Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Long Term Vision #5

Closed
WebsByTodd opened this issue Feb 3, 2017 · 4 comments

Comments

@WebsByTodd
Copy link
Collaborator

commented Feb 3, 2017

I think our values are clear: transparency and truth. Sticking to those values there is plenty of room for a bigger vision. Let's start the brainstorming discussion.

Expanded Foundation of Facts.

I think we should clearly define two directions for expansion.

  1. New sources of facts that are still first-hand records of the events as they occurred without opinion or analysis. (e.g. Court Documents, Statistical Studies, etc.).
  2. Same types of facts, but more allowable content. (e.g. shaky camera footage, peer-reviewed studies, etc.).

Presence at New Events

Our own videographers at new events. No reporter necessary. Just dictators (the voice to text kind, not Kim Jong Un kind) and translators.

Other ideas for the vision?

@nedtwigg

This comment has been minimized.

Copy link
Collaborator

commented Feb 4, 2017

Our goal is to build trust and consensus in an area where there is none. Many organizations have this goal, but none are currently succeeding to the extent necessary for a respectful debate to take place.

Our strategy, unique in the current ecosystem, is to achieve trust and consensus by limiting the scope of allowed evidence until it is so small that every thing it contains is undeniably true, and undeniably free of bias. If we allow political speeches, but 90% of the speeches we contain are from Hillary, then that's true but biased.

Discovering mechanisms which can build a large foundation which is both undeniably true and undeniably free of bias will be extremely difficult. The discussion itself will be a point of contention which may sew distrust and disagreement - which is the opposite of our goal of building trust and consensus.

If we agree to some irrefutable facts, can we start to agree on political issues that we couldn't agree on before?

That is the experiment we are running. If the answer to that question is "No", then we're done, and it doesn't matter if we had an expansion plan, because we can shut it down.

If the answer is "Yes", then we've already succeeded in a huge way. Follow-on experiments would be great, but this first experiment would be a huge achievement on its own, and we must be sure that our expansions don't threaten this achievement.

Figuring out how to grow will be extremely contentious. For example, those who support Black Lives Matter will want to include eyewitness video. People who support Blue Lives Matter will feel that such videos are a biased sample of the evidence, and that we should be looking at crime statistics rather than cherry picking isolated incidents. Both sides are correct, and they already have platforms where they can cite statistics and videos to talk past each other. We're trying to build a place where they can talk using the evidence that they agree on - Cops feel they're being tried in the media and denied due process, and BLM feel that they're being racially profiled and denied equal protection under the law. They've both got the constitution on their side, so let's help them see that.

The broader our focus, the smaller our chances of success are. At this time, I think the most important part of our long term plan is to make a credible promise that we will never dilute an indisputably solid but small foundation by expanding it to contain facts which are difficult to verify or possibly tainted by sampling bias.

One way to allow expansion without accidental dilution is to have a "Foundation" and a "Frontier". The Foundation will include only the unassailable stuff, and the Frontier could include trickier stuff. Takes which use only the Foundation will be presented using a formal serif font, and takes which use the Frontier will have to use a handwriting-style of font.

@WebsByTodd

This comment has been minimized.

Copy link
Collaborator Author

commented Feb 4, 2017

Our goal is to build trust and consensus in an area where there is none.

Currently that's our only goal, and our only hypothesis to validate is:

If we agree to some irrefutable facts, can we start to agree on political issues that we couldn't agree on before?

This is certainly one lofty goal, and one huge assumption. If we meet the goal and validate the assumption, we will have achieved what we set out to achieve. If we fail to validate the hypothesis, then at least we will have succeeded in finding an answer to our question.

Then what?

Currently, none of us can truly answer this question because it's still very early to be asking this. There are things that will come up as we are validating our first hypothesis and I don't want beneficial ideas to go unshared because they are currently out of scope.

I think using the adjective "small" to describe our Foundation is limiting. "Indisputably solid", however, is very apt.

@theglennewman

This comment has been minimized.

Copy link
Collaborator

commented Feb 4, 2017

Alright. I think I've heard in a few different places that right now, the project is completely ridiculous to handle once these allowed data sources get expanded too much.

Discovering mechanisms which can build a large foundation which is both undeniably true and undeniably free of bias will be extremely difficult.

Extremely. Exponentially. Maybe I'm being dramatic.

There are things that will come up as we are validating our first hypothesis and I don't want beneficial ideas to go unshared because they are currently out of scope.

Okay, good point. Up to now I have been really hesitant about this Foundation idea because it seems like a black box that is impossible to implement. I like the idea that we finalize a first list on the Outline of Facts (#2) and then we start populating it, even if it isn't finished before the first prototype.

One Idea for Expansion

Here's one proposal (and yes, let's look at many more, keep spitballing). I am thinking of a foundation as in a house. It's solid and never moves. Its definition can't ever change. So this layer is the absolute most indisputable. Above that is some set of data that is slightly less so, we'll call it the first level, and then one level above that that is even slightly less less so, the second level.

I am wondering if we can leverage the users to submit the data sources (urls) to be included and also rate the data. Or we look at its popularity in all the Takes. This could be useful for breaking news.

A newly submitted or included URL (a URL included for the first time by a Take and not yet appearing in any others) doesn't appear in the house yet. As it's used more and people trust it, it's moved to the second level. It's popular but it's unverified (and clearly marked - maybe some trending badge and some unverified badge) As it's used all over the place it's moved to the first level. It's more trusted now. The tricky part would be moving it to the foundation. Either it passes some threshold in the user base / rating system, or it's approved by some control board (like a CM kind of thing). Who decides really that it should be in the foundation?

I know my idea has a lot of bias problems, but I'd like to be thinking about how users can submit stuff that isn't in the foundation. I'd also think about how they can affect whether stuff is included. I think if we don't include users in that process, the foundation never grows, or the scope of the foundation grows too quickly and there are too many gaps in the data. (If you think about it like a number system - we can define bigger and bigger sets of data really easily. But writing them all out in each giant set becomes harder and harder.)

@nedtwigg

This comment has been minimized.

Copy link
Collaborator

commented Feb 5, 2017

I think the problem of "what if the foundation never grows" is exactly the same problem as "what if Twitter never supports more than 140 characters". It means that the platform will continue to serve its original market well, but might fail to expand into new markets.

It's interesting that "what is a fact and how do we decide" is the most hotly contested issue we have. I think that's the most hotly contested issue in our entire culture right now, and lots of people have spent lots of money and effort trying to solve it:

We need to build something that users of Wikipedia and Conservapedia think is fair. Look at all the systems people have tried. Based on how much effort has been put into this already, I don't think it is dramatic to say that it is extremely difficult to succeed in this effort.

Consider these two missions:

  1. Our goal is to decide which facts are true.
  2. Our goal is to see what we can say using only the facts which are indisputably and obviously true.

Look how much effort has been spent on mission 1, and to what benefit? Why do we want to engage mission 1 at all, when mission 2 is completely unexplored territory, and can be explored without engaging mission 1 beyond "The Constitution and The Debates are true and unbiased".

When I look at any system for determining a "fact", I look at it through the lens of Black / Blue lives matter. If the foundation is only the Constitution and the debates, then both sides will be citing the same evidence - there is a shared foundation. If our system allows shaky eyewitness footage and crime statistics, then BlackLM will talk mostly about the footage, and BlueLM will talk mostly about the crime statistics - the foundation is no longer shared, regardless of what we call it. People can already disagree about which facts matter on Facebook or Twitter, and they can already decide which facts are real by choosing which channel to watch.

The fact that this issue is so important to us is a sign that it will be important to our users as well. Perhaps we could settle it with something like the following:

The Foundation will be completely static until we have at least 1,000 users in every U.S. State. From that point on, the Foundation can be changed using the same rules as the U.S. Constitution:

The Congress, whenever two thirds of both houses shall deem it necessary, shall propose amendments to this Constitution, or, on the application of the legislatures of two thirds of the several states, shall call a convention for proposing amendments, which, in either case, shall be valid to all intents and purposes, as part of this Constitution, when ratified by the legislatures of three fourths of the several states, or by conventions in three fourths thereof, as the one or the other mode of ratification may be proposed by the Congress; provided that no amendment which may be made prior to the year one thousand eight hundred and eight shall in any manner affect the first and fourth clauses in the ninth section of the first article; and that no state, without its consent, shall be deprived of its equal suffrage in the Senate.

That means:

  • A new Fact Rule can be proposed by a 2/3 vote of users.
  • That Fact Rule will be accepted into the Amendments to the Foundation if a majority of users in 3/4 of the states agree to it.

A little more in-depth:

  • The Foundation is a list of Fact Rules.
  • A Fact Rule defines a list of Facts.
    • For example, if the Fact Rule is "Video Footage from Presidential Debates 2000 to present", then the Facts are the footage from "Oct 3 2000, Oct 11 2000, Oct 17 2000, Sep 3 2004, Oct 8 2004, Oct 13 2004, ..."
    • If the Fact Rule is "The content of the U.S. Constitution" then the Fact is the full text of the U.S. Constitution.

I'll be very honest though - even this is quite scary to me. Early adopters tend to be young, and the young tend to be liberal. If our Foundation sways left, then we're no longer "see what we can say using only the facts which are indisputably and obviously true", we have started to challenge our users on "which facts are true", and I think the moment we touch that point we have failed. I think we'd need some mechanism to ensure that our users' votes are scaled to match actual voting demographics. Not sure how to do that.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
3 participants
You can’t perform that action at this time.