tw.tter

[jimbob]

I do have a bluesky invite code if anyone wants

How quickly do new sign-ups get their own invite code? MrsG has signed up using Gfamdottir's invite, so I'll hang on until I can use MrsG's.

A couple of weeks for me.

[Gfamily]

I do have a bluesky invite code if anyone wants

How quickly do new sign-ups get their own invite code? MrsG has signed up using Gfamdottir's invite, so I'll hang on until I can use MrsG's.

A couple of weeks for me.

Cheers.

[jimbob]

How quickly do new sign-ups get their own invite code? MrsG has signed up using Gfamdottir's invite, so I'll hang on until I can use MrsG's.

A couple of weeks for me.

Cheers.

Just use mine, it'll be replaced soon enough anyway

[Gfamily]

Just use mine, it'll be replaced soon enough anyway

I'll have a route in soon enough - so if someone else is looking for one...

[bjn]

Musk showing who he relies on for an ‘unbiased’ source information…

[Grumble]

Just been on Xitter. Was reading a Democracy Docket tweet about gerrymandering and about three replies down it was all photos of improbably proportioned young women. I mean, it’s a pretty dry subject to start bombarding with escort ads or whatever they were.

[dyqik]

Just been on Xitter. Was reading a Democracy Docket tweet about gerrymandering and about three replies down it was all photos of improbably proportioned young women. I mean, it’s a pretty dry subject to start bombarding with escort ads or whatever they were.

Phishing scams, blackmail gathering attempts and attempts to set up advance fee fraud is mostly what they are.

[temptar]

Elon has picked a fight with Thierry Breton. This could be popcorn worthy….

[Woodchopper]

Elon has picked a fight with Thierry Breton. This could be popcorn worthy….

This is not a fight that Musk can win.

[dyqik]

Particularly when coupled with NPR's experiment of stopping Twitter promotion, which has had miniscule effect:

https://indieweb.social/@tchambers/111219082248580986

https://niemanreports.org/articles/npr-twitter-musk/

[Martin Y]

Yaxley-Lennon and Hopkins reinstated, because hate speech is the new small talk.
https://www.bbc.co.uk/news/technology-67331288

As Wilde said, there's only one thing worse than being talked about and that's being talked about by Katie Hopkins.

[Woodchopper]

The European Commission has opened formal proceedings to assess whether X may have breached the Digital Services Act (DSA) in areas linked to risk management, content moderation, dark patterns, advertising transparency and data access for researchers.

https://ec.europa.eu/commission/pressco ... IP_23_6709

Much more at the link.

[Woodchopper]

Fidelity has again marked down the value of its shares in X Holdings, which the mutual fund giant helped Elon Musk buy for $44 billion when the company was known as Twitter.

By the numbers: Fidelity believes that X is worth 71.5% less than at the time of purchase, according to a new disclosure that runs through the end of November 2023 (Fidelity revalues private shares on a one-month lag).

This includes a 10.7% cut during November, during which time Musk told boycotting X advertisers to "go f**k yourself" during an on-stage interview with the New York Times.

In terms of publicly traded comps, Meta stock rose 4.9% in November while Snap shares climbed 38.2%.

https://www.axios.com/2023/12/31/elon-m ... uation-cut

[Brightonian]

I've read that invite codes are no longer required to join Bluesky.

[shpalman]

I've read that invite codes are no longer required to join Bluesky.

I just joined.

[Woodchopper]

Fidelity has again marked down the value of its shares in X Holdings, which the mutual fund giant helped Elon Musk buy for $44 billion when the company was known as Twitter.

By the numbers: Fidelity believes that X is worth 71.5% less than at the time of purchase, according to a new disclosure that runs through the end of November 2023 (Fidelity revalues private shares on a one-month lag).

This includes a 10.7% cut during November, during which time Musk told boycotting X advertisers to "go f**k yourself" during an on-stage interview with the New York Times.

In terms of publicly traded comps, Meta stock rose 4.9% in November while Snap shares climbed 38.2%.

https://www.axios.com/2023/12/31/elon-m ... uation-cut

And they continue to down write the value of the company: https://fortune.com/2024/03/30/fidelity ... -takeover/

[IvanV]

There's been a correspondence in the Economist criticising their repeated use of the phrase "X formerly known as Twitter". The first letter, paraphrased, roughly went "Grow up." A follow-up calculated how many pointless words they had printed, by failing to use a one word name, but did not say which one-word name that correspondent preferred. Another preferred "X which everyone except its narcissistic owner calls Twitter".

The BBC still calls it "X formerly known as Twitter".

Despite haemorrhaging users, Xwitter still has 1000 times more than Truth Social. Xwitter has lost a lot of money for Musk, but unfortunately, despite its dire business position - costs 10 times income in a recent SEC filing - Truth Social could well make a lot of money for Trump, according to the analysis in this LegalEagle analysis (youtube 20 mins), which also exposes a remarkable amount of legal shenanigans. (This one is presented by Liz Dye who is actually rather better than the usual LE presenter.)

[Grumble]

I was under the impression that my “following” tab was meant to be people that I was following. Today I found a tweet from Tommy Robinson on there, who I don’t mind telling you I don’t follow. I try to keep off the “for you” tab because that’s bound to be sh.t but I hoped I was ok on “following”.

[Woodchopper]

I was under the impression that my “following” tab was meant to be people that I was following. Today I found a tweet from Tommy Robinson on there, who I don’t mind telling you I don’t follow. I try to keep off the “for you” tab because that’s bound to be sh.t but I hoped I was ok on “following”.

I think the ‘following’ tab also includes retweets from people you follow, and perhaps their likes.

[Grumble]

I was under the impression that my “following” tab was meant to be people that I was following. Today I found a tweet from Tommy Robinson on there, who I don’t mind telling you I don’t follow. I try to keep off the “for you” tab because that’s bound to be sh.t but I hoped I was ok on “following”.

I think the ‘following’ tab also includes retweets from people you follow, and perhaps their likes.

It wasn’t a retweet. I may have been on For You by mistake. I hope so anyway.

[jimbob]

I think he sometimes pays for adverts,

[Grumble]

How about no?

[Brightonian]

In this thread, about "likes" now being more anonymous, there's a chart from Musk showing spikes of likes on the hour, every hour. Presumably those are bots scheduled at xx:00, to amplify tweets.

But there's also a sharp dip ("anti-spike"? "negative spike"?) just before the hour. I would've expected a fairly uniform distribution in the minutes starting hh:01 to hh:59, but there is definitely this sharp dip at hh:58 or hh:59. Why might that be?

[dyqik]

In this thread, about "likes" now being more anonymous, there's a chart from Musk showing spikes of likes on the hour, every hour. Presumably those are bots scheduled at xx:00, to amplify tweets.

But there's also a sharp dip ("anti-spike"? "negative spike"?) just before the hour. I would've expected a fairly uniform distribution in the minutes starting hh:01 to hh:59, but there is definitely this sharp dip at hh:58 or hh:59. Why might that be?

It could be a slow down in Twitter's backend caused by bots all connecting and liking on the hour that causes a delay in registering and time-stamping likes.

I doubt Twitter is designed to carefully preserve and ensure monotonically increasing time stamps at the second and minute level, with transactional database operations designed to ensure that everything happens without race conditions and with self-consistent metadata, as that's just not required for it, and is computationally more expensive.

[Brightonian]

In this thread, about "likes" now being more anonymous, there's a chart from Musk showing spikes of likes on the hour, every hour. Presumably those are bots scheduled at xx:00, to amplify tweets.

But there's also a sharp dip ("anti-spike"? "negative spike"?) just before the hour. I would've expected a fairly uniform distribution in the minutes starting hh:01 to hh:59, but there is definitely this sharp dip at hh:58 or hh:59. Why might that be?

It could be a slow down in Twitter's backend caused by bots all connecting and liking on the hour that causes a delay in registering and time-stamping likes.

I doubt Twitter is designed to carefully preserve and ensure monotonically increasing time stamps at the second and minute level, with transactional database operations designed to ensure that everything happens without race conditions and with self-consistent metadata, as that's just not required for it, and is computationally more expensive.

Thanks, makes sense, more precise timestamps really aren't warranted (unlike, say, stock market transactions).