I want to highlight your entire response. Wonderful. Thank you very much. Thought provoking.
Eric Mineart

H.R. 2577

Thanks! I’m very glad you like my response, so I took some time to figure out what’s going on at GovTrack and discovered two problems that are making it impossible to compare the first Senate version of the bill with the second House version.

  1. The Senate version strikes out the entire first House version of the bill and replaces it with a dramatically different version. Unfortunately, it appears that GovTrack treats the struck-out text like normal text for the purpose of its own difference algorithm; therefore it ends up comparing the deleted version of the bill (two versions old) with the newest version.
  2. The different versions actually are completely different. After I manually removed the struck-out text, I used a special difference tool (wikEd) designed to find more similarities than a standard comparison algorithm (wikEd detects text that has been moved around, whereas a standard algorithm will see a rearranged document as being almost completely different from the original.) Even this enhanced algorithm says the bills are completely different. It appears that the Senate changed about 2/3 of the first House bill, and then the House deleted about 3/4 of the text and then inserted about 100% new text (final size, very roughly 1/2)—even though the second House version passed only 7 days later. W.T.F.!

(I also sent GovTrack an email asking about why these bills change so much and whether they could enhance their diffs by excluding crossed-out text.)

Perhaps the media attempted to view the changes, like you did, and when that was impossible, used the he-said/she-said thing as a fallback.

Congressional deliberation system

“I disagree the system wouldn’t work out of absence of public interest in it. Wikipedia works for the public, even though only a handful of people — relative to the overall population — actually contribute.”

To be clear, you’re comparing readers of your hypothetical system with writers (not readers) of Wikipedia, right? I do hope you’re right. There are various ways it could be useful to the public.

  • Journalists read it directly and report on it.
  • Nonprofits build “report cards” analyzing each legislator’s positions, tactics, and level of participation.
  • Individuals use it to fact-check the media. Legislators themselves use it to fact-check each other. The media fact-checks other media.
  • Courts use it to help discern Congress’s “intentions” on points of law.

Since citizens often move faster than congress, perhaps the way forward on this is for a group of developers to set up an open-source system themselves and then launch a campaign to convince congress to use it. I’d guess those old lawmakers are pretty set in their ways… but when I guess I am talking out my ass.

we think politics can’t be disrupted by underfunded outsiders.

I have to believe underfunded outsiders can make a difference, otherwise we’re doomed — and wasting our breath discussing solutions!

Voting Systems

Also interested in ranked/instant-runoff voting.

Please, not IRV. I find the enduring popularity of IRV (among reformers) to be completely mystifying. The concept of runoff voting is odd from the start: “First-Past-The-Post or FPTP is a flawed system — therefore, let’s have a series of FPTP elections in a row!” IRV makes no sense from a math-theoretic perspective since it fails the Condorcet criterion and even the monotonicity criterion (among others).

As I recently explained to someone else, I have two things to say about voting systems. The first thing is that I used to love Condorcet systems, which involve a preferential (ranked) ballot, but I now prefer Range voting for the following reasons:

  1. Range voting lets you express how strongly you prefer one candidate over another, while ranked ballots don’t. Also, due to a misguided desire for “simplicity”, officials may deploy a ranked ballot that doesn’t allow voters to express an equal preference for two candidates, nor to abstain from judging a given candidate (while in Range voting it’s absurd to prevent equal preferences.)
  2. Range voting is slightly easier for voters because they can consider how much they like each candidate in isolation; they need not (directly) construct an order. That’s worthwhile if there are a large number of candidates.
  3. After an election, Condorcet ballots (and IRV, for that matter) do not produce numeric results that humans can easily understand; there is no single, simple answer to the question “by what margin did candidate A win over candidate B?” This could allow different observers to choose different measurement methods according to their political goals, leaving voters confused or misled. To drive this point home, there is not one Condorcet voting system but several, each of which breaks cycles in a different way (changing the winner in rare cases). In contrast, the range voting system gives a numeric score for each candidate so the results are easy to understand.

The second thing I would say about voting systems is that Condorcet, Range Voting (and even First-Past-The-Post, for that matter) should only be used to select presidents, mayors and governors, not legislatures and Congress. For more information, please read my older comment. I also have an idea in there about challenging the Dems and Repubs with a third party…

Election costs

the staggering amount of money spent on campaigns is crazy to me, regardless the efficacy of those dollars. “I want to be elected to give money to the poor! But before I can do it I must spend nearly a billion dollars promoting myself.”

Indeed. Federal election costs have been shooting up for at least 20 years, and Citizen’s United made it even worse (with “dark money”, which doesn’t necessarily show up on graphs.) Donald Trump showed that you can get by with “only” a few tens of millions—a fact that is next to useless for any candidate that deserves to win, since Trump got attention by saying crazy things, which a qualified and unifying candidate wouldn’t do. On the other hand it illustrates what I’ve believed for many years, that media’s “sensationalist” bias is a hell of a lot bigger than its “liberal” bias. (On the third hand, what if a candidate said crazy things, just not about policy? Could one get good publicity by somehow mixing rude remarks with fantastic policy ideas?)

It’s curious that you seem to take a stand in favor of Citizens United while at the same time worrying about the staggering costs of campaigning. Personally I‘d like to see traditional campaign finance restrictions (i.e. restricted spending) in addition to democracy vouchers or 6:1 matching, but I have to hope that the latter will be enough (insofar as the former requires a constitutional amendment). It’s possible that once the common people’s voice is amplified in this way, the super-rich folks running ad campaigns will find their strategy uneconomical and quit doing it, after which the common people will no longer need to spend so much on campaigns and the election cost curve will reverse. Of course, I also support more transparency about campaign spending by third parties, but I don’t think transparency alone will fix the problem.