Tangentially related, Musk now wants NPR to tweet again, under the threat of giving away NPR's handle for disuse, contrary to Twitter's inactive account guidelines.
To preempt "musk hate" arguments, I'm attributing this to Musk and not Twitter because the emails were sent directly from Musk to NPR.
This does not sound like a reasonable way to build trust between brands and Twitter.
It truly boggles the mind why Twitter/Musk/wtv is so aggressively attacking the goodwill of advertisers and brands with this whole blue check and state/public/government media debacle.
I’m pretty sure Musk is not a good manager, but he can select good operators for his vision. Tesla and SpaceX’s successes I believe at this point should be attributed to their other executives and leadership, with Musk just being the marketer for a bold idea and not the actual executor of that vision. Musk is hands on with Twitter and it is a disaster.
Twitter was managed well before? They were burning money on useless people like no tomorrow, hardly rolled out any new useful features and spent all their time policing the platform.
Whatever Musk is it's certainly not worse off then before. NPR threw a tantrum because of a label that wasn't the same as what Chinese and Russian media had, which they disingenuously claimed was the same. Then they threw a tantrum when Musk in response removed all labels from ALL media organizations.
Musk is certainly not building trust with the advertisers, but neither are the advertisers acting in good faith. It was also quite the spectacle to see all those blue checkmarks flip out at losing something they got for free to the extent that they went around saying that peoples lives would end because of it.
Fun fact: You can think the current owner and manager of Twitter is insane and running Twitter in an insane way, without also simultaneously thinking the previous managers were running things in a perfect way without any errors or issues.
Without a comparison, then the only conclusion is that Twitter simply cannot be run "correctly" by anybody, given that everyone has a vision of how it should be.
That's a false dichotomy. Saying that neither of two examples is perfect does not mean that one was not clearly better, and does not mean that the task is impossible.
The only thing Musk has achieved at twitter is accelerating the exodus of users from the platform, and the fracturing of the "microblogging" space. If things continue as they are now, there won't be much of value left at Twitter.
Billionaires are unreasonable. The brand is built. He is not interested in brand building like a startup. He is interested in removing "tyrannical" cruft.
The obvious response to this is malicious compliance. OK, every week or every day NPR tweets something to the effect of "to get the latest NPR news, go to [URL]" where URL is a page on NPR's site with its preferred social media outlets.
No news. No updates. Nothing that would help engagement on Twitter, but the account is active. It tweets.
Of course, Musk will try to force them to engage with Twitter as if it's business as usual and try to invent some other reason to either reap the account if they don't comply.
When dealing with tyrants and those who feel they are above the rules, try to use a gentle voice and carry a big stick.
NPR reporter should have replied back that NPR is held to the terms of service agreement and giving away a handle can only happen when both parties agree to a monetary exchange. Musk would probably change the TOS but without NPR using the website again, New TOS would not apply.
Since I recently tried Nostr, I am much less worried what goes on with Twitter and Co. Nostr is crazy cool and feels a lot like it could become the future of publishing.
I basically just started posting on the first random client I found on Google (iris.to) and ... it works:
No signup, no decision to make which host to use, no nothing. Just save the private key (which iris.to creates for you) and you are good to go forever. The content you post gets relayed everywhere:
So it does not matter if the client you used changes their policies, their algorithm, deplatforms you, goes down or whatever. You can just use any other client and keep posting.
And since the protocol is using websockets, one can write a client in a static html page without any server backend. And it needs surprisingly little code to do so. I really enjoyed this video of a guy writing a Nostr client in a simple html file with like 50 lines of code or so:
A massive usability hurdle for any hope of mainstream adoption.
People say Mastodon is hard to join because you have to pick a server. It’s on another level if you also have to pick a service to hold your private key, and you can lose your account if you don’t do this.
> A massive usability hurdle for any hope of mainstream adoption.
Compared to managing passwords, screwing around with authorization codes sent by email and/or text message, and all the other folderol associated with that model?
All the folderol exists so that ordinary people don’t lose their accounts permanently when they inevitably forget a password.
With just a private key, no such protection. Unless you take active steps to store the key safely, you’re a few Backspace key presses away from losing your account forever (or at the other end, having it stolen if you leave the key lying around).
I just marked the key with my mouse, pressed ctrl+c, went to a text file, pressed ctrl+v and saved it.
If someone is not capable of doing that, I am not sure I want to read their articles.
Not sure what you are referring to with "pick a service to hold your private key". The whole point of Nostr is that you don't need any service to hold anything for you.
It's actually a little more complex than that. Many people go through computers several times between moving to each new social network. You have to also A) print and store the key (yikes), B) make sure you move the file to the new computer (easy now with migration assistance) or C) make sure you are backing up your computer (a lot less people use backups than should).
How is that different from passwords, exactly? Okay, maybe you can remember the passwords for services you use on a daily basis, but I must have a couple of hundred passwords to deal with, at least. Password managers can help with the data overload, but only at the cost of introducing the same single point of failure as a private key, and arguably a less-secure one.
It’s not different from just a password. That’s why no service uses an unrecoverable password as the only authentication mechanism because it would be a terrible usability and security problem.
Ordinary people expect to be able to reset forgotten passwords. And because people are bad at managing passwords, reasonable services try to get them to use 2FA so their accounts wouldn’t get stolen. A service that uses plain private keys has both problems.
In addition to the other points, "regular users" unfortunately still quite often use only a few passwords, shared between all their services. So only 3-4 passwords (and variations) to remember!
> Not sure what you are referring to with "pick a service to hold your private key". The whole point of Nostr is that you don't need any service to hold anything for you.
Serious question: Didn't the cryptocurrency people say the same thing? Look how that went -- the tech nerds tended to do it "right," and the non-tech-nerds tended to use exchanges-or-similar that were a bit less than stable, to say the least. Am I misunderstanding?
Your comment reads exactly inverse of the original dropbox comment, sure you can Ctrl+c and ctrl+v in a text file. How are you securing it and why, what does it do. Why do I have to open a text editor, to create an account.
Sure I can rsync or host my files on sftp, but there is a reason why Dropbox got more popular than any of those among avg users.
Post a couple of times that the account is not to be trusted anymore and that people should check your HN account / Twitter / FB / Website to see your new account.
In contrast to Twitter & Co, a compromised account does not mean you cannot post anymore.
I am kind of surprised there isn't a media company with a twitter clone.
We're your one stop shopping for information. Want to know what Jones bought today? Here's a picture they uploaded (to their feed or w/e)! Want to know what the weather is? Here's an account for NYC weather!
It's a bit risky slotting ads next to random people's content but you'll know which feeds are very commercialable. The weather feed is always going to be SFW, etc. Plus you can always use unfilled ads slots with ads for your own content (just focus on ads for the lower tier-content to avoid associating Greys Anatomy with some beheading video).
> I am kind of surprised there isn't a media company with a twitter clone.
The same reason there isn't a traditional publisher with an Amazon clone. One-stop shopping does win, but very few traditional media companies would willingly distribute content for their competitors (probably you could replace "very few" with "none"). Even if they were willing to do that, the antitrust people would be on their butts in short order.
IIUC, you're referring to Notes [1]. I'm not familiar with it but mostly yes I think so although I was thinking more of a legacy media company from the one thousands.
Read the article, it's quite enlightening. Russia today's follower count jump...
My take:
This happens just as another Tesla factory in China opens. Anyone who thought buying Twitter (now X corp) is about some free speech absolutism lives in a fantasy world. Musk has Teslas to sell (and make). CCP can single-handedly take away the 2nd biggest market and expropriate all IP from factories, and they know he owns Twitter.
Your comment is a straw man. They were labeled as state media but not censored, anyone could follow and read them. The difference is that now their tweets are actively fed to people.
Also, nah, free speech absolutism starts with removing algorithmic feed and stopping profiting from ads. By not having the guts to do that Musk already proved himself not a free speech absolutist. This incident is just icing on the cake.
But still I have to ask, those countries that themselves do not have free speech and do not allow their own citizens to read Twitter but spend big bucks on information war against the west? You see amplifying their content and giving it the same weight as regular personal accounts as free speech absolutism?
Imagine you removed all rules on people behavior and make it free for all, except some people have bare hands and a knife while others have tanks and rocket launchers. Ask yourself, is it absolute freedom? Freedom for whom exactly?..
you claiming that government accounts are amplified is a strawman
> free speech absolutism starts with removing algorithmic feed and stopping profiting from ads
if you can't make money off it, the platform will cease to exist. its cherry picking to say that something isn't perfect because of this when its the same as everything else on the internet.
> if you can't make money off it, the platform will cease to exist
charge for it, if it's good people will pay
> you claiming that government accounts are amplified is a strawman
no, read the article, it is about that
> Platform Use Guidelines on state-affiliated accounts stated, “In the case of state-affiliated media entities, Twitter will not recommend or amplify accounts or their Tweets with these labels to people.” By 23:39 GMT that evening, however, Twitter had removed this sentence from its platform policy
Unrelated to the content of the article, the effect on the left sidebar (white text on green, blending in diagonally with the same text green on white as you scroll) looks visually interesting.
> blending in diagonally with the same text green on white as you scroll
I took a quick look - initially I was thinking they'd be doing something smart, like using CSS blending filters or a text-mask technique as to avoid duplicating HTML markup - but alas, no: duplicating HTM is exactly what they're doing: there's two copies of the sidebar HTML (`div.gta-article--scroller--top` and `div.gta-article--scroller--bottom`), the one with white-text is z-ordered above the scrolling gradient, the other (with green text) is z-ordered behind it.
I'm surprised at the extent people still compromise their HTML for CSS effects when it really isn't necessary - the curse of divitis still exists :/
The other way of looking at it is, this is the brute-force way of doing it that works and is probably reliable on lots of browsers.
Trying to do it in "proper" CSS could be fiddly to get right, and may or may not work the same in all browsers. (I don't know off-hand about this particular effect, but I've done some work with CSS gradients recently and they really don't compose well with other features.)
yep. why spend more time to achieve the same effect? as long as you've marked the elements as "invisible" so a screen reader wont read them both out, and it works, what really matters.
of all things, the extra few divs and css isn't gonna cause issues here.
The largest change shown on those graphs is a gain of about ten thousand followers, the smallest shown is a couple hundred. Without more data, it kinda feels like this could be natural fluctuations.
This website is in itself state-controlled, if only on a second degree (or 3rd degree, if one wants to get technical and not call the Atlantic Council a state-controlled entity):
> The Digital Forensic Research Lab (DFRLab) at the Atlantic Council
what is the problem with this. State controlled media are useful accounts to follow, they will tell you the policy position of the governments in question. If they gain followers presumably this is one the volition of those who choose to follow, free competition for ideas and all that. Suppressing accounts is preferring comfortable ignorance over discomforting awareness.
State-controlled media are used to publish disinformation and misinformation. There is no "free competition for ideas" when certain actors will blatantly lie; removing warning labels guarantees that users will stumble across demonstrably false information from an unreliable source and falsely believe it credible.
agree with your assessment - this is still a competition, perhaps less for ideas, more for 'narrative dominance'. US state controlled media has an extraordinary advantage in this area, I see no reason why it should require a platforms additional support to win this contest. It's like prime Mike Tyson needing the referee to hobble the guy in the other corner - totally unncessary!
the twitter community notes are amazing at handling disinformation and misinformation. better than anything else on the internet that i've seen. its open source and it requires multiple different echo chambers to agree in order for it to come into effect.
multiple politicians have already been caught lying on the spot and it points it out. any individuals, corporations, or state actors trying to do the same thing will face the same consequences.
personally, i care a lot about receiving and working off truthful and factual information. i can't think of a better platform for all these assholes to have to communicate through.
Bluesky is destined to fail for the same reason every other attempt has failed, it only attracts a certain crowd, which ruins the point of the platform. Twitter is full of diverse thoughts and opinions; with news you can't get elsewhere. People also aren't going to dump their millions of followers to have to rebuild their brand elsewhere, so they'll post to both; and unlike with Tiktok, etc where your posts would be unique to that platform, you are just posting the same things you'll post to Twitter. I don't see Bluesky lasting more than a few years. The best bet is that Elon will get bored with Twitter and appoint a CEO that is competent.
its been quite a long time for them to get that fact straight. that they're not just lying, they're gaslighting. this is disinformation, that's the point.
They uploaded sanitized point-in-time snapshot of the algorithm code. It did not include all the information necessary to actually understand how the algorithm works, nor is it an accurate reflection of how the algorithm currently works.
I do not see how saying "While open-source researchers cannot access Twitter’s algorithms *directly*," is disinformation.
That is their complete recommendation algorithm and the repo is regularly updated. Soon they will even be uploading their test suite so you can build and even test the entire recommendation system. This is complete disinformation coming from the Atlantic Council. [1]
To preempt "musk hate" arguments, I'm attributing this to Musk and not Twitter because the emails were sent directly from Musk to NPR.
https://arstechnica.com/tech-policy/2023/05/musk-threatens-t...