Where answers come from

Some pages can ask a language model questions. Set this up once; your choice is remembered in this browser for the whole site.

Model source
Use another source instead

Smallest-to-largest weight downloads. Qwen3.5 0.8B is our starting choice for page Q&A. GPU estimates are not total RAM: leave room for the browser, context and temporary buffers. Hardware changes speed and capacity; answers still need checking.

Session-only loading avoids saving model weights, uses extra RAM and downloads again after unloading. Turn it off to save downloads. Delete-data-on-exit or private browsing may impose a much smaller storage limit than the displayed estimate.

You sign in on OpenRouter; it gives this page a key for your account.

A free account does not make every model free. Choose a model with a listed :free variant; merely adding that suffix does not create one. Check free-model availability and pricing.

Chat

AI assistant, not Shawn Hartsock. It does not speak for him.

Inspect the page excerpt

Up to 4,800 characters, not the whole site. Recent conversation turns are retained within the model’s context limit.

Ask about this page, or just say hello.

Ready when you are.

Not saved or uploaded automatically. Closing keeps this conversation in this tab; changing pages starts a new one. Model replies can be wrong.

Convenience To Security Scale

Originally published .

I was thinking about a few things… we’re doing a kind of security system audit at work and I’ve come up with a few koans about it:

*

Convenience is inversely proportional to Security.

*

*

The more secure something is the more inconvenient it is.

*

*

The easiest security system to use is one that doesn’t exist.

*

In one measure the goal of security is to make things hard for people. Specifically, it should make things extremely hard for Bad People (tm) and not too hard for Good People (tm). These are both user groups that you are attempting to distinguish between. Both the good and bad people are people attempting to use your system and you are making a call as to which group a person belongs in.

So to that effect I’ve created the Convenience to Security Scale (patent pending).

On a scale from 1 to 10 a system that rates a 1 has no security whatsoever and is (presumably) very easy to use. And, a system that rates a 10 is perfectly secure because it is completely inaccessable. The goal is to judge where on the scale your system needs to be.

This model works even for insecure and frequently crashing programs since the frequent crashes are viewed (in this model) as a security feature preventing attackers from using the service. So making the program more stable and thereby more useable makes it more attackable… lowering its rating on the scale. A program that can’t execute is perfectly secure since it can never be attacked.

The “secure” in this model is also referring to the protection of information… no “secret” information is divulged by the program. So a program that can’t produce output is perfectly secure… it is also perfectly unusable.

Explore the map · Browse by topic and date