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

Open Discussion #3

Open
romanman opened this issue Mar 6, 2017 · 2 comments
Open

Open Discussion #3

romanman opened this issue Mar 6, 2017 · 2 comments

Comments

@romanman
Copy link
Contributor

romanman commented Mar 6, 2017

[ @conor10 ]

Hey Roman, I’ve gone through the article & fixed up a number of
grammatical errors - you should have a PR for it.

Questions:

  1. entities such as doorways <- do you have a citation for this?

  2. Proof of identity by linking the social accounts, which were used
    in hack.ether.camp season 2, also can be useful if it's impact
    is not too big. <- do you talk about how you weight different types
    of accounts? E.g. Linkedin is more trustworthy then Twitter

  3. First upvoter receives maximal reward, reward will decrease for every
    subsequent upvoter <-surely this means there’ll be a rush to get the
    first upvote in all new content, and subsequently it will die out,
    causing the number of upvotes to be a diminishing measure of actually
    how good the content is - e.g. weighting of 15 versus 10 is actually
    more like 50 versus 10 in a traditional karma system such as Stack Overflow or Reddit

  4. Having simple picture of the final proposed solution would be useful

  5. An overview at the start of the proposed solution would be useful

  6. Could you reference then Stack Overflow & Reddit repuations systems
    versus what’s being proposed? As they’re very well established

[ @iurimatias ]

seems to the one in which it's attempted that, the amount of work
X requried to get the reputation score Y for Z number of accounts
would be the equivalent or even more than doing the same or less
work X to get the same reputation score Y with just 1 account
my concern is that, this systems are always vulnerable to
'wolves in sheeps clothing' sybil attacks
i.e malicious users that do everything right in the beginning
who then increase the reputation of their own sybil accounts
until the time is right to do the attack or manipulation
systems with tokens can discourage that bkoz there would some cost involved
I'm also concerned about the rewarding users for upvoting, looks like a way to game the system
perhaps a good alternative would be to rewards the first users who upvote something that later becomes popular

@romanman
Copy link
Contributor Author

romanman commented Mar 6, 2017

@conor10 : thanks for your feedback, I'll try to answer here:

  1. entities such as doorways <- do you have a citation for this?
    @evilsyncope , can you help here?
  1. Proof of identity by linking the social accounts, which were used
    in hack.ether.camp season 2, also can be useful if it's impact
    is not too big. <- do you talk about how you weight different types
    of accounts? E.g. Linkedin is more trustworthy then Twitter

The ignition of the new system will involve some bonus for people from the
previous events. Regarding social accounts contribution: clearly it has to be
valuable in the system including reputation (karma) increase. Linkedin is more
trustorworthy then Twitter. - agree.

  1. First upvoter receives maximal reward, reward will decrease for every
    subsequent upvoter <-surely this means there’ll be a rush to get the
    first upvote in all new content, and subsequently it will die out,
    causing the number of upvotes to be a diminishing measure of actually
    how good the content is - e.g. weighting of 15 versus 10 is actually
    more like 50 versus 10 in a traditional karma system such as Stack Overflow or Reddit

You are right the outcome can be perverted but we do need to give some motivation
to people to act - especially once the community is still few hundred people.

  1. Having simple picture of the final proposed solution would be useful
  2. An overview at the start of the proposed solution would be useful

We trying to have open discussion here. There are serious questions regarding current
reputation system as Reddit or Twitter. We have to answer it first And later to bring it to
short very clear definition.

  1. Could you reference then Stack Overflow & Reddit repuations systems
    versus what’s being proposed? As they’re very well established

I wouldn't put in one claim Reddit and SO. Reddit is completely fucked up:
https://www.youtube.com/watch?v=FxNvUWN3vYk

Regarding SO, I think they are much better - but there is no much gold to motivate
hack on it - if you have idea how to test it - I am curious

@romanman
Copy link
Contributor Author

romanman commented Mar 6, 2017

@iurimatias : The basic idea for now is to give karma on good valuable content -
nothing new here. The 2nd feature that should help here is the HKG deposit -
https://github.com/ether-camp/public/blob/master/reputation.md#hkg-deposit

The idea of the deposit in short is to have combination of good content and value
invested. In that case there is no motivation to grow sybil - while you can grow one
valuable account.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant