There are a lot of features under the hood that I'm not sure what to do with yet - various contextual analyses, a storytelling mode that will fly you around and voice-over a series of events automatically, a system which makes and later scores predictions.

There's also an entire "newsroom" editorial and journalistic layer which writes and publishes articles based on developments, using its own judgement.

Currently running (astonishingly well at the price point) on deepseek-3.2, which tends to reject Chinese military news. American models tend to refuse on Iran-Israel.

I've had a lot of fun and felt very well-informed on current events while building this. Would this kind of thing be useful to HN users?

Comments URL: https://news.ycombinator.com/item?id=47970178

Points: 2

# Comments: 0

Presumptuous and wrong "memories" from a one-off command which affect all future commands, repeated/nonsensical phrases in messages, novel display bugs which make going back in the conversation impossible (I can't tell where I am), lack of basic forking features (resume a current convo in a second CC instance -> fork = no history for that convo?), poor/unclear reasoning, a new set of unclear folksy phrases (it really wants to "cut code" all of a sudden).

Qwen + Opencode has been a game changer: which runs very well on a 4090 for basic/exploratory/private tasks, and being able to switch to and between frontier models (using openrouter in my case) to avoid vendor lock in feels like basic hygiene.

There's also the homo economicus psychological difference between having a token budget to use up, and a cost per token. I'm more thoughtful about my usage now.

It's more a data-driven pub test, I think it explains itself well.

The question is - are those apps actually so simplistic? Or is this a strawman.

If AI slop is replacing the content you were consuming, it was already slop.

Perhaps it'd be more difficult for him to broadcast if he had an anti-china perspective, but the content itself seems legitimate.

Your comment has the effect of being flippant, condescending, and seemingly callous to the subject matter. When called out, you have backed up to an alternative explanation which is, again, massively condescending (I don't need channeling mate, certainly not from you).

You have not engaged with the content in a good faith manner.

So, standing back and looking at your comment in terms of its effects rather than what it claims to be its effects (AND the effect that making those secondary claims have - doubling down on condescension), it looks more like you're trying to bully me into changing my behaviour and viewpoint without meaningfully engaging with the content.

Ironically, I'm feeling psychological reactance, so your comments polarized me against you (see the Backfire Effect) and deepened my convictions.

I won't engage with bullies any further but to call them out, I'm hesitant to bring the conversation down to this level and give you any kind of air to begin with, but I think it's important to analyze discourse as it happens.

Meta knowingly hurt children for profit. It worked.

If we are in any way serious about technocratic solutions to social problems, this would be untenable, the company would be bankrupted, a new company would fill its place. No tears would be cried, nothing of value would be lost, half of hacker news would be chafing at the bit to build a better alternative for the newly opened market.

But that's not what happened. We allowed children to be knowingly hurt for profit.

The system is functioning as intended.

You can purchase a scam ad it'll be up in 10 minutes. Lie to every anxious child they have ADHD and need meth, lie to every dejected boy that they just need to manosphere up and buy supplements.

They think the public is stupid. They might be right.

Got away with it again, good profit, will repeat.