Hacker News Reader

Comment by mlmonkey | Hacker News Reader

Unfortunately, it requires 3 clicks ("account" -> "join date" -> "about account") to get the required information.

It should be visible on the post itself: where it was posted from, and where the author's account is mostly located. Just 2 little chips stuck on top of each and every post.

View on HN

It wouldn't be difficult to include a little "" icon on each post that, when hovered, shows probable location or the creation origin. It's true that VPN's will often muddy this info, and even things like phone number confirmation for 2FA or billing address for Verified badges can't be 100% reliable... but it shouldn't be too difficult to say in those cases "User likely masking location" if the pattern is detected.

It's not about being easy it is about being profitable. I don't care where you fall politically, accounts like this increase engagement and thus profit. Companies like profit so this type of thing despite being much easier than some might assert is going to be slow to do. You can have legal requirements around identification for accounts over 2K followers. It basically would never inconvience people unless they are trying to be an influencer

The largest effect is that X suddenly (out of nowhere?) started to mention this, publicly. Which, given mr. Musk's political stance (who himself has been the target of a campaign, likely from mr. Putin), wasn't intended to have this effect.

If you were to share this information publicly from the start, the troll/propagandist/agent would know this information is being shared as-is from the start, and they could adapt. Very likely they would adapt. Therefore, the effect would be much less severe.

It is also very obviously PII, which would hurt innocent people (likely a minority but I can imagine it happening).

My take: intelligence agencies were already interested in said information and mr. Musk was like "whatever, we'll just share this with everyone".

This is exactly what happened.

I mean, it is just my theory. I don't have a source to back it up.

Yes, X is the new /pol.

How many of these accounts are Russian? I bet hardly any. My money is mostly from a small country in the middle east.

How much money are we talking?

russians been interfering with other countries through all possible channels for decades.

They integrated with every single Ukrainian election (including the most recent one in 2019), they tried hijacking (with various success) elections in Romania, Moldova, Lithuania, Czech Republic, Estonia, etc..

This includes radio, tv, printed periodics, social networks, even popular music, movies, tv series and books...

So how much are you willing to bet that this time it's definitely not russians?

Yeah looks like the data proves im right. how much you want to wager again?

shitty little mena countries suck.

What data?

Followed immediately by these groups making VPNs a standard part of their toolkit. This data is enlightening but its reveal doesn't solve the problem.

Twitter can require verification on their end or require ID upload or something along those lines. The social media companies can solve this problem, it’s just way too profitable to not. They make a lot of money when you read about non-existent events from an account curated in China or India or Belarus and then go look at a bunch of ads for Patriot Bars (R) or Rainbow socks (L) or whatever.

They could by default for example require an identification of some sort, and then allow “non-ID” accounts to exist but require specific opt-in to view broadly or something along those lines.

More easily though you can just delete your account and then you don’t have to care about any of this crap.

There's already pretty sophisticated setups to get fake live ID verifications where it's already a cat and mouse game between the tools to verify and the tools faking live verifications (sometimes including just scamming people into acting as the 'user' for verification). Ideally I'd also not have to provide my ID for yet another inane service and risk that ID getting leaked as seems to be inevitable.

Yea that’s fair. I guess the easy/best solution is to just not use the product in that case.

But I’d be in favor of even an in-person verification system though the costs to do that would be unpalatable but I guess you could stand up a 3rd party to do that. Maybe there are better solutions out there that I’m not thinking of. I do know it’s very much against what the larger social media companies would want though because they actively want you to be outraged and misinformed since they make a lot of money off of that.

I wish they had included it as part of the RealID system. It would have been pretty easy to make them smart cards that could use simple public private key tech to verify IDs. Private keys go into the cards and public keys get registered with the state agency issuing the IDs. It's not perfect but it'd at least be a pseudonymous way to verify ID possession.

I've recently (last several years) started framing my use/consideration of services and products as a dependency. I find myself arriving at "I'd rather go without than depend on xyz" quite often.

It seems like they are using a more sophisticated way to determine location (App Store download county, etc) not just IP.

Twitter reveals when it thinks users are using a VPN and tries to guess the origin location regardless, though with quite low accuracy

yeah it needs to be robust enough to keep a trail so it can detect leaks and predict, and perhaps eventually identify vpn nodes, like Reddit will often block my request when I’m on VPN so there may be a way to smartly disregard certain requests from being counted as the true location, this seems really complicated the more I think about it though

I built a little Chrome extension that shows the flag for where the account is based in on the profile page itself: https://grokify.app.

I'm thinking of adding it next to the Tweet source when you focus a Tweet:

https://twitter.com/ControlPanelFT/status/199301008332080747...

In the old days 'where it was posted from' used to be displayed on tweets, and Elon intentionally changed it to 'Earth'. Interesting to see a backpedal here, albeit concealed.

This is incorrect.

The old Twitter and its API had "source" field which showed the client app a tweet was sent from.

{ "created_at": "Mon Nov 20 12:34:56 +0000 2023", "id_str": "1726578932412345678", "full_text": "Example tweet", "source": "<a href=\"http://twitter.com/download/iphone\" rel=\"nofollow\">Twitter for iPhone</a>", ... }

If you had access to the twitter feed (and I did at one point, as our company got the firehose from Twitter), it showed the Lat/Long of where it was posted from IIRC.

I still remember looking at that tweet when the helos went in for OBL, from someone in Abottabad PK, saying something like "helicopters hovering at 2AM... this is a rare occurrence" (or something like that).

You can still opt in today to tag your location in a post, that never changed.

It's still there, they just stopped showing it in the UI. If you poke about in the React state of the Twitter UI, you can see it in the `tweets` entity cache (everything else is still Twitter under the hood too).

Didn't that show the client the user posted from, not their location? Which was mostly useless anyway after they all but killed the API.

they really should. i thought they'd have a flag on the actual profile.

Reply on news.ycombinator.com