nick evans

@[email protected]
75 Followers
255 Following
153 Posts
i should be better at this by now...
GitHubhttps://github.com/nevans
Net::IMAPhttps://github.com/ruby/net-imap
nick evansMar 16
Pascal CostanzaMar 13
I have now seen this a few times in mailing lists: There is some discussion with an explicit or implicit question, somebody asks some random LLM and then posts its answer into the discussion, without double-checking it. Then somebody else double-checks it, shows that it's wrong, or at least in need of some correction. All of this is a waste of bandwidth, and frankly speaking just impolite. No matter what your stance on LLMs is, anybody can ask an LLMs themselves, there is no added value whatsoever in posting an unfiltered LLM answer to a public discussion. There should be an etiquette on banning this kind of behavior. I assume that people who do this act in good faith and believe that they are helpful, so it's ok to tell them in a gentle way, but this behavior should stop IMHO.
nick evansMar 13
JenniferplusplusMar 10

"AI can make mistakes, always check the results"

I fucking loathe this phrase and everything that goes into it. It's not advice. It's a threat.

You probably read it as "AI is _capable_ of making mistakes; you _should_ check the results".

What it actually says is "AI is _permitted_ to make mistakes; _you are liable_ for the results, whether you check them or not".

Except "you" is generally not even the person building, installing, or even using the AI. It's the person the AI is used on:
https://thepit.social/@peter/116205452673914720

nick evansMar 13
Show threadBradley SchaeferMar 11

@nick_evans i think simple and direct wins: https://github.com/lobsters/lobsters/blob/main/AGENTS.md

It came to mind because of this thread: https://lobste.rs/s/x0qrlm/agents_md_as_dark_signal#c_zwlclq

Show threadnick evansMar 13
@soulcutter that's marvelous and very sensible, and something I may copy for some of my own projects. But it's not quite the level of cathartic hostility I was looking for. 😉
Show threadnick evansMar 13

maybe I can move on to some other games now?

Although... I still haven't gotten to Ante 39...

nick evansMar 11
I'm wondering if there are any (open or proprietary) projects with a hostile AGENTS.md. Something to just tell all of that stuff to fuck off, in spectacular fashion, or subtly. Ideally in a way that makes me laugh (but may make the user running the agent cry).
nick evansMar 11
Jan Lehnardt May 25, 2025

I’m officially done with takes on AI beginning “Ethical concerns aside…”.

No! Stop right there.

Ethical concerns front and center. First thing. Let’s get this out of the way and then see if thre is anything left worth talking about.

Ethics is the formalisation of how we are treating one another as human beings and how we relate to the world around us.

It is *impossible* to put ethics aside.

What you mean is “I don’t want to apologise for my greed and selfishness.”

Say that first.

nick evansMar 7
𝐿𝒶𝓃𝒶 "not yet begun to fight"Mar 7
A.I. musicians are COOKED. I just recorded an ENTIRE SONG using nothing but my instruments and a few microphones. This would have cost upwards of hundreds of dollars in expensive A.I. subscription fees but for me it was completely FREE. Don't get left behind. This is the future!
Show threadnick evansFeb 25
@janl @mattly I've been running either Debian or Ubuntu as my primary OS since '98. And, although I'm not currently dual booting or running VMs, I did dual boot (both Windows and Mac OS) for well over a decade. I even dual booted ChromeOS for a while (before there was an official VM option). But it's never before been as viable an option as it is today.
Show threadnick evansJan 14
But I still haven't defeated Simon!