This really isn't possible to implement in any practical or consistent way without contradictions in usability. If you made such a law, trolls such as Nazi-proponents would use it as a weapon to infect sites such as HackerNews with their walls of hate speech.

Moderators couldn't flag nor delete the hate speech as "off topic" because the Nazi group would complain of "censorship". Imagine that scenario: it would be against the law to delete hate speech here even though most us on HN would rather talk about programming languages or tech startups.

The way marginal groups have free speech now is at the level of DNS servers by buying a domain name. Instead of trying to get Facebook and HackerNews to host your hate speech by expanding "free speech to private entities", you create a Nazi domain (e.g NaziPartyUSA.com) and put your hate speech at that URL. Same for any ISIS group that wants to infect commercial websites like Facebook/HN with recruitment propoganda to attract followers who are willing to behead infidels.

The reason free speech advocates don't like that option is that they don't get an audience -- such as the millions of eyeballs at Facebook. Well, free speech doesn't guarantee an audience. The quantity of the audience is orthogonal to the free speech and the corporate censorship.

I don't like it because "public areas" are owned and governed by private entities. It's nonsense to say you're in public if there's no public listening. That's not free speech, it's free speech zones. As in, "Say whatever you want, just say it to yourself in that shed over there."

> Nazi-proponents would use it as a weapon to infect...

I think Nazis are dispicable but quantitatively not a big enough problem to be worth the cost of genuine public discussion and debate. A group idiots the size of Raton, New Mexico running around in tiki torches in Charlottesville is sensational, but not really our biggest issue, even if we limit the discussion to racism and hatred.

If trolls show up on message boards, the solution is curation, moderation, law enforcement (digital assault should be enforced more), and good counterarguments.

Right now, people are worried about totalitarian government and corporate police punishing them for thought crime. If you asked me a decade ago, I would have called that paranoid, but these days the supporting examples and surveys seem to be piling up.

And those editorial mechanisms mean you no longer have "freedom of speech" -- which breaks the very law you're proposing. That's what I meant by "contradictions" in my previous message. A new law that extends "freedom of speech to private entities" is not possible for society to implement in a consistent manner.

>Right now, people are worried about totalitarian government and corporate police punishing them for thought crime.

Well, maybe we're having 2 different conversations. I'm not talking about situations like China and their Great Firewall suppressing Tiananmen Square, Taiwan independence, and other topics of dissent. That's the _government_ doing the censorship and not _private_ website like HN censoring by deleting posts. That's a different topic from "extending freedom of speech to private entities" that you were talking about.

Instead, I'm talking about websites with unpopular topics like Unz.com which spawned this thread and neo-Nazi website like Dailystormer.ws. If DailyStormer can reject articles about "programming in Lisp", HackerNews should likewise be able to reject posts about "eliminating the Jewish population".

Trying to craft the wording on a new law such that both sides like HN and DailyStormer can post off-topic content on each others' website without censorship is a futile goal for lawmakers.

Right now, in the USA, American citizens can go point their browser url to Dailystormer.ws and read all the articles. There is no Great American Firewall nor ISP blocking access.

B and C are simply exercising their freedom from speech, which is as important as freedom of.

Of course, this has its own problems: like online fora devolving into cliques of people who think the same way.

People should be exposed to trolls and critically evaluate and re-evaluate their beliefs.

They're either curators and editors and somehow responsible for the speech on their sites, it their carriers and just providing a forum.

If Reddit isn't responsible when people post libel or legally actionable threats, then they shouldn't be taking responsibility for policing their content for ideological alignment, much less politeness and courtesy.

I'm saying it's time to explore whether we need common carriers of speech like we have for raw data.

This is a pity. This piece was entertaining and informative as it points to the subtlety of these effects. Nobody necessarily needs to do anything - any algorithm or process that can optimise for any metric and produce this result - humans need not be involved in any way. This problem can only get worse as more automation, i.e. machine learning is applied and where the humans involved in the day to day running of the systems do not have the faintest clue how it works.

I'm not claiming there is zero censorship at the DNS level. I'm saying a marginal group (whether ISIS or Piratebay or Wikileaks) needs to find a TLD (top level domain) willing to hold them. (That's what The Daily Stormer did -- they tried Albania tld ".al", then Russia ".ru", and now it looks like they're at tld ".ws" (Samoa))

Creating a new domain at a tld is much more realistic to implement than passing a new law requiring Facebook/HN to pay for and host your objectionable content. For marginal groups, that DNS-level option to spread your ideas exists today without any new laws.

Companies do not owe it to you to let you promote your views using their infrastructure. Maybe it is in their best interest, or the best interest of their users, or maybe not. But historically, we have valued companies' (and, transitively, company owners') free speech rights too much to force them to promote views as common carriers. There are of course a few exceptions to this where there is a compelling public interest, but you're going to have an uphill battle to prove that in the case of posting certain types of news to Facebook or Google News.

There are other political views I oppose where I would want corporations to allow those views to be expressed on their platforms. For example, I'm opposed to a lot of economic conservativism. But I don't see it as being anti-epistemic, and I don't think corporations should suppress those viewpoints. Fortunately, that's not what is happening. I would be shocked if anyone had ever been banned from Twitter, Facebook, or Google News for promoting these sorts of views.

So the problem here is that all views are political in a sense, but views are not merely political. It's not rightist political views generally that are being suppressed. It is other, more specific types of views that tend, at this point in time, to be associated with the right.

Who said anything about support? Do DNSs "support" every website they host? Why would forums be supporting all the speech they host then?

> Companies do not owe it to you to let you promote your views using their infrastructure.

I don't think of it in terms of obligations. I do think supporting free speech and earnest dialogue is a social responsibility. That's why the first amendment was written to start with. The same negatives seen when government curtails speech also apply when corporations do the same thing.

Yes, in the sense we use in the tech industry where we say we "support" a particular use case -- i.e. enable it/allow it to happen.

> I do think supporting free speech and earnest dialogue is a social responsibility.

I don't. Or, rather, I think it exists in a constellation of other social responsibilities, and is subordinate to some others.

Perhaps there is a solution with large, well-funded libraries or universities where there is a clear mandate for creating an environment that stimulates people's minds. here the value of their brand would be the main attraction. People who we content with mainstream media could still read their personalised content on Google, Facebook, etc. However I think it will be some time before the technology is sufficiently well developed and cheap enough to make this a reality - assuming the usual political hurdles can be overcome or worked around.

Then... you think that the government should compel me to allow any and all voices to use my private property as a megaphone?

I guess that is no different from making Christian bakers attend in support of a gay wedding or forcing Catholic nuns to fund birth-control - so ... crack on.

Imagine Facebook and Google together decided to go completely partisan and support one candidate or another so much as they possibly could. They could present a huge chunk of the world's population with an entirely false reality. That power is something that should be very disconcerting in a democracy. If this were able to be put into effect, and prove workable, companies like Facebook and Google could effectively silently overthrow any democratic government - replacing them with the puppets of their choosing.

But I hesitate to grant any particular national government to do what I will with my property.

I must think on this. Thank you.

Significantly and uniquely, Facebook, Google, etc. are selling forums and mediums for expression. They are common carriers of speech; they are not partnering in creating speech.

If I were Facebook (I am not) ... wouldn't I - shouldn't I - be able to say: "Get off my lawn"?

"Any" and "all" are absolute words like "always" and "never", so no.

I think there's room to create something like network neutrality but for speech. It doesn't need to apply to every single site on the internet, just like network neutrality doesn't apply to every single server connected to the internet.

But there is an analogy between the companies that provide the backbone of the internet and the companies that provide the backbone of public discourse.

> "If I were Facebook..."

Morally and ethically, facebook should not be able to influence public discussion to fit their worldview, including banning thoughts it finds too abhorrent. The (recent, to the internet) push to do so is just a new wave of the things described in the Scarlet Letter, Fahrenheit 451, and 1984. It's getting pushed too far and things will happen people will regret.

Should there be legal protections? At the moment, I think the discussion is worth having. If you asked me ten years ago, I would have said no, but it's apparent to me that something is different this time. I don't know if it's network effects or what, but it seems like a few companies are carriers for the public speech of average Americans and if they all decided that the thoughts of, say, libertarians were dangerous, we could end up with another round of the Hollywood blacklists, but that affect everyone and in much less tangible ways (algorithms that change and cannot testify, lack of objective data to prove suppression happens, etc.).

For one person maybe, but try getting all of society to do the same.

Edit: You do realise we are both using a "digital forum run by a private entity" right now?

Besides, frequency of something isn't a good measure of its importance. "How often do you feel the need to vote?" is also the wrong question to ask.

While I think people are genuinely quite smart on average, I also think most are not particularly well informed. Adblocking is something that directly and immediately provides a vastly better experience. But the majority of people still do not use them and are, for the most part, likely more or less ignorant of their existence or their effect. Now somehow getting people to engage in behavior for an abstract high level change that would, in the short run, result in some negative consequences for themselves (for instance Google search is certainly still the best search engine by shrinking but still wide margin) -- that's something that is going to be very challenging to make happen.

[1] - http://www.businessinsider.com/pagefair-2017-ad-blocking-rep...