mikenew

I have around 15k games in SC2 and I can honestly say; this is awful. I love it.

I'm increasingly not seeing the point of google search. If I have an agent that can find whatever I ask it to, including with constraints like "peer reviewed journal articles only" or "working code published on github", why would I be using google's kinda-shitty LLM to do a worse version of that? Search results are just SEO-stuffed AI slop so I find nothing there of value either.

I genuinely think I have not used traditional web search in months.

There is no one left at Blizzard with any RTS experience. They've long since left for smaller studios.

SC2 is my favorite game of all time and I would love nothing more than an SC3. But if there ever is to be a real spiritual successor it won't be from Blizzard.

That's true, but if you wanted to boil down the art of acting to one sentence it would be "do not pretend; actually do".

Jobs was a great actor in that he did not pretend to be excited. Whatever coaching is being given here is exactly in the opposite direction of acting.

Ha, well that's good because I would hate to be labeled an "LLM advocate". I've been on the negative side of the conversation many times, and I think the downsides are very real and often ignored. But calling them useless is just not reality.

As far as system prompts and skills, my approach has been to start with nothing and add very specific instructions as I notice issues or feel like I need them. Mostly around trying to keep it terse, kill the obnoxious rhetoric ("It's not X. It's Y"), and give specific guidance on how to write code. Skills are things like "use edge-tts to generate spoken audio for this answer and send it to me" kind of thing.

> The chatbot recommended some of the dumbest shit you could possibly do

The "I tried it and it sucked" is borderline conspiratorial at this point. There are enough talented, thoughtful developers saying there's something real here, and it is worth believing them and investing some time to understand it, even if you come out the other side and decide you don't want to use LLMs.

Use a very good model. Set up a good harness. Spend some time on your system prompts and skills. Develop your intuitions about how the model works, what it's good at, how to scope the work, and how to steer it. Recognize when it's alleviating menial work and recognize when it's making choices you really need to understand yourself. Be patient and accept that the failures are going to be very painful for a while.

Don't write it off until you've genuinely seen the upsides.

I've decided that the main thing I'm building is my own mental model. You can take notes, create docs, put graphs and websites together, but unless I'm just trying to generate some reference material the only real objective is to develop the understanding and intuitions inside my own brain.

So I have the LLM offer a very short explanation of something, and from there's it's just me asking questions. Anything that feels fuzzy or not fully internalized is something I poke at until I'm satisfied.

It really has helped me develop a sensitivity to what I understand vs what I don't, and the ability to drill into any part of it is amazing.

It's not a tactical failure of the US; it's a strategic one. The US operates a tremendously powerful military, but that doesn't change the fact that the war with Iran was a tremendous strategic blunder. "Moronic" is a pretty fair characterization.

Apple is on the side of privacy if it serves their marketing. Which is why they would rather build and normalize CLIENT SIDE CONTENT SCANNING so they can continue to market iCloud as "secure and private".

If Apple's interests sometimes align with ours then great. I'll take it. But don't attribute to this ~5 trillion dollar company some kind of altruism.

Yeah at 120hz each frame is 8ms. So you're only missing a single frame around 30% of the time.

I certainly want my latency as low as I can get it. But I'm pretty skeptical that anyone is truly feeling the difference of a couple ms.

Current top 5 played games on the Steam Deck are: Slay the Spire 2, The Binding of Isaac, Dave the Diver, Stardew Valley, and Baldur's Gate 3. BG3 is the only one you can consider AAA, but Larian is hardly an EA or Blizzard.

The big AAA studios recycle the same formulas and push graphical quality (mostly downstream of Unreal Engine improvements) because they are risk averse. Similar to big comic-book-hero films. It's not what consumers want and it's increasingly starting to show. AAA is struggling badly while indie has been on an absolute tear the past few years.

Arcan is such a fascinating project that I've never quite managed to get my head around.

That was my experience a couple months ago, and until someone shows me real evidence of something valuable they've made with it, I'm not wasting my time on this stuff again.

Authenticity requires vulnerability and that's not something Apple can do.

I use Opencode, both directly and through Discord via a little bridge called Kimaki.

https://github.com/remorses/kimaki

GLM 5.1 was the model that made me feel like the Chinese models had truly caught up. I cancelled my Claude Max subscription and genuinely have not missed it at all.

Some people seem to agree and some don't, but I think that indicates we're just down to your specific domain and usage patterns rather than the SOTA models being objectively better like they clearly used to be.

I've found GLM 5.1 to be extremely good, as long as you keep the context under 100k or so. But I do agree; the actual service is rough. It's often unusable during peak hours.

I imagine they're trying to slow down the acquisition of new customers because they're so overloaded. But yeah at double the price it doesn't seem worth it anymore.

I can't find an announcement on this, but here's the archived subscription page from a couple days ago: https://web.archive.org/web/20260410092340/https://z.ai/subs...

The Max plan has doubled, and the Lite and Pro plans have more than doubled. No change in usage limits.

Even if this is largely due to a change in how PCs in China are being counted, it's still amazing to watch Linux usage continue to climb like this.

It's really the only opposing force to Microsoft's enshittification of Windows.

This feels like an existential threat to HN, and to the general concept of anonymous online discourse. Trust in the platform is foundational, and without it the whole thing falls down.

Requiring proof of identity is the only solution I can think of, despite how unappealing it is. And even then, you'll still have people handing their account over to an LLM.

I really struggle to imagine a way around it. It could be that the future is just smaller, closed groups of people you know or know indirectly.

The side effect of trying to enforce this kind of sensitivity is that you make certain things taboo to talk about. And this is a good example of something that should be easy for someone to talk or even joke about because it makes dipping into that conversation much easier.

Pretty much every software engineer I've talked to sees it more or less like you do, with some amount of variance on exactly where you draw the line of "this is where the value prop of an LLM falls off". I think we're just awash in corporate propaganda and the output of social networks, and "it's good for certain things, mixed for others" is just not very memetic.

I think in the context of business communication; yeah a lot of people are doing that. Which, to be honest, I don't think it the worst thing ever. Most corporate communication is some basic information padded out with feigned personal interest and rehearsed politeness, so it's hardly a huge loss.

For personal communication between friends it would be horrible. Authenticity has to be one of the things I value most about the people I know. Didn't mean to imply from that example that I did or would communicate that way.

I played with it extensively for three days. I think there are a few things it does that people are finding interesting:

1. It has a lot of files that it loads into it's context for each conversation, and it consistently updates them. Plus it stores and can reference each conversation. So there's a sense of continuity over time.

2. It connects to messaging services and other accounts of yours, so again it feels continuous. You can use it on your desktop and then pick up your phone and send it an iMessage.

3. It hooks into a lot of things, so it feels like it has more agency. You could send it a voice message over discord and say "hey remember that conversation about birds? Send an email to Steve and ask him what he thinks about it"

It feels more like a smart assistant that's always around than an app you open to ask questions to.

However, it's worth stressing how terrible the software actually is. Not a single thing I attempted to do worked correctly, important issues (like the discord integration having huge message delays and sometimes dropping messages) get closed because "sorry we have too many issues", and I really got the impression that the whole thing is just a vibe coded pile of garbage. And I don't like to be that critical about an open source project like this, but I think considering the level of hype and the dramatic claims that humans shouldn't be writing code anymore, I think it's worth being clear about.

Ended up deleting it and setting up something much simpler. I installed a little discord relay called kimaki, and that lets me interact with instances of opencode over discord when I want to. I also spent some time setting up persistent files and made sure the llm can update them, although only when I ask it to in this case. That's covered enough of what I liked from OpenClaw to satisfy me.

I like how Xcode installs a bunch of gigantic, multi-gigabyte artifacts for like ios runtimes or whatever, fills up the hard drive, can't update because it's out of space, and then tells me I'm not allowed to delete them because of SIP.

I've been using it for the past couple days. Like most AI products right now, it is both incredible and incredibly stupid.

Virtually everything I've tried (starting with just getting it running) was broken in some way. Most of those things I was able to use an LLM to resolve, which is cool, but also why doesn't it just work to begin with?

I still haven't gotten it to successfully create a cron job. Also messages keep getting lost between the web GUI and discord. Trying to enable the matrix integration broke the whole thing. It seems to be able to recall past sessions, but only sometimes.

I've been using OpenCode with various models, often times running several instances in tmux that I can connect to and switch between over ssh. It feels like the hype around openclaw is mostly from bringing the multi-instance agentic experience to non-developers, and providing some nice hooks to integrate with email, twitter, etc. But given that I have a nice setup running opencode in little firejail-isolated containers, I'll probably drop openclaw. Way too janky, and I can't get over the thought of "if this is so amazing, why doesn't it work?"

I've recently switched to OpenCode and found it to be far better. Plus GML 4.7 is free at the moment, so for now it's a great no-cost setup.