Jim:
>>Want to post how much you hate (to pick an example) mexicans on twitter? Go for it, it's unlikely you'll be banned.
>
> Are people who post how much they hate Trump, or hate people who support him, going to be banned? Somehow, I don't think so. It's PC nonsense. Selective enforcement.
Well, no. Just like you're not going to be for posting that you hate mexicans, which is kind of my point.
> Oh, I see! Posting such statements are "actions", not "speech".
Well, if that trick were used, people could be prosecuted for uttering sounds that happen to sound like "hate speech". See the problem? Of course you don't.
I think you're being intentionally dense here.
This isn't people tweeting "build a wall", it's throwing abuse at a specific individual.
> If the speech involved was objected to by 99% of the population, or even 95%, you might have a point. but I suspect you are really only referring to the opinions objected to by (at most) 51% of the population, or in fact far less. (Say, speech objected to by 75% of the PC crowd, or maybe 10% of the population.)
There's no way we can objectively measure this, but I think your figures are wrong.
49% may not object to you tweeting "build a wall", but I think a much smaller percentage would be happy with you (and others) racially abusing an individual user (to pick an example). That's not "being PC", that's thinking you're a prick for continually tweeting coon at a single user just to see how they'll respond
>> "Reddit had to"
>
> HAD TO? Really? Are you absolutely sure about that? Like, somehow, their computer servers would melt down, or something, if HTML data with certain data patterns were stored in it?
Well, yes. They had to, in that they felt they couldn't attract the revenue they wanted to in the current state. Businesses, funnily enough, want to make money.
> "descend, en masse, on someone else"??? You mean, like, physically attack them? Or do you merely mean using speech that offends them? If the former, I don't see much of a problem. If it is the latter, I see a huge problem.
It is a little tricky to physically attack someone whilst on the internet. But there are situations online that we consider worthy of the term "attack". A DDoS Attack, for example, doesn't do any physical harm and yet is still considered an attack. Why? Because it's a targeted and deliberate attempt to interfere with operations.
If you're receiving thousands of abusive tweets sent specifically at you, and/or those same people are filling your mailbox with the same, I'd call that an attack. YMMV
> Private property is different. Twitter MIGHT justify discrimination based on the fact it's privately-owned. However, that's a very imperfect defense. First, Twitter hasn't previously claimed that it has an institutional bias.
You don't need to have previously expressed a public opinion on something to suddenly find you don't agree with it.
Corporations (and media) aren't required to explicitly state whether they're left or right leaning (which in some ways is a shame)
> Further, imagine a different service, one that presents itself as being Twitter-like, but which openly engages in what we'd agree is hateful, racist speech, and discriminates in employment. If a person could use the court system to try to shut them down, then it isn't purely a "private" operation, is it? It's subject to government regulation and control. In that case, I think that Twitter itself (following arguendo, the same logic) should be subject to legal attack.
Now that's a complex area of law, and varies between countries. But, those who were banned would (AFAIK) be perfectly at liberty to try and sue Twitter.
Strange though, whenever something like this comes up, you sometimes hear threats of legal action, but they're never followed through. It's almost as if the complainers haven't got a case under law. Admittedly, it might well be, of course, that the cards are unfairly stacked against them.
> Except you forget: Up to now, Twitter hasn't identified itself as a "left-wing" organization. Now, apparently, it has.
See earlier comment - there's no requirement for them to have done so.
And just to be a little picky. In the rest of the world, Twitter are not a left-wing organisation, nor are most of the US's "lefty" politicians much further left than centre-right. But I take your point
Juan:
> The difference between your previous scenarios and the twitter thing is that if there were any threats, I doubt they were credible at all.
That's a valid point (though I did say analogies were flawed). Receiving those kind of threats is undoubtedly unpleasant, but you're right, they shouldn't lead to quite the same fear for safety as in my scenario.
On the other hand, I could understand someone fearing that they were just the beginning. The person sending them might not, physically, be able to reach you themselves, but the next step might be swatting. But, to be fair, at that point we're away from speech and well into real-world actions.
> I mean, what kind of threat is an anonymous idiot with a fake account half accross the world saying he'll kill you? Your own account probably not giving any real info about you either.
It's scary the amount of information people do put online, though most at least have the sense to not publicly post their address. That's often not enough, though. Both corporations and governments seem to take a laid back approach to protecting the data they're given (and in the case of govt, the data they compel us to give them), so with various leaks floating about there's always the possibility that information can be gathered. Again, though, I'll conceded that that's a seperate issue
> Of course for those people tha ability to block anybody who called them out was essential.
There's definitely groups of people (some of those Jim would probably correctly identify as "PC") who seem to revel in the ability to block dissenting views. An individual blocking someone else, at most, is only harming their ability to learn something about the opposing view. But others do try to take it further by trying to have "it" blocked at a wider level by coercing government (or private orgs) to act.