I wonder what Reddit's current shareholders think of that idea?

They are the ones driving this fight. The venture capital invested in Reddit has long soured and everyone is looking for the exit, and that demands an IPO at a good enough price to make whole, and preventing the most engaged users from bypassing ads is a prerequisite for a reasonable IPO valuation.

> decisions are generally a lot less arbitrary than people imagine. It's just
> that when someone gets suspended they tend to only post misleading
> reports about it

Strong disagree. Twitter's rules don't accomplish what Twitter says they accomplish, and the enforcement is uneven, arbitrary, unreasonable, and utterly opaque to users. I've seen countless examples of utterly arbitrary enforcement, and regularly repeated claims that perfectly normal content is somehow a violation. Not only in my case, but in many, many cases.
In this case, a woman was gleefully discussion the fact that human face mites do, in fact, have anuses, and therefore our faces are covered in face mite poo. The conversation was already about this topic - merely suggesting that we affix tiny diapers to them so we can study the poo is not hostile in anyway. This person didn't understand or like the suggestion and instead of muting the tweet or the tweet author, they used Twitter's rules and the company's pro-complainant bias to remove my ability to use the service. That's neither just nor reasonable nor sensible.
I'd suggest examining your instinct to defend the giant faceless corporation over real people, but figure that'd be a waste.

I'll take over and fix Twitter if the people there finally acknowledge that suggesting that we affix tiiiny little diapers to human face mites is not "hostile conduct" and unlock my account.

Adobe's neural filters are not 'deepfake tools,' they're tools. Blaming the makers of a general purpose tool (of any kind) for specific uses that tool is put to is pretty weak sauce. This simply is not a case of someone making a tool specifically designed for a nefarious purpose, and CEO's assertion is essentially correct.

So much anger, so little understanding of the world.

edgy

Shit, Stang don't even answer my calls anymore. I remember 1993, Halloween, Dead Or Alive, Ivan and I were front row. Now? not even a friendly text. :(

Astronomical observations don't always need to be reported as "mysterious," tbh.

I'll stay with Windows 7 for a while longer. But I won't be installing Windows 10 willingly, if ever.

Microsoft: "We let you choose how much spying we do on your activities!"
Users: "Uh...please don't spy on us at all. Like, none."
Microsoft: "OK, here's some revised settings that don't let you actually turn data collection off! One is called 'Basic'!"
Users: "..."

You're mixing terms here.

Oversampling is always divided out.

What you're accusing polls of is having a turnout model which over favored Democrats by up to 15%. That's a completely different polling error and there is no reason, nor evidence, that it was intentional.

Maybe it's because the plan was the skew the polls to suppress the GOP vote by oversampling Democrats?

Minorities (not Democrats) are oversampled because they are minorities.

But oversampling doesn't mean what you're implying. It means if you are trying to accurately track the behavior of a minority group and only have the budget for a limited (often times in the low hundreds) data set you MUST poll more minorities constituents than otherwise would come up, or else (as happened in the LA times tracking poll) a single outlier participant can skew the results for all.

So you poll more people than their share of the electorate, BUT THEN YOU DIVIDE BY THE OVERSAMPLE FACTOR. This is statistics 101.

Once. He closed the gap once.

The convention bounce closed it the other time, not him.

Ad spending in the US was $200billion in 2016, for a US labor force of 160mil. That's basically an annual $1200 tax on everyone for the privilege of having ads shoved in their faces.

Another, I would argue more representative, way of putting things is that advertising costs ~1% of GDP.

One could view that as a small price to pay for consumer lubrication.