CNN photoshopped a “Verified” tag next to my name (2017) | Hacker News Reader

CNN photoshopped a “Verified” tag next to my name (2017)

twitter.com

The twitter thread shows a plausible reason. No photoshop involved.

Pray tell, what you see as the plausible reason is, so we don't have to sift through all the noise?

CNN tweeted that an automated system identified them as verified by mistake

Whose system? CNN's?

The claim is that CNN forged the verified mark in the supposed screenshot. "But they thought they should" is not a defense.

Say we go Hanlons razor and not malice: they used a Twitter API which draws the markup, and the API identified them as certified. No hand on the tiller, just a bug

"Forged" goes to intentionality. They've said they didn't mean to do it.

Maybe this is just a corner fight about linguistics.

Hmm, so are you saying that Twitter API used to render these shows different info than the Twitter UI at twitter.com?

Yes. Through third party wrappers which do the render, and a bust cache or some other stuff up which misattributed a flag.

What I don't think they did is screen grab a tweet and ask their photoshop guy to add a blue tick. À

It seems they have some system which generates the markup of a tweet from the pure content. I honestly think that's pretty cool, but apparently it had a bug. Forgery implies intent.

And there are some missing steps between "code got a bool wrong" and "tweet shown with verified mark"

This implies that they don't just screenshot the tweet, they somehow generate it? Some kind of automated "photoshopping", you might say.

Yep, what I get from the explanation is that they don't do a manual screenshot but a software pulls the tweet from Twitter's API and generates it.

If I had to guess, to be consistent with other screenshots, match TV resolution and shape, transparency or backgrounds, and maybe have a trail log of how and when that screenshot was generated.

I obviously don' know if this is what's happening, but a friend of mine works for a company that makes software similar to this, but for live video broadcasts.

I assume big media companies either use such products, or maybe even have in-house solutions.

At least to me (and maybe I'm just weird), to say that they "photoshopped" it implies that they modified the real image. It's a little weird that they reconstruct the appearance of a tweet, but not that weird! It's a consistent style, and doesn't leave any risk of artifacts. Plus you can presumably plug it right into your production system. Like CD for tweets lol

Link to said message?

Looks off axis and done by hand. Does it look the same on other screenshots of real verified users?

Reply on news.ycombinator.com