Out of many value propositions which I could/(or could not) think for social media, one which has just struck me is "Reputation and Trust management".
In the ideal world of high Internet penetration, where every body is on some sort of social networking site. We will have every body connected to somebody (with some exceptions). The creation and identification of trust from reputation would be much easy, especially in the places where it is not that easy now!
Like we already can see role of recommendations on LinkedIn for getting a job. Recommendations do exist in real world but hard or costly to have. In online networks they exist in black and white, easy to access and persists for ever!
Some of the things we (Govt.) ought to do in physical world, where people intend to do some fraud for greedy reasons, like what is mentioned here in an article of Times of India. These problems could be solved effectively with use of online social networks and crowdsourcing.
I am taking some basic assumptions about India here:
1. Coming from India, I know that we have a society where most of the people know about people around them, which is very different from west! So, it is very easy for people to vouch for others (their near and dear ones).
2. Too many people, lot of contribution possible. Success rate for crowdsourcing the effort!
Now one of the problem posed in the article is that govt. want to issue id cards and then they are issued to people who do not even exist and execution fails.
So, what I propose is a wikipedia kind of solution where people can create people database. Let's call it peoplepedia.
Allowed functions for peoplepedia.
1. Creating a node, which is essentially a living person (say in India ) which can be later dead and the status change can happen accordingly.
- Node will contains the name.
- The gender.
- System will issue a unique id to every node.
- Anybody can create/ delete a node.
- The system will have a CAPTCHA like solution to prevent non-computers to make fake nodes.
2. Linking to another node, if one thinks they know each other.
- People can also indicate the relationship when linking. Like relative, friend, NGO, third party etc.
3. Bootstraping this database from existing social networks like Facebook and Orkut. (as chances of having real living users on those networks are high)
4. To users people have a graphical interface to search people based on their unique id, name or social graph neighbors.
5. Optional could be users adding more details to a node which could be private data like passport number etc but then which might need a login/password protection. Although this is not the real value of this database.
6. Since it is a people driven project and people identify the people they know in real life , which gives the trust in the numbers. Also, more people linking to a person will add to the reputation of a person and database can have an indicator based on who links to the node and how many links to the node. Let's call this indicator lively index, which would mean, the chances that person is real and exists/existed in this world ! However, the loss of privacy will happen in form of identifying somebodies social network but since we do not have any other details about that person (apart from name and sex) the privacy loss is minimum.
The cost of the project must not be high but the value it can provide with a small percentage of population actively involved would be high.
It would be a dynamic snapshot of number of real people in the country and also giving the break up (like men and women)
This platform can then be used for vouching for people, for distributing cards / other goods. Although it would not be highly reliable but could boost the confidence of identifying the real person.
I welcome readers of this post to improve the concept as it is in very initial stage and I have the feeling that it can be improved a lot!
