Re: Crazy idea: How trust in darknets enables secure democratic censorship

Matthew Toseland wrote:
> Here's a really whacky idea I came up with on the train back from
> Strasbourg (please read the whole email before flaming me):

*snip voting idea*

not only do I think this won't work for philosophical reasons, but I think 
it underlines one of the most fundamental *weaknesses* of the Darknet. Let 
me explain:

- Firstly, One of the first things that came to mind when you began this 
quest to determine "trust", was that it creates a very real chance of 
"groupthink" within freenet. Why? Because, while it is difficult to quantify 
when the method to *determine* this trust is still largely unknown, the 
reality is, there will likely be some "like-mindedness" in those who *are* 
determined as "trustworthy" - this is further underlined by numerous 
comments that the non-dark-freenet would still be there for the "masses" - 
the implication is they won't be trusted and likely will have no means to be.

Now, take this a step further. Lets assume for a moment a trust system 
similar to PGP, where a number of existing nodes/Toad's 
choices/known-good-nodes/etc become the "base" for the darknet - it's 
assumed that you will need to somehow gain the trust of these core people to 
join the darknet itself, which means, at some point, you'll be gaining the 
approval of the "person" behind the node, in some way.

Now, in a global community where in some parts of the world things like 
same-sex marriage are a done deal, and in others it's a crime worthy of 
execution, for example, how do you propose to ensure you have an acceptable 
cross-section of "minds" behind the darknet? It's seems inevitable to me 
that the "prevailing winds" of the collective morals of this core group will 
go a long way to creating the prevailing moral authority for darknet, even 
*without* a voting system in place - since you're not likely to "trust" 
individuals with differing views from your own in most cases, and unless 
this "trust" mechanism is somehow disconnected from people (which would mean 
that a nefarious party would need only complete these machine requirements 
to enter the darknet, so is likely not to be the case).

Now, add voting to the mix. Allow this set of "prevailing morals" to be 
*enforced* within freenet. This about guarantees, given the likely nature of 
trust, that the aggregate "morality" of the founding Darknet nodes would be 
preserved within freenet itself, since these founding nodes would have, 
through overt act or merely by likelihood-of-association, chosen like-minded 
trustees, who would vote as they do.

You don't have to go to extremes like child-porn to see where this kind of 
thing could have real-world consequences. Just take a few minutes and look 
at the US mass media, even in the "land of the free" people are deeply, 
deeply divided over issues far more benign, some of which on religious 
grounds, some on moral, some on traditional grounds, some due to reasons I 
have not even personally considered. Do you honestly believe that your 
"trust-based" system will truly encompass even that diverse a set of 
thinkers, in one country, let alone the rest of the world, especially in 
cases where people may be far more polarized on certain topics?

I see large cultural voting-bloc-type situations that, largely, will 
maintain the status quo. But that's the thing - the "status quo" is exactly, 
IMO, what the freenet should NOT be. You *want* the radical, free-thinking 
underbelly of the world to have somewhere to go where they can dissimate 
information that *may be* morally objectionable to large swaths of the 
populous, even the populous of freenet (think religious or cultural or 
political here - Tibet, scientology, women's rights, etc). What good will 
freenet be if someone who thinks "against" the prevailing beliefs of a 
region (but is otherwise "moral", for whatever that term can mean in this 
context) is simply "Voted down" by those within their own culture that may 
not agree?

I think the voting would lead to almost a "constitutional" form of 
"group-morals" that may or may not actually *be* altruistic - I think you'd 
find human nature would lead to large groups of people voting down 
information objectionable to *them* on cultural or religious grounds, that, 
in the broader sense, should not be.

I think this will be a darknet-wide problem in and of itself, mind you, but 
voting will likely make the problem all the more extreme.

-- 
Ken Snider