates a combination of trusted user vouching (you can see who is a friend of a friend), and monetary vouching (advertising).

Couldn't a social media site implement support for custom blocklists or moderators, and have users opt into them? Facebook can have their own Official TM Facebook Moderation Blocklist, which users can subscribe to and filter out all spam and hate according to Facebook's algorithm. Or users can choose not to and subscribe to someone else's or no blocklist at all. Therefore, there wouldn't be any issue with common carrier status since users can post without being hindered, but other users can tailor their feeds to only see posts they care about.

I think the biggest problem is less spam and more harassment. Simply blocking harassers isn't as effective as blocking spam due to the way social networks generally work. The harasser can just move on to your connections which is potentially even worse than them harassing you. Imagine an ex-partner sending revenge poor to all your contacts. Fighting that requires the ability for users to report potential bad actors and for the platform to have a centralized moderation mechanic to stop those bad actors.

Or not making porn.

Not all online harassment involves porn. I'd wager that the vast majority doesn't. You can be squeaky-clean porn-wise and still be the target of online harassment campaigns.

He could just have done that using phone calls or standard mail. The government handles those cases, we don't expect phone or post companies to inspect what we send and ban us if they don't like what we send.

I think this is a common failing for legal analysis of tech problems. Some years ago Lessig made the argument that radio bandwidth should be unregulated and belong to all, instead of licensed to cellphone carriers. The technical basis for the argument was that's how Ethernet hubs work, and they work great. Of course, nobody uses a hub today.

Could you explain what the common failing is?

To underestimate how technical changes affect real-life usage?

I often see the opposite fallacy: That tech people assume technically optimal solutions to be socially and legally optimal, which they often are not. To transfer this to your example: Perhaps the social and economic landscape of society would be better off if the spectrum belonged to us all. Maybe we would have developed legal and social norms to handle the hub nature?

Assuming some existing technical solution is optimal. Hubs existed for a time because they were cheap, not because they were good. Or assuming that some solution to an unsolved problem will be pulled out of the magical innovation hat. We'll just devise more efficient time sharing algorithms.

Thanks, the first fallacy is great, I had not thought of this before. It's similar to how people will not deem change possible until it has always existed (i.e. when convinced, they overwrite the memory of their own past opinion).

User has control over what is marked as spam, and can see spam messages if needed. If you block some content the way gmail blocks spam, it won't be censorship.

> If you block some content the way gmail blocks spam, it won't be censorship.

Are you thinking of the spam folder? Because the primary way that gmail blocks "spam" is by never delivering the email at all. Not to your spam folder, not anywhere.

I learned about this when I was unable to receive email from a personal friend.

Yes i was talking about the spam folder, i have never heard about this other way of blocking, and can't find anything about it on the internet. Is this behaviour documented anywhere?

https://support.google.com/mail/troubleshooter/2696779?hl=en

https://glockapps.com/blog/remove-ip-address-gmail-blacklist...

I guess part of my question with spam lately is why can't we just go after the spammers? It seems to me that acting as a public nuisance would be enough for the government to take action.

Because most of the spam is coming from other countries and the cost of tracking down every spammer is prohibitive.

Reply on news.ycombinator.com