VgtA: Questions

Ask a New Question

Vote all the Answers you Agree on

How to avoid "spammers" and "trolls" in Vilfredo?
by giovani in Vilfredo

avoiding spams isn't an issue because it is really easy to identify (a "report spam" button could play well if a minimum fixed of users use it). Plus, using a good captcha should resolve the problem at the source knowing that most spams are posted by bots.

"Trolls" are another question, it is sometimes a matter of opinions, the lonely way I see is to open a question to statuate if the related question is a troll or not. It is not a really efficient solution but it keeps the democracy thing.
The traditional method on blogs is to moderate comments, i.e. remove comments that contain spam.

I propose that anyone be allowed to remove a question or proposal; however, anyone else can appeal the decision to remove, in which case a vote is held. If more than 50% of votes are in favour of keeping the question or proposal, it should be restored.
Untrusted/new users must fill a CAPTCHA. After a number of some real posts/votes, Vilfredo trusts them and the CAPTCHA disappears.
An old saying states: "don't go looking for problems, let problems come looking for you".

Vilfredo goes to Athens is quite a unique service. As such the way in which it can be abused is not going to take exactly the same form that we have seen on other web sites (wiki, blogs, etc...).
No doubt there will be people who will participate with the aim to bring attention (or revenues) to themselves; like spammers tend to do.
Regarding "trolls" the position is less clear, as some undoubtedly participate just to get attention, and with no constructive aim; while others, sometimes, seem to be "trolling" because they find themselves part of a minority who are not given their due voice. Vilfredo is here also trying to correct this situation: provided someone is willing to actively participate in a question (generation after generation), Vilfredo, because of its voting structure and its selection mechanism, should help a fair and complete exchange of ideas between different people. When this happens antisocial trolling behavior might lack the exact environment that would encourage it. In other words the more people are allowed to speak, and the more their vote counts the less they will tend to act as trolls. (Or act in ways that other people will regard as trolling, often people called "trolls" would not see themselves as Trolls). So, here, the troll problem might be less disruptive than on other services.
Not only the amount of people willing to act against the social etiquette might be fewer, but also the form will be different. Right now a person can interact (and abuse) in three separate ways:
the creation of a question.
the proposal phase of a question.
the voting phase of a question.
And it might be difficult to understand in some cases what behavior was really done with a disruptive intent, and what really represent a different, honest, opinion on that matter.

My suggestion is that we should not do anything until we a problem really emerges.
And then we should deal with it.

An example about this can be:
In the past we have seen people making a suggestion and then voting for their own suggestion. By doing this a member can make sure that their proposal will be in the pareto front. Although this behavior could be representative of a genuine divergence in opinions, it can also be representative of a person who is simply trying to show that the system is not going to work. In other words a person testing the system... and the patience of other members. If this happens a few times this is not a problem. When a person does this generation after generation he is, effectively, not participating in the discussion. Like someone sitting in a corner of a room, and just speaking to itself. There is nothing against that behavior, in real life, but of course we have all learned that in society it is not a constructive way to interact with others. So we avoid it. Here on Vilfredo, we have noticed in the first questions that if we make public who has voted for what, people will tend to act less in this way. In other words when partially stripped from their anonymity people will tend to act in a more social way. They still will do it, but now there will be a social cost in doing so. People who do it are seen as asocial. It is not a big issue, but it is in general enough that people tend to avoid it... unless they have a good reason. In which case the social cost also drops (of course you voted in that way in that occasion, because you thought that...). This is exactly the sort of behavior that we were looking for to let Vilfredo run.

This is an example of how Vilfredo goes to Athens can be abused in a specific way, and what form did the prevention take. Both were very unique to this medium, and would have been hard to fathom before the problem arose.
This suggestion is the synthesis of two different ones. One that suggested using CAPTCHA, and another that suggested to wait until a problem arise. Since the second was much longer, the first is shown in bold.

An old saying states: "don't go looking for problems, let problems come looking for you".

Vilfredo goes to Athens is quite a unique service. As such the way in which it can be abused is not going to take exactly the same form that we have seen on other web sites (wiki, blogs, etc...).

No doubt there will be people who will participate with the aim to bring attention (or revenues) to themselves; like spammers tend to do.

To avoid this once the problems presents itself we can use CAPTCHA:
Untrusted/new users must fill a CAPTCHA. After a number of some real posts/votes, Vilfredo trusts them and the CAPTCHA disappears.
Regarding "trolls" the position is less clear, as some undoubtedly participate just to get attention, and with no constructive aim; while others, sometimes, seem to be "trolling" because they find themselves part of a minority who are not given their due voice. Vilfredo is here also trying to correct this situation: provided someone is willing to actively participate in a question (generation after generation), Vilfredo, because of its voting structure and its selection mechanism, should help a fair and complete exchange of ideas between different people. When this happens antisocial trolling behavior might lack the exact environment that would encourage it. In other words the more people are allowed to speak, and the more their vote counts the less they will tend to act as trolls. (Or act in ways that other people will regard as trolling, often people called "trolls" would not see themselves as Trolls). So, here, the troll problem might be less disruptive than on other services.
Not only the amount of people willing to act against the social etiquette might be fewer, but also the form will be different. Right now a person can interact (and abuse) in three separate ways:
the creation of a question.
the proposal phase of a question.
the voting phase of a question.
And it might be difficult to understand in some cases what behavior was really done with a disruptive intent, and what really represent a different, honest, opinion on that matter.

My suggestion is that we should not do anything until we a problem really emerges.
And then we should deal with it.

An example about this can be:
In the past we have seen people making a suggestion and then voting for their own suggestion. By doing this a member can make sure that their proposal will be in the pareto front. Although this behavior could be representative of a genuine divergence in opinions, it can also be representative of a person who is simply trying to show that the system is not going to work. In other words a person testing the system... and the patience of other members. If this happens a few times this is not a problem. When a person does this generation after generation he is, effectively, not participating in the discussion. Like someone sitting in a corner of a room, and just speaking to itself. There is nothing against that behavior, in real life, but of course we have all learned that in society it is not a constructive way to interact with others. So we avoid it. Here on Vilfredo, we have noticed in the first questions that if we make public who has voted for what, people will tend to act less in this way. In other words when partially stripped from their anonymity people will tend to act in a more social way. They still will do it, but now there will be a social cost in doing so. People who do it are seen as asocial. It is not a big issue, but it is in general enough that people tend to avoid it... unless they have a good reason. In which case the social cost also drops (of course you voted in that way in that occasion, because you thought that...). This is exactly the sort of behavior that we were looking for to let Vilfredo run.

This is an example of how Vilfredo goes to Athens can be abused in a specific way, and what form did the prevention take. Both were very unique to this medium, and would have been hard to fathom before the problem arose.

Generation=2;
Recently 1 human beings voted on 5 possible solutions,
1 produced by 1 human being(s)
and 4 inherited from the previous generation.
Minumum time: 3 days;
Maximum time: 7 days

Propose an Answer to a Question

Questions that reached unanimity

If you do not agree with tha answers you can reopen them, by adding other proposals.

How to avoid never-ending questions? by giovani in Vilfredo
Agreement found between: vilfredo jb555 BrendanP giovani

If a question is never-ending (in a loop) it has not yet reached consensus. If solutions or proposals continue to be submitted, it is likely that the person proposing thinks there may be a better solution. Which should (and does) trigger a new round.

But what informations of the past consensus should be shown?
As the question progress from consensus to consensus, it should be possible not just to see the history of all past rounds (as it is right now), but to see specifically the past consensus. Something like:
we discussed for n turns;
we reached consensus m times (with m <= n );
those were the consensus that were discovered:
...

Even better; not all the past consensus should be shown, but only some:
A consensus A hides a consensus B if both are true:
And then we show all and only the past consensus that are not hidden.
So new consensus should (only) hide older consensus that were less (or equally) popular.

How should we handle the "wall of text" problem? by vilfredo in Vilfredo
Agreement found between: battocchia vilfredo Ford canderson Ed Pastore

Abstract:
Require abstracts for long articles; calculate a complex score of the proposal body; give visual feedback of that score; and present proposals in order of lowest-score first.

Proposal:

1. Proposals longer than 1,000 characters require an abstract. Abstracts are hard-limited to 500 characters. (These numbers could be tweaked.)

2. Give visual feedback about the quality of writing of the proposal body, perhaps in the form of a bar across to top of the input box which contains a gradation from green to red. As the difficulty score of the input goes up, the bar fills up (or a slider moves) toward the red side. Also, the number score is shown. Additionally as the score goes up, more and more text warnings/FAQs start to appear (using CSS visibility), advising against length/complexity and giving tips on how to write more concisely. Metrics for the difficulty score could include:
- SMOG score
- Length of proposal - weighted heavily
- Number of other proposals by same author on same question (ie, a power law)
- Possibly "readability-votes" by other users
- Possibly others (though note SMOG is pretty inclusive. See source.)
3. Proposals are presented in order of their difficulty score. So the simpler your writing, the more likely your proposal is to appear at the top. This eliminates the need to build a karma system right away, while still providing a strong incentive (but note, it is not mutually exclusive with a karma system, and one certainly still could be implemented).

(This proposal is a synthesis of several others.)

How can the brainstorming potential of 'Vilfredo goes to Athens' be complemented by microblogging? by grahamperrin in Vilfredo
Agreement found between: vilfredo canderson jb555

A certain amount of Vilfredo's process happens outside of Vilfredo. Talking on Google Chat, or on Twitter, about the different positions, helps integrate the positions. This is fine with small numbers of participants, but as Vilfredo grows, this cannot scale. We need faster, more informal, ways of talking about proposals than the formal Vilfredo system. Integrating with Twitter might be one way to do this, but people might not want to clog up their Twitter stream with lots of Vilfredo chatter.

How can we tell is Vilfredo is successful? by canderson in Vilfredo
Agreement found between: vilfredo LukeSchubert canderson

I think the way to see how successful is Vilfredo in each question is by seeing how many people have voted on the last round of it (the one where an agreement is reached).

There are questions where many people have participated, but at the end only few have remained. Since the users receive email when they need to take action we can assume that they consciously chose not to participate anymore. Which is an implicit way of saying that over there Vilfredo has failed. So maybe without asking more information from the user we could show for each question a graph with:
on the x lines the generations,
on the y lines one line showing the number of people participating in each generation (either voting or submitting)
a second line with the total number of people who have participated.

This graph should give a pretty clear idea of how each question is going.
A successful procedure should show:
nearly everybody who participated should be present in the end;
the number of people participating in each generation being either stable or growing,

Viceversa a failure should show the number of people decreasing over time, as people drop off out of boredom or frustration.

Also, eventually there is a plan to keep track of where does the material from the proposal come from (this was copied from this proposal, this from that), and we will be able to say how many people have indirectedly participated in composing the final proposal.

This also should give a fairly good idea of how successful was Vilfredo applied to a particular question.

What should be the semantics of Vilfredo? by canderson in Vilfredo
Agreement found between: canderson vilfredo

The semantics used should grow with Vilfredo.

Soon we should introduce the words:

being derived from

which should mean that a proposal has been generated by copying another proposal and modifying it. Thus the two proposals would have a (usually strong) syntactic similarity.

Once the being derived from relationship is there we should be able to draw a phylogenetic tree of proposals. Showing each users what branches does he endorse.

Maybe Phylogenetic Tree should then be another word.
Of course also
Pareto Front
Supporting
Ignoring
User

Who must be involved in a meta-question for it to have standing? by canderson in Vilfredo
Agreement found between: canderson vilfredo

Right now the community is still small. As such the programmers are reading every question. In future probably the problem might open up.

I suspect some of the programmers responsible in VgtA should participate in the questions, and someone who is willing to code for the changes that the question proposes.

But then this leaves open the question of the general community. How many people should be involved, and what kind of advertisment should the question get.

Probably we can ignore the issue of how many people participate. Since this method of voting gives veto ability to each person, we can assume that who has not participated either is not interested in the community or it is not interested in the question.

In other words we need to make sure that the person who are interested in the community, will only have a fair chance to participate in every question concerning it.

For this I would suggest that all those questions should be discussed in the same place: room Vilfredo. And should have a minimum update time of 3 days. Also it would be good if each room had an RSS feed to permit to anyone to keep an eye on every room they want to participate in.
Existing developers do not need to be involved in the process, as the project can fork, and other developers can build another Vilfredo-style system elsewhere. The algorythm is not copyright, it is open source.

Why should Vilfredo be open? (or why not?) by canderson in Vilfredo
Agreement found between: canderson vilfredo

Vilfredo is still not User Friendly enough. The whole UI have to be re-planned.

In the net, once an interesting project it is well known it tend to spread like wild fire. When this happens it is a great opportunity for the project to be known. Unfortunately people tend not to test the same project two times. If they try it one time, and they did not like it, they will hardly try it a second time. For this reason it is better that before we open up Vilfredo, we work on the user interface, and fix all the little problems that can stop people from participating.

Some of those are:
-The wall of text discussion should reach an end, and we should implement what have been decided."
-The user interface is still in great need to be restructured. Not only should it be moved to css+xhtml+ajax; but what can be seen should change.
-There have to be better descriptions about what should the user do, and they have to be embedded inside the system. A system of red boxed messages that tell the people what they should do, for example
-Recovering of part proposal must be made easier. The history should be visible on the side of a question.
-A series of question marks should be pplaced all around, giving to users the possibility to know what is what. FOr example a question mark near the place where the user can look at the FAQ about rooms.
-Possibility to log in with open id, should be there.
-The questions about Vilfredo should be moved to the room "Vilfredo". At least the once that are supposed to have "legislative power" over Vilfredo. And a link to that room should be placed on the Header.

And so on.

In any case right now Vilfredo is already much more open than before. People can log in using Facebook, which already lowers the entry barrier quite a lot
Once you permit to people to look at what is going on, without logging in, the problem becomes "should they be allowed to propose solutions/vote without being logged in". It is easy to say that they shouldn't. But the truth is that in many context they actually can and should.

In particular since the pareto front does not change if the same person votes more time. Provided he always votes in the same way.

But still not everybody will want their question to be open to anyone. And in particular to people who are not logged in. So we need to make some major changes tied to that, to make it right.

But changing the code to let people view the answer will either require the coders to write even more code to handle whays to tell the people they cannot participate unless they register. Or we need to handle people who write questions, and then are pissed off because they lost them.

Instead we want to make the process of letting people in as smooth as possible. But while we are planning this smooth process, the website remains in pre-alpha stage. And is only available if people register in.
For many reason. Here are the main two.

1) Vilfredo is still not User Friendly enough. The whole UI have to be re-planned.

In the net, once an interesting project it is well known it tend to spread like wild fire. When this happens it is a great opportunity for the project to be known. Unfortunately people tend not to test the same project two times. If they try it one time, and they did not like it, they will hardly try it a second time. For this reason it is better that before we open up Vilfredo, we work on the user interface, and fix all the little problems that can stop people from participating.

Some of those are:
-The wall of text discussion should reach an end, and we should implement what have been decided."
-The user interface is still in great need to be restructured. Not only should it be moved to css+xhtml+ajax; but what can be seen should change.
-There have to be better descriptions about what should the user do, and they have to be embedded inside the system. A system of red boxed messages that tell the people what they should do, for example
-Recovering of part proposal must be made easier. The history should be visible on the side of a question.
-A series of question marks should be pplaced all around, giving to users the possibility to know what is what. FOr example a question mark near the place where the user can look at the FAQ about rooms.
-Possibility to log in with open id, should be there.
-The questions about Vilfredo should be moved to the room "Vilfredo". At least the once that are supposed to have "legislative power" over Vilfredo. And a link to that room should be placed on the Header.

And so on.

2) Once you permit to people to look at what is going on, without logging in, the problem becomes "should they be allowed to propose solutions/vote without being logged in". It is easy to say that they shouldn't. But the truth is that in many context they actually can and should.

In particular since the pareto front does not change if the same person votes more time. Provided he always votes in the same way.

But still not everybody will want their question to be open to anyone. And in particular to people who are not logged in. So we need to make some major changes tied to that, to make it right.

But changing the code to let people view the answer will either require the coders to write even more code to handle whays to tell the people they cannot participate unless they register. Or we need to handle people who write questions, and then are pissed off because they lost them.

Instead we want to make the process of letting people in as smooth as possible. But while we are planning this smooth process, the website remains in pre-alpha stage. And is only available if people register in.

In any case right now Vilfredo is already much more open than before. People can log in using Facebook, which already lowers the entry barrier quite a lot

How should we proceed to open up Vilfredo? by vilfredo in Vilfredo
Agreement found between: vilfredo grahamperrin canderson battocchia

If we can develop a standard format for proposals (as in the Metagov Standardization project), then it can simply be a stage in an open process. It would be fantastic if questions could be ported from one system to another depending on what is appropriate for the discussion at that time.

What shall we do about proposals that are against the law? by vilfredo in Vilfredo
Agreement found between: vilfredo jb555

There are two classes of illegal proposals to consider:
1. There are those where the proposed resolution is illegal, were it to be adopted.
2. Those where the very act of writing the proposal in a public forum is illegal.
Regarding the first:

The system presents us with a mirror of what the community really thinks of something. As such any distortion would modify our perception of reality, making it less faithful.

But something could be done in general. People should have the possibility to comment and explain why they are not endorsing a particular proposal. Those comments could appear during the next endorsement phases. If a proposal is against the law, people could comment, and clarify why they are not endoring it. Thus giving the possibility to other people to rewriting it, in a different way that follows the law!

A different proposal will speak about the second case.
There are two classes of illegal proposals to consider. There are those where the proposed resolution is illegal, were it to be adopted. And there are those where the very act of writing the proposal in a public forum is illegal.

In the first case, the fact that the proposal could not be reasonably instituted might weigh against its viability as a solution. It depends on the goals of the question. So long as participants know that enacting the proposed solution would be against the law, the VgtA mechanism should address this pretty well through endorsements.

The second scenario could potentially be a bigger problem -- here, the very existence of the proposal text on the site could constitute libel or defamation -- but Wikipedia has successfully defended against this threat in US courts (see this CNET article on the topic). The court ruled, essentially, that the website has no liability and that users have to be responsible for their posts.

Nevertheless, to defend against the possibility of the second scenario, vilfredo.org may want to present legal text to a user submitting a proposal that states the terms and conditions of that submission. These terms should exempt vilfredo.org from any liability and state that the liability will instead be assumed by the user submitting the content.

What are the most URGENT things to be done on this website? by vilfredo in Vilfredo
Agreement found between: jb555 vilfredo

during the proposal phase, one should be able to 'modify' an existing proposal more explicitly. Perhaps where the inherited proposals are listed, the listing includes a link to 'modify' which will only insert the text of the existing proposal into the answer box.

this seems trivial, but it may encourage people to improve on existing proposals rather than only use the proposal phase to give new ideas.

How should we refer to this website? by vilfredo in Vilfredo
Agreement found between: vilfredo Ed Pastore jb555

We shoud call it "Vilfredo", as in: You know you should really ask "Vilfredo" about this? Or what would "Vilfredo" say? Or I added a message on "Vilfredo".

What Domain Name should this Website use? by vilfredo in Vilfredo
Agreement found between: vilfredo Ed Pastore jb555

Vilfredo.org is available. It has a nice ring to it and clearly there is a connection there.