kyle-rb

The more obvious bias is towards Premium subscribers over us freeloaders. The reply-ranking algorithm was a nightmare for a while; they strictly adhered to the rules of gold check > blue check > normal account. It's still not great, but eventually they added enough slack to let normal accounts with good quips outrank 30 paying users saying "lol", and tweaked some things so it's no longer so easy to shill tshirts/whatever by tagging Grok for the gold check double-boost.

But they're still reticent to ban any Premium account without a reason good enough to give up the $8/month. I have some sympathy for trying to moderate in the era of LLMs that can generate perfectly on-topic replies that even take images into context. But I think they need a more targeted approach against OnlyFans bots replying to popular tweets with a standard formula of some sexual reference and an insane claim. Sometimes they're less sexual and more straight up horrifying [0]:

> i breed mantises for fun and once tried grafting two ooths together just to see what hatched. got a whole clutch of nymphs that all just twitched in circles and died by day two. butterfly chimera energy hits different when it's your own disaster.

The other bot replies to the same OP are the more typical sexualized sort; they tend to reply in swarms.

[0] https://x.com/ariexsha/status/2110924715103514871

A few weeks ago Donald Trump put a series of polls via Truth Social and Twitter with 3 choices for a replacement name for Artificial Intelligence (which according to him makes it sound like it's fake when it's actually real).

The initial choices were: "Supreme Intelligence", "Superior Intelligence", "Extreme Intelligence". [0]

Then he removed "Supreme Intelligence" because he said it reminded people of the failing Supreme Court, and it was down to just "Superior" or "Extreme". [1]

Then, presumably after some industry people talked him into a compromise, he decided it was actually between "Superior" or "Super Intelligence", and the latter won. [2]

[0] https://x.com/realDonaldTrump/status/2101350559328416142

[1] https://x.com/realDonaldTrump/status/2101698074166882517

[2] https://x.com/WhiteHouse/status/2102067457506611585

> Cloning someone's SIM

Yeah I don't trust SMS either, that's why I said email. You can do a lot more to protect your email account than your phone number.

It's very disconcerting to think about how many malicious apps like that must exist in the Play Store, but again, gaining access to someone's Google account is still game over if that's how you're syncing passkeys.

> Emails are super easy to hack

Source? And if somebody hacks my Gmail account, won't they be able to access my Google-synced passkeys?

Magic link auth isn't any less secure than any site that has a password-reset flow.

Important context: IKEA just released a mod where you can play as a shelf.

https://www.youtube.com/watch?v=iZODN0QUgjI

https://news.ycombinator.com/item?id=49639647

Microsoft Bob wasn't wrong, it was just early.

My initial suspicion was that my account was flagged because I only used the web version (the PWA experience on mobile is pretty good) so I've been using the Android app the last few days, and I'll see if that has any signal on my account.

In another view days I'll try filing a support ticket, maybe claiming something like "I was mass-reported by DSA leftists for my right-wing views"; things along those lines have gotten some peoples' permanent suspensions lifted.

I haven't left because there are still circles of people I follow that haven't bailed. Once a large enough majority move to Bluesky, I'll start browsing that app more.

On the other hand, I might start posting on Bluesky more because as of a week ago, I'm shadowbanned on X. I know people throw that term around, but nobody on the site can see any of my reply tweets, even me (unless I scroll my account's 'replies' tab); people who've followed me for years are asking why their tweet has a ghost reply.

I don't think X has that much of an incentive to filter out bots while keeping around accounts like me (200 followers, will never pay for Premium). The slightly higher effort bots that glaze tech CEOs will be basically indistinguishable from the real people glazing tech CEOs because they're wannabe founders trying to network/cloutfarm. And that style of dead internet is fine with Elon.

It's worth noting that the particular recording they include[0] is not the original performance, it's a CC0 (public domain) recreation of the sample.

So the recording itself has no copyright attached, but yes, there's still a question of the copyright on the composition.

[0] https://freesound.org/people/Dolfeus/sounds/43389/

The arbitrariness of app store review hasn't been the greatest on the Google side, for me lately anyway. I got dinged for having an E-rating for my TV show info app that displays an official promo image for an episode of Landman, where a character has a wound with some blood on his face.

I pointed out that there were a couple other E-rated apps that show the exact same image, and many more T-rated ones that only warn of "In-App Purchases" — no "Violence" label. One of the latter is the official Google TV app [0], so I sent some screenshots of that, and asked if maybe they were "crushing the competition" by avoiding their own review process, to see if that would escalate my case (no such luck).

[0] The image is still the same in the Google TV app, and Google Search surfaces the image in question S1 E10: https://www.google.com/search?q=landman+season+1

Wow a lot of emdashes, are we sure she didn't use Claude?

> Taste.

Ok but what if I told you it's also serverless?

It's AI-native Microsoft SharePoint

I agree that the overall premise is somewhat eugenics-y, but I overlook that on the basis that it's a comedy movie from 2005 and it's easier to swallow than something like "aliens abducted all the world's smart people"[0] or something. There are still some good themes on over-reliance on automation without understanding it.

The Idiocracy line that I think about the most these days is "Brawndo stock dropped to zero and the computer did that auto-layoff thing to everybody."

[0] https://www.youtube.com/watch?v=lllrpcd8vbk

The CSS selector they used (`a[href*="://"]::after`) is meant to only target only external links, but assumes any internal links will be using relative paths like `href="/about"`. The problem is that this site uses absolute URLs (`href="https://hhh.hn/about"`) for its nav links, so every link ends up with an icon.

You could fix this by adding an exception to the CSS rule so it skips links starting with your site's name:

  a[href*="://"]:not([href^="https://hhh.hn"])::after

Not sure how long the subreddit has existed but I always assumed it was inspired by the episode of The Office where Michael asks Oscar to explain what a surplus is like he's 5. The broader usage on social media and the ELI5 shorthand are definitely the result of the subreddit though.

This article is definitely abusing the terminology, because I wouldn't expect even tech-knowledgeable people to know what Nostr is or does.

The term AGI was coined almost 30 years ago; is that when the marketers got ahold of it?

If you don't like BlueSky managing your keys, the alternative to did:plc is did:web. Unfortunately, afaiu, it's not possible to migrate[0]. If this is something you really care about, and you trust yourself to manage your own private keys, you should probably bite the bullet and do it now. Make a pinned post on the old account to point to the new one.

That's the one part you can't migrate; everything else you can change while retaining your followers, posts, etc.

Setting your handle to a domain name via DNS[1] is intended to be easy for non-technical users (especially with registrars adding one-click forms for doing so) and also decoupled from actual PDS hosting. Many users just want their handle to reflect their identity on the web, and are ok trusting BlueSky to host their PDS for free.

Migrating to a new PDS is possible in both the cooperative case[2] or in the adversarial case[3]. The gist is that A: you should take regular backups by exporting your account data, and B: if you're using did:plc instead of did:web, you should register a backup key that has a higher priority than the key held by your PDS, so a malicious PDS can't simply migrate you back.

[0] https://github.com/bluesky-social/atproto/discussions/2705

[1] Alternatively there's an option for HTTP handle resolution instead via `/.well-known/atproto-did`

[2] https://atproto.com/specs/account#pds-account-migration

[3] https://www.da.vidbuchanan.co.uk/blog/adversarial-pds-migrat...

Well actually, that's not what an AppView is for.

If you just want to post messages on your website that get syndicated to BlueSky (as blog post links or full threads or whatever) you could have your website server make API calls to your PDS (whether that's self-hosted or not).

If you're talking about building your own custom BlueSky app, "AppView" is not synonymous with "viewer app"; basically it's an "AppView" because it provides a "view" of the network necessary to power an end user "app". I.e. it holds a copy of every single post, repost, and like, so you don't have to. Your custom app would simply make requests to the logged-in user's PDS, which routes most of those requests to that user's preferred AppView.

I view Bluesky as being decentralized-optional. It's cool because you actually can join first and wrap your head around it later. You can continue using your account you created on the fully 1st-party stack, and you still have the option to switch to self-hosting the parts you care about, without losing your posts or followers.

- If you just want to use your domain name as a username, you set a DNS record.

- If what you care about is the client, you can build your own website or native app. You don't really even need to host a server other than for your own static assets, since the app can request Bluesky network data directly via the logged-in user's PDS (they even have CORS headers!)

- If what you care about is data sovereignty, you can self host your PDS (personal data server) on a low-end VPS. It's cheap because it pretty much just holds your data, passes events to Relays, and proxies data requests to your preferred AppView.

- If what you care about is not needing to trust Bluesky to reliably gather and collate events from each PDS, then you'd need to host a Relay ($30/month) and an AppView (even more expensive) so you'd be best off pooling resources with other people you trust. But that's kind of the nuclear option.

- With a narrower scope though: if you noticed that Bluesky was censoring a handful of legitimate accounts and you still wanted to follow them, I think you could probably have a personal Relay+AppView that only listens to the censored accounts' PDS's, and proxies other requests to the 1st party AppView. (I'm not 100% sure if that would be allowed.)

Is the reputational risk of pirating terabytes of books worse than the reputational risk of shredding (destructively scanning) millions of books?

https://arstechnica.com/ai/2025/06/anthropic-destroyed-milli...

Yeah this is the best implementation of A Young Lady's Illustrated Primer so far.

In the California gold rush, the people who got rich were the ones selling shovelware.

I do think it's a tempting use case, but there are precautions that up I would take if I was doing this sort of thing. Off the top of my head:

- Set up a separate inbox just for the agent, with a forwarding rule that passes on most things that land in my Gmail, but omits as many verification token/password reset emails as possible.

- Have the agent alert me when any sensitive emails do make it through, with suggestions on how to update the forwarding rule.

I guess what's unusual is that the scope includes inbox access.

IMO it's probably a bad idea to have an LLM/agent managing your email inbox. Even if it's readonly and the LLM behaves perfectly, supply chain attacks have an especially large blast radius (even more so if it's your work email).

Yeah, I think the italics compounds the problem in their comment example: // Notify aZZ Zisteners

Plus "a dev typing real fast" from the XKCD Stack (https://xkcd.com/1636/) is now feasible.

It seems like they'd have a stronger negotiating position if they had an alternative contractor waiting in the wings before they accused Anthropic of being woke traitors, as opposed to a threat to migrate away over the next 6 months.

But again, the sophistication of their strategery might also have a negative correlation with Hegseth's BAC.