I've been hearing all these predictions and statements. And they are certainly not that far off. These new(ish) technologies are certainly going through their growing pains, but they will certainly challenge the old(ish) ones.
The article about O'Reilly's presentation goes on to say:
Tim showed screen shots from a Microsoft Research project that could answer questions such as who you communicate with around this particular topic. The question that follows is how we build tools for creating networks and managing our contacts. These tools could end up as part of Outlook and proprietary software, or they could become a connection between Orkut and GMail. "We have to Napsterize the address book and the calendar so that we own the data about our social network but we are able to query our friends about who they know.
I agree with the vision. But, I am hoping to bring the conversation down a few hundred feet and talk some specifics about issues:
Firstly, distributed event listings and calendars are becoming the norm:
Marc Canter pointed me to RSSCalendar.com today. We've relaunched WhizSpark with crazy-mad-sortable-wicked-searchable-republishable event rss listings. Upcoming.org has been around for awhile doing similar stuff. SocialWeb.net, although local (and not using xml), has been publishing event listings on other sites for years. Evenevite /IAC is waking up to the fact that they can't control the event listings of the world (Holding breath on that one!).
So, I don't think distributed and sharable event and schedules are that far off.
But, distributed social networks?
I think there are some issues re: FOAF that need to be addressed, b4 social networks will integrate it in a meaningful way. And when I say meaningful: anything more than allowing them to import profile data and upload connections to be invited.
I've posted some questions up on the FOAF wiki regarding these issues and haven't gotten any answers.
So, here they are again in statement and question form.
Most social networks (friendster, ryze, linkedin, IM) create bi-directional connections. This is ideal for creating many connections quickly, because both people have incentives to create the connections. The incentive is that they can collaborate. Depending on the network, the collaboration can take a different form. However, for marketing relationships or "fan/nod" relationships, this isn't ideal. To make an analogy to political ideaologies: if you ascribe to socialism and think that all people are created equal and should be treated equally, bi-directional connections are ideal. But, unfortunately (of fortunately), each of us performs differently and each of us has a different status in society. So, this is where these social networks break down. Since, connections between people are not equal, the incentive for "high" status people to join and use these social networks wanes as more people join and
abuse the service.
Orkut is an extreme example of where this "jamming equality into unequal relationships" is highlighted. By forcing people to receive an invitation, there are a million requests for invitation that go out to the members. You thought receiving Friendster invitations got annoying, try receiving 150 please invite me messages to orkut. That is how many I have received in the last month.
LinkedIn is an example of where this type of connection really works. The system is designed to screen people b4 passing along messages or information requests. And ultimately, the goal of the users is to collaborate with people. Since Rupert Murdoch probably doesn't want to collaborate with the street vendor selling newspapers, this system works for this purpose. The business people that use linkedin don't just pass out bi-directional connections on a whim, which prevents people from wasting time with requests that don't deliver value to both parties. Bi-directional connections are suited well for finding and forming mutually beneficial business relationships.
Another type of prevalent connection is outbound uni-directional. Examples of this are address books, FOAF, evite & blogrolls. The connection is defined by one person (the sender) and no approval by the receiver is necessary. This is ideal when people want to show their appreciation and respect. Blogrolls, using this type of connection and are what created the infamous A-List of bloggers. (I read 280 blogs and have a link on my blog for each. However, only somewhere between 5 and 10 people have me in their blogroll.)
Outbound uni-directional connections are what allowed evite and hotmail to grow quickly, back in the day. And if we couldn't store our addresses in an address book, think how difficult it would be to use email.
However, this type of freedom to message who-ever we want, can result in unwanted communications. Since a spammer doesn't need permission to send email to an account, they use outbound uni-directional connections to send their shit. Interestingly, the solution that many people are using for spam-blocking is whitelisting, which is in effect, making email connections: bi-directional connections.
The last type of connection is inbound uni-directional. This type of connection is defined by the receiver and approval is either inherent or optional from the sender. Permission email marketing or double-opt-in marketing is the prime example of this. The marketer advertises a list and the receiver signs up and confirms that it is their address. There isn't really an equivalent of this in Instant Messaging in the US, but in Europe (I believe) permission IM marketing is fairly common.
Plaxo also uses inbound uni-directional connections. For example, I have sent my plaxo card to Bill Clinton, but he hasn't returned the favor. So, I gave him permission to message me, but I don't have permission to message him. I've signed up for eMarketer's mailing list, but if I try to reply, the message bounces. I give permission. And don't get it back. I receive emails, but can't respond.
In this "connection framework", does the fact that friendster uses bi-directional connections make it obvious that fakesters will never have a purpose? Whether they were real, created by a member, or created by Friendster themselves, there were many accounts of celebrities on Friendster. But, the whole concept is pretty ridicilous. Imagine if Britney Spears was forced to use bi-directional connections to communicate? How could she possibly use bi-directional connections to communicate with people like this? The only social network that a celebrity could join and use would be one that used inbond uni-directional connections, because the celebrity can allow people to subscribe to them; to be a fan; without being a fan back. The same logic applies to any media company. A media company cannott possibly listen to all of its listeners.
Here comes the commercial: My Company, WhizSpark, has also built a social network which relies on inbound uni-directional connections (see mailing lists). (We relaunched the site last week and would love feedback, btw.) We've designed the system for the promotion of events. Whereas evite uses outbound uni-directional links to get-people-together at mostly non-commercial events and generates revenue from online ads, and upcoming.org requires bi-directional connections to share free event listings, WhizSpark was designed around the purpose of promoting events where the promoter/planner makes money(or the event is a marketing expense). In this scenario, getting permission to market-to is necessary, and thus, we use inbound uni-directional connections.
So, to start addressing Tim O'Reilly's statement about how noone has reinvented the address book yet, I think we need to keep in mind all of the types of connections that are required by different people. In this "connection framework", It is easier to conceptualize what features of different communication and collaboration technologies/applications (IM, RSS, email, social networks, FOAF) will make sense for what purposes. Then, maybe our blog discussions can progress beyond what technology will win and what technology is the best. I know football is exciting and our politics have certainly regressed to two sides fighting it out like it is the super bowl. But in technology, there are certainly still some gray areas left. Right?