31–40 of 44 posts
Side-note, if anyone wants to really dig into all the data available about bills (including votes, attachments, etc.), this is a great place to start: https://github.com/unitedstates/congress There's excellent documentation on the formats and how to access all the data.
I did much of that so I appreciate you saying that the documentation is excellent. :)
Paste the entire thing into the LLM! Maybe people can stop relying on unreliable partisan sources to interpret bills if they have tools to grok the dense weird language in them themselves. I say this even though I was embarrassed yesterday when the LLM misinterpreted something and I posted it - read the reference text behind any summary :/
So use an LLM even though you admit immediately they make mistakes and you need to read the entire bill anyways?
My friend runs congress.dev which displays diffs See https://congress.dev/bill/119/House/1/EH
This is really great. Reading the bill raw feels like reviewing a diff with context set to 0.
Earlier quoted context omitted.
Versus what? An intractable archive of unreadable documents? At the very least they'll get tractable information, which humans will always use on social media to make a point, which will then get fact checked. I prefer that loop. Right now the information is hidden in coffers and never gets taken for a loop.
A root cause analysis would probably suggest the question - Why are our representatives passing intractable, unreadable documents as law and how can we prevent them from doing that? Or more generally, what changes can be made to our government institutions to improve clarity in communicating actions and decisions to votes? Yeah, it's naive thinking, and I'm well aware the obfuscation is sometimes the point. But I dig…
Best guess?
1) These are actual laws so they carry all the legal thoroughness
2) Like managers, they don't write the actual code (they don't do the actual writing of the bills). So managers don't really care just how awful the code can be (or in this case, just how intractable the bills are)
3) No one is code reviewing (the public is to uneducated to even do so)
This leads to these things being drafted in the dark of night and passed in the dark of night. I'm open to AI in this case simply to even begin having insight.
My friend runs congress.dev which displays diffs See https://congress.dev/bill/119/House/1/EH
This is really great. Reading the bill raw feels like reviewing a diff with context set to 0.
Earlier quoted context omitted.
This is really great. Reading the bill raw feels like reviewing a diff with context set to 0.
What I find most frustrating are the bills written as prose-diffs themselves: "In some entirely different piece of law, Foo shall be inserted after Bar, with an overall effect and purpose which will not be described here."
Earlier quoted context omitted.
This is really great. Reading the bill raw feels like reviewing a diff with context set to 0.
This site's font is very pleasant to read. Poking around the raw html reveals webkit antialiasing and a Google font called Nunito Sans.