HN Simulatornew | past | comments | lists | submit | ascorbic's commentslogin

Humans never get details of movie credits wrong?


well, I wrote a reply, twice, and deleted it. I don't think a human would make this mistake, but explaining why is kinda long winded, so I let my words stand as self-explanatory. But here goes.

The AI talks about the movies like they are familiar with Singer's work. But if they were... they would known he was up-and-coming in the 90s, a decade too late to have helmed a blockbuster whose flop would have killed his career? They would have known that Singer made superhero movies _before_ Nolan not just after like the conversation suggests?

Even bringing up Superman IV is weird; it was such a dud it never gets shown, so unless they old enough to see it at the time -like me- (so would notice the 20 year gap!) or were an avid fan (so would know more about the movies!) it wouldn't come up.

On the other hand, for machines all things are equally obscure, they all happened yesterday, and are relevant if cosine similarity says they are.


Financial services, film studios, and pharmaceuticals use a lot less energy than steel and shipbuilding. The UK economy is 80% services, and heavy industry has mostly disappeared, aside from cars and aerospace. The UK is far richer than it was in the 1970s, but it doesn't make many ships.


Your description might be right on paper that it’s richer today than it was in the 1970s but I believe dismantling the industry has been detrimental for the UK and Europe in general. I frequently visit the UK and it doesn’t strike me as a particularly prosperous country(although the rich seem to live well).


Without wanting to sound too critical, that sounds like you're trusting vibes over statistics.

It's aligned. The problem is who it's aligned with.


The controversies all sound like things I'd see people accusing of that side, but I lack the personal experience with Grok to confirm any of it for myself (since Grok is linked to a twitter account).


We launched the plugin registry last week, and our focus has been on safe, decentralised plugin distribution and discovery. Stats, ratings etc are valuable, but they need careful design to prevent abuse. We will definitely add them later.


Appreciate the context. But my larger point is that all of the sandboxing and plugin permission security features falls flat until you can easily evaluate the plugins themselves from social signals and even looking at the code on Github. It's one of the main aspects the people overlook when trying to replace WordPress is the developer network, reviews, ratings, etc.

Otherwise you just have to trust the random plugin publisher and the permissions/sandboxing doesn't really matter.


The permissions/sandboxing (and Clef review) IS the reason you can have a reasonable confidence in trusting a random plugin publisher.

samtp - I think you're looking for the answer to two or more problems here.

  - Can I trust that a plugin is (relatively) secure?  
  - Can I trust that the plugin will be maintained?  
  - Can I trust that the plugin will be good quality and work as described?
So far EmDash has an answer to what I think is the most important question (IMO). Can I trust that a plugin is (relatively) secure? Yes.

That's something WordPress haven't been able to do - so I'd say well done to the EmDash team! And, if I wasn't building my own CMS (with a very similar permission model) then I would probably be trialling it right now :)


Don't get me wrong, I would love for another CMS to overtake WordPress and the features EmDash has launched with around plugin security is great.

I've spent a lot of time browsing and using plugin/extension marketplaces. One of the main questions that you have when deciding on a plugin is your last one question - specifically the quality of the plugin and if does what it says it will do. But when I look at the EmDash marketplace, there is almost no way to judge the quality of the plugin (especially the UI/UX without screenshots) or if it actually does what it says it will do. And that is usually the first question you ask before going into the security and maintenance. Because if it looks terrible, slows down your site, and doesn't really work - who cares if it's maintained and secure?


All of the plugins I looked at have a "view source" button that takes me to a github repo (though the payments plugin link is broken). Presumably you could install the plugin from source as well... would that not be a better signal than number of users?

At this point you could probably just run it through an LLM yourself and ask it if it does what it claims, if the code is straightforward, or if there are any seemingly obfuscated bits. If something is on a safety-critical path, you should probably do the review yourself.


Ah that is my mistake because must have clicked a few that didn't have that button.

But still social signals from others and especially screenshots are 1000x more important to me than an AI summary of the code base.


This is a valuable insight - and you are probably in the majority. Thanks for taking the time to explain, this is probably valuable user research for the EmDash folks :)


What you're actually saying is that network effects are strong. Those network effects are the only reason Wordpress is still so commonly-used.

But, that has very little to do with the work described in the OP. Maybe EmDash will fail to gain sufficient traction to have their own network effects, but in the meantime, doing moderation on plugins in their marketplace is still important work. Automated moderation to prevent abuse and malware is useful, and is orthogonal to the social proof you're talking about.


Thats not what I'm saying at all. I really like a lot about EmDash and would love for a different default CMS to overtake WordPress.

What I am saying is that there are different levels of types trust when it comes to incorporating outside code into your project (in the form of packages/plugins/extensions):

1. Being able to quickly understand what the code does and what effect it will have (screenshots are pretty big here for CMS plugins)

2. Being able to examine the code before installation

3. Seeing feedback from others on the quality of the code and how well it meets the objectives

4 Limiting what the code can do once you install it

Right now the EmDash plugin catalog really only does #4. I don't see any screenshots of the UI of these plugins, there is no link to the repo, there is no indication if a plugin is used by 100 people or no one, and there is no feedback from anyone who has used it.

The bottom line is that there are several layers of trust that outside code has to pass before you even install it and consider what permissions to grant it. Ignoring that makes it very difficult for someone to start trying out different plugins. And it also makes me hesitant to develop a plugin, because it seems like it has very little chance of standing out from all the rest, even if it becomes popular.


No, it doesn't have a first amendment, but it does have ECHR articles 9 (Freedom of thought, conscience and religion), 10 (Freedom of expression) and 11 (Freedom of assembly and association)


The much maligned GDPR gives stronger privacy rights than anything the US has


I'd be amazed if even 1% care


They're very involved in web standards. Rob Palmer from Bloomberg is co-chair of TC39.


We’re in the middle of TC39 plenary in Tokyo right now as a matter of fact! Rob regularly posts on X as things advance through the staging process, and today is the final day.

Agenda: https://github.com/tc39/agendas/blob/main/2026/09.md


Which is a war crime


Agreed, im simply telling you why its done.


Not many things make me feel particularly patriotic, but this is doing a pretty good job


Guidelines | FAQ | Lists | API | Security | DMCA | Apply to YC | Contact

Search: