Live data from Hacker News

Twitter's Fucked

degoes.net

371–380 of 398 posts

Re: Twitter's Fucked

#371
post #222

Earlier quoted context omitted.

How is Twitter not a good business? It's still worth $11 billion.

It's never made a profit.

This is a good point. So I looked up some Twitter financials.

In 2015 they posted a Net Income of -$521.03M and EBITDA of -$137.21M.

I don't exactly understand how those two numbers can differ so much, but either way they're pretttty bad.

Re: Twitter's Fucked

#372
post #85

Earlier quoted context omitted.

Explain to me how effective blocking individuals is when you're being spammed by dozens and hundreds of people with so little to do that they will happily go create egg accounts to continue harassing you. It's not practical or feasible to block all of them. So people should delete their accounts because they get brigaded? To hell with that, and with the people who think that is a social norm worth protecting. "Crybul…

For that, there is another button: "stop receiving notifications". And another one: "don't allow anybody but the people I follow to spam me with shit". Both are easily activated on Twitter. What other button we need to invent? "Activate our hordes of ML algorithms to make sure that all replies you receive are about kittens or pink ponies"? Can be done, but we already have Facebook for that. For me, all this drama loo…

Do you realize that a big part of Twitter, for a lot of people, is making connections with different groups? I've made multiple friends on Twitter, people I now know and talk to through other channels (as well as Twitter). Even aside from those friends, I have had many conversations with people I don't follow and who don't follow me about issues--interesting ones that, in your weird little world, simply wouldn't happen because instead of fixing the problem of superannuated juveniles who have nothing better to do than blow up your phone with guro and death threats, you offer complete non-solutions to begin with.

But, then. "Crybullies." Res ipsa loquitur.

Re: Twitter's Fucked

#373
post #137
post #85

Earlier quoted context omitted.

Explain to me how effective blocking individuals is when you're being spammed by dozens and hundreds of people with so little to do that they will happily go create egg accounts to continue harassing you. It's not practical or feasible to block all of them. So people should delete their accounts because they get brigaded? To hell with that, and with the people who think that is a social norm worth protecting. "Crybul…

There are social networks (Facebook, Tumblr) where you can turn off the ability to be messaged by people you don't friend/mutually follow. Getting attacked by trolls? Shut it down, and enjoy the silence.

As I mentioned in a sibling comment, a big part of Twitter for me is meeting new people. Not only have I had good, thought-provoking conversations on Twitter (for all the 140-characters jokes I don't find it that difficult) but I've made friends who are now RL, in-person, at-the-bar-on-Fridays friends there. That doesn't happen when you're blocking external folks in that way, and that's a bad tradeoff to make just because "well, those people are dicks, we can't control them, emojishrug!".

Re: Twitter's Fucked

#374

Earlier quoted context omitted.

I hear and respect what you're saying, but I don't think you can technologically fix a social problem. Twitter needs moderation , and I don't think that's practically accomplished with a purely technological solution. Assholes simply don't have a right to be there, and Twitter has a responsibility to its userbase to not subject them to it at all, rather than merely "well, machine learning" to a merely intermittently…

FWIW, even attempting technological solutions sends a signal of a sort. It's probably a "Red Queen's Race" to stay ahead of determined abusers, just as it's difficult to conquer spam, but one side-effect of making a concerted effort to try is that Twitter would be sending an unambiguous message about limits on behaviour.

This is a good point. That some movement is happening is, unambiguously, a good thing. It demonstrates at least some recognition of the situation. Whether it's the right movement, or whether Twitter will consider it to be "enough" and stop, is still an open question.

Re: Twitter's Fucked

#375
post #311

Earlier quoted context omitted.

Instagram, WhatsApp, WeChat, Line?

Up until earlier this year, WhatsApp actually had a for-pay business model: it cost $1 per year. They stopped charging in January, evidently, in theory to remove a barrier to entry. (Being owned by Facebook now probably helps them make that move, though.) I've wondered how Twitter would have been different if they'd taken a similar approach: $1 per account per year. They'd certainly be smaller than they are now, but…

The problem is that people treat free very differently from very cheap. Rationally, they shouldn't, but they do. So I suspect the number of users with the dollar-a-year plan wouldn't just be lower, it would be really dramatically lower. An order of magnitude, probably. Two, even.

Re: Twitter's Fucked

#376

Earlier quoted context omitted.

If your beliefs change all the time, they can't have been held very strongly or justifiably in the first place. Maybe you should stop having beliefs like that. That's direct evidence that you shouldn't be confident in many of your beliefs, given they change so often. A person with stable beliefs, even if they aren't ever challenged, has no such direct evidence to suggest they should be less confident, they can only h…

> A person with stable beliefs, even if they aren't ever challenged, has no such direct evidence to suggest they should be less confident That's like saying someone who comes up with a bunch of scientific hypotheses and never tests them should have more confidence in them than someone who does. Holding on to your beliefs even in the face of evidence/arguments to the contrary is not a sign of having better or more con…

You cut off the rest of my sentence. It's not that never having your beliefs challenged is good, it's that it's not a strong signal that your beliefs are actually wrong, and so your confidence can't actually diminish much from pointing out this fact, because everyone should already account for this in their confidence factors for beliefs they haven't actively tried challenging. Whereas if you're frequently changing your mind, well, that is a strong signal that your beliefs are frequently wrong, or that you're interpreting evidence poorly, either way that is a signal you should be even less confident in your beliefs than general. If you're frequently testing your beliefs, then it's the results of the tests, not the fact that you're testing, plus your ability to have your beliefs appropriately shifted by the results of the tests, which determine how good your beliefs are. The latter half of that equation is probably the bigger factor for what causes wrong beliefs anyway. Ideally your confidence in the belief should match its actual probability distribution, and shifts should be in correspondence with Bayes' rule. In reality many people will see a known-to-be-uniformly-independently-random binary process output a 'run' of Trues and feel more than 50% sure the next hit has to be False because it's 'due'. (https://en.wikipedia.org/wiki/Gambler%27s_fallacy)

Re: Twitter's Fucked

#378
I wonder how much of root of this can be reduced to election year political nonsense. A year from now, it's going to be much less annoying to be on the Internet.

The author is right that the product has some major flaws, especially around harassment, but no one else seems to want to take over the space. And Facebook, Tumblr, G+, etc. are all much worse platforms for the kind of discussion that happens on Twitter.

So I think the toxic environment can definitely put a ceiling on the number of new users, so does the timeline issues, but it could probably just keep plodding along for another 5+ years as-is until these issues are worked out or something much better is put in place (not likely, but possible, IMO).

Re: Twitter's Fucked

#379
post #365

Earlier quoted context omitted.

> Nothing in my comment as it stands now is false. Nothing in my responses to you are false. When I noticed something incorrect, I modified it, noted it, and accepted responsibility. But you didn't go back in time and read the article before your comment. That's the point. > The fact that my comment is similar to the article doesn't make it false. The fact that I didn't notice the similarity immediately doesn't make…

Please stop. Personal attacks are not OK on Hacker News, and we ban accounts that do so repeatedly.

Are you a bot?
Post reply on HN