What if I made a bot that just posted the title of any bill supported by a majority of US GOP House reps and added comments like "This will harm millions!" or "Wealth transfer to the wealthy!" or "Income inequality will become even worse!" or "This will marginalize minorities and women even more!" without reading the text of the bill at all.
Then I let that run for a few years.
How often would the comments be factually wrong?
(Note: I can't make bots. If you do this, just gimme a shout out)
@tomstearns The primary risk to your bot's truthiness will be bills with major bipartisan support, like continuing resolution funding bills (not much changes), bills naming post offices, and bills regarding specifically Congress' operational procedures.
@Ontploffing Hm. Good point. Then I should look at partisan bills... if I ever learn how to code a bot.
@Ontploffing Thinking more about it in more detail, this problem (assuming I could easily scrape the relevant data) has potential solutions.
One that comes to mind is to create a simple cross-product or cross-sum for each bill of the # of GOP vs. Dem supporters of each bill, weighted by N. I visualize a 2-axis Cartesian space. If N is X and proportion of Republicans (right, natch) vs. Democrat (left) supporters is Y, the bot could choose only bills in some part of the upper right.
@tomstearns Fortunately, most Congressional independents caucus with the Democrats. :p
@tomstearns The hardest thing about this would be aggregating the bill sponsorships. Don't know if Congress has an API you could hit.
But actually making the bot would be pretty easy if you got that data.