cd /news/ai-tools/how-to-think-about-the-difference-be… · home topics ai-tools article
[ARTICLE · art-136730] src=danielmiessler.com ↗ pub= topic=ai-tools verified=true sentiment=· neutral

How to Think About the Difference Between Choice and Score in Jev

A blog post on Jev distinguishes the tool's "choice" and "score" primitives by arguing that score ranks options on an ordered scale, such as cold/warm/hot or critical/high/medium/low, while choice returns probabilities for an unordered list. The author, writing on the Typesafe AI blog, says both primitives return probabilities and proposes renaming them "choice" and "scale," stating 85% confidence in the explanation. The post follows earlier coverage of Jev and cites a 2021 chart-scale-types reference on nominal and ordinal scales.

read2 min views1 publishedSep 22, 2026
How to Think About the Difference Between Choice and Score in Jev
Image: Danielmiessler (auto-discovered)

One thing that has been tripping me up with Jev is the difference between choice and score. I keep mixing them up.

I finally took the time to debate it with my digital assistant, and I came up with a better way of describing it to myself, which I will share here.

They're both actually returning probabilities. So I was like, well, if they're both choices that are scored by probability, why not call them both choice? Or more importantly, why have a score one that's separate from choice, if score is just giving you back probabilities on the options that you gave, just like with choice?

Okay, so score is very similar to choice in many ways, but the main difference is you provide it in order, as a ladder.

So it's like cold, warm, hot. Or for security: critical, high, medium, low. Or for customer service: about to cancel, super agitated, and so on.

So between the structure that is being used, the way you phrase the inputs, and the state that you provide, which is the context, the system itself understands all of that and does a better job of ranking them because it understands that it's a scale.

In the case of choice, you just get probabilities for whatever list you provide, and it could be in any order because it's not on a scale.

Nominal Scale: Only the selection name matters, not any particular order. Ordinal Scale: Only the order matters, but you can't tell the relative difference between the items.Chart Scale Types, 2021

So in my mind, I'm going to think of these as choice and scale instead of choice and score, because both of them provide scores.

I'm 85% confident that I have this right. If anyone has a better way of explaining it, please let me know. But hopefully this helps someone.

── more in #ai-tools 4 stories · sorted by recency
── more on @jev 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
Live at https://your-agent.zahid.host
Get free account → Pricing
from €0/mo · no card required
LIVE [news/how-to-think-about-t…] indexed:0 read:2min 2026-09-22 ·