Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Elevated error rate across multiple models

status.claude.com

Post reply on HN
Print view View original

261–270 of 293 posts

Re: Elevated error rate across multiple models

#261

Post by arbll »

bflesch wrote: post #247
arbll wrote: post #228

Earlier quoted context omitted.

A lot of those scripts are wrappers around package managers. Creating them is extra work for distributors, but they still do it because package-manager installs are not truly one-liners and offer far less control over the installation experience. Users need to figure out which of the 10+ package managers they should be using, then run several commands. If something fails, the error messages are often cryptic and not…

It's about trust and having an official account for packaging on each platform where my customers getting their software from.

Most official repositories have policies that are incompatible with the needs of software vendors (release timing, supported versions, bundled dependencies, etc...).

IMO a lot of the blame falls onto the package manager ecosystem refusing to take into account very valid needs and claiming they aren't real / desirable.

Re: Elevated error rate across multiple models

#262

Post by arbll »

8note wrote: post #244
arbll wrote: post #218

Earlier quoted context omitted.

The "like an adult" is what has and will continue to hold back linux on the desktop. Always gatekeeping less technical users instead of acknowledging adoption and ease of use are critical.

i dunno, nothing about most computing is particularly easy to use or intuitive. what has worked over time is having computers of various types in schools, where teachers teach students and let them play with it. nobody teaches about the command line, so nobody knows what to do with it. its also inscrutible without a useable help view, unless you already know how to use the terminal

Windows, macOS, iOS, and Android are definitely much easier to use and more intuitive than Linux today. That’s because their developers are incentivized to put themselves in the shoes of less-skilled users and figure out how to build a good experience for them.

I’m all for higher Linux adoption on desktop, but there’s still a lot of resistance to making less-skilled users the primary target instead of power users.

Teaching can help, but if it takes 50 hours to learn the basics of Linux versus 5 hours for Windows, it’s a losing battle.

Re: Elevated error rate across multiple models

#263

Post by agentcooper »

yanis_t wrote: post #18

I suppose it's a good time to encourage people trying out pi[1] with any cheap model from the openrouter rankings page[1]. [1] https://pi.dev/ [2] https://openrouter.ai/rankings

With Agent Client Protocol (ACP) you can keep the same UI and switch not models, but entire agents, that means using tools/prompts/compaction/etc that are tailored for the model.

Try Zed[1] for GUI and pool[2] for TUI.

[1] https://zed.dev/

[2] https://github.com/poolsideai/pool

Re: Elevated error rate across multiple models

#265

Post by CBLT »

agentcooper wrote: post #263
yanis_t wrote: post #18

I suppose it's a good time to encourage people trying out pi[1] with any cheap model from the openrouter rankings page[1]. [1] https://pi.dev/ [2] https://openrouter.ai/rankings

With Agent Client Protocol (ACP) you can keep the same UI and switch not models, but entire agents, that means using tools/prompts/compaction/etc that are tailored for the model. Try Zed[1] for GUI and pool[2] for TUI. [1] https://zed.dev/ [2] https://github.com/poolsideai/pool

Linked your own project with an "All rights reserved" license? The only thing my company will allow me to do with that software is have AI steal it

Re: Elevated error rate across multiple models

#266

Post by agentcooper »

CBLT wrote: post #265
agentcooper wrote: post #263

Earlier quoted context omitted.

With Agent Client Protocol (ACP) you can keep the same UI and switch not models, but entire agents, that means using tools/prompts/compaction/etc that are tailored for the model. Try Zed[1] for GUI and pool[2] for TUI. [1] https://zed.dev/ [2] https://github.com/poolsideai/pool

Linked your own project with an "All rights reserved" license? The only thing my company will allow me to do with that software is have AI steal it

We don't have any client-side telemetry. Conversations with Poolside models are stored, but you can use any ACP agent with pool. And we have plans to open-source it eventually.

Re: Elevated error rate across multiple models

#267

Post by mik3y »

arbll wrote: post #259
tovej wrote: post #251

Earlier quoted context omitted.

Your arguments do not make even a little sense. In what world does a user have to choose between 10 package managers? Each distro has exactly one. There are also only about three, maybe four main package managers out there. A shell script being piped into bash has so many more ways to break than a package. And if yhe theory is that package managers are fickle (they aren't), then how does adding more complexity help?…

It's crazy to me that even after seeing so many major software distributors choose `curl | sh` as their entry point, people like you will still argue to the ends of the earth that there’s no problem with the package manager ecosystem. I'll stop there. I'm not interested in continuing this discussion when it's being conducted in bad faith.

Bad faith, or perhaps just ignorance. It reminds me of purist junior engineers - and I have been one - refusing to understand or tradeoff in the world beyond their own.

Rather than argue with those of us who are pointing out messy realities, this commenter might be better served filing a bug against any number of the projects that offer installation this way, asking them to remove it, and see if it lands any better.

Technical purity/superiority isn’t the only factor, or even the most important one, driving projects to offer quick installers like this.

Re: Elevated error rate across multiple models

#268

Post by girvo »

agentdev001 wrote: post #217
Carrok wrote: post #209

Earlier quoted context omitted.

If you care about privacy at all, you can route your Opencode requests through an inference provider that does not retain any logs or data. It is also much cheaper. So if your boxes include `Privacy` and `Affordability`, then no, they don't tick the same boxes.

You can use the Codex harness with non-openai providers if you want.

Not very well, if you're using the latest version of Codex. They broke a lot of stuff that made that possible.

Re: Elevated error rate across multiple models

#269

Post by HDBaseT »

qarl2 wrote: post #127
tovej wrote: post #106

Earlier quoted context omitted.

Both of them provide that option. I've never installed rust without a package manager. Why would I?

> Why would I? Because then you can install it without depending on a package manager?

Depending on a package manager has its benefits.

I personally try and use my operating systems package manager for all applications (in this instance, dnf on Fedora). The moment you start adding a million repos, third party package managers, Flatpaks, Snaps, random curl install scripts, etc, it becomes extremely unmaintainable.

What happens if the curl application depends on openssl, maybe with a legacy or specific cryptographic function? I assume the curl script will either install the required version, or include the relevant libraries right? Now that is outside of the system package managers scope, meaning updating openSSL to avoid some vuln now requires extra work.

What happens when you go from Fedora 44 to 45? You should be checking all your applications are supported on whatever version or operating system you are running. There is a decently high chance you run into dependency issues when some lib version is updated.

Package Managers are (generally) more secure and simple to use for an end user (they are using the OS to begin with).

Curl scrips are easier for the software developers.

Re: Elevated error rate across multiple models

#270

Post by sometimelurker »

cromka wrote: post #138
yanis_t wrote: post #18

I suppose it's a good time to encourage people trying out pi[1] with any cheap model from the openrouter rankings page[1]. [1] https://pi.dev/ [2] https://openrouter.ai/rankings

Is pi better than opencode?

Ive tried both, and pi's core is a lot smaller but can be extended to have as many features as opencode. personally I like using pi with llama.cpp, and use stuff like pi.dev/packages/pi-llama-cpp-stats to combine them better. there's a lot of stuff you'll feel like pi is missing, but its just optional.
Post reply on HN