Hacker News Reader

> This only follows if you believe you're under an obligation to use social media.

Were the town residents in Marsh v Alabama, required to inhabit the company town? I think the fact that they did is what made is a public space, regardless of whether they were required to.

I don't think that's quite right. The issue there is that the company town wanted to be an actual town- they were providing services, inviting people to move in (including people who didn't directly work for them), and generally saying "anyone who wants to can be here"- they were very explicitly trying to create "public spaces".

Facebook is not the same thing- people aren't moving there and people don't live there. I know we want to pretend that digital and physical are similar, but this particular case that everyone is citing was very specifically about the company towns and precedent afterwords has been pretty clear.

> What public space offers infinitely scalable many-to-many communication? Public space is distance limited, these platforms aren't. Just because these platforms chose not to heavily regulate the content doesn't preclude their ability to ever do so in the future, nor does it make a comparison to a theoretical public space that never existed any more valid.

You're still extremely limited on Facebook and Twitter by natural space and time constraints. Just because these take a slightly different form and aren't based on physical distance between communicating parties doesn't mean that communication on the platform is just completely unfettered.

So "infinitely scalable" does not mean that each person can actually consume all content produced by all other people. That's not possible in all current iterations of social media and presumably is fundamentally unachievable. Rather, it means the system can scale to an arbitrary number of producers and consumers, each of whom is still subjected to natural physical constraints, whether those be determined by physical location, screen space, the loudness of your voice, the speed at which you read, the speed at which you type, etc.

I believe that with network effects far more pervasive as things move online, monopoly/competition law needs to play a greater role. The basic idea is that any party that controls more than (TBD) percent of a significant market/distribution channel/medium/etc. (significance by total $, total # of users, etc., TBD) loses the right of a small, private party to unilaterally ban users.

If you create a FaceBook competitor, identical in every way except dozens of other users instead of hundreds of millions, it would not be a realistic alternative. Likewise for Twitter, or real estate listing services, and a growing number of other things where it's not the tech so much as the other users.

When network effects dominate, as is so common online, and users gradually lose the realistic ability to just switch to a different service with terms they like more, service providers must also lose the ability to impose any terms THEY like and to refuse service to users. What they can charge, who they can ban, etc., should become matters governed by regulators.

There's no objective, consistent way to define a "significant market" for highly differentiated products. What market is Facebook in? Is it all web sites? All mobile apps? All media? All social media? All communication tools? Those categories are totally subjective and arbitrary.

As someone that refuses to use Facebook / etc... I still see the free speech issue in this.

Imagine it from this perspective; whom can a telephone company decide to deny service?

If they have stepped beyond merely disagreeable speech (and I agree, it's often very disagreeable) where was that line crossed so that our court system might process their transgressions?

Telephony is a commodity service with a legally defined market. The rules for such services are irrelevant when deciding how to regulate non-commodity Internet services.

Legal status is frequently based on reasonableness. When you have a service with hundreds of millions of users, you already have a reasonable argument for "significant" even if the lower bound of significance is fuzzy. If you then survey the users of a giant system regarding practical alternatives, and the answer is overwhelmingly "even if there were one, I couldn't switch to an alternative offering the same features and price, because it would lack the critically important feature that everyone uses this one," you have a reasonable, practical, realistic means of identifying some of the entities that are too powerful, relative to their users, to be allowed to just do whatever they want with prices, who they will do business with, and any number of other policies.

If you want complete freedom to refuse to do business with users for whatever reason you want, charge anyone whatever you want and change as you like, and so on, you can have it as long as your users can reasonably exercise similar freedoms by doing business with alternatives. If network effects remove those options from your users, they should remove them from you, too.