What makes Facebook successful, and always relying on search bad.

Facebook became popular not because of the almost universal and explosive adoption but because it provides something that had been missing from communication on the internet for a long time. Most means of internet communication are centred around knowing people and having them in your MSN/Googletalk/Y!Messenger contacts or email address book, or whatever means of storing the people you know are. These are referred to as strong links, and strong links are good, because they allow for communication direct with those that matter to you most.
In a social perspective, these strong links can be identified mostly as close friends, family and work colleges. There are of course also weak links, these weak links are generally friends of friends, acquaintances, and professional contacts of various forms. The irony of this is that weak links are actually more important than strong links in many ways, social networking being a perfect by the numbers example of how, but we can also see town markets and church congregations as other ways that these weak links can become more powerful.
The importance of weak links is related to the value of strengthening the link, if you strengthen a strong link not much is achieved, a further solidification of the already existing relationship. Strengthening a weak link creates a new strong link, and therefore the value of this is immeasurably higher than a strengthening of a strong link.
At GCDS an example of google shooting themselves in the foot in this regard came up, although I didn’t have much time to explain my dislike for the feature I’ll cover it all here; Robert McQueen brought up that googletalk has a feature that will hide the contacts that you don’t speak to often, these people aren’t important and therefore shouldn’t be shown in the main UI, although you can jump through a hoop to find them if you need them. To me this is a fairly senseless feature as it takes away from the table the impulse action which can lead to the valuable result of strengthening a weak link.
What would be more interesting is if google had a sorting method based on your frequency of communication, but also promoted those weak links that may be relevant, relevancy could be based on the number of shared contacts, suggestions for new contacts could also be made; however, we’re now talking about social networking not just communication, instead of locking it down, expanding instant messaging to incorporate social networking features in the same way that the social networks are incorporating instant messaging features.
Google failing to do this already is unbelievable, Microsoft were talking about a 3 degrees style instant messager a long while back so the idea isn’t new. However developing the client in the completely opposite direction is a bizarre detour from the obvious.
How does this relate to search? and why on earth would I group the two of them together? In many ways search is broken as a concept, the most obvious problem with search, as I’ve heard brought up over the years is;
You need to know something about what you’re looking for to find it.
This doesn’t mean that all kinds of searching are bad, for instance if you do know enough about what you’re looking for, search is great. But in most cases a better solution is filtration, for instance filtering a list of files to produce a specific subset.
This all boils down to the one problem that computers have never really solved which is locating things, and making sure that you can find them again and again without ever really loosing track of them.
Solving this problem of locating things is a very difficult one, and is generally based on one of two methods, browsing or searching. These are polar opposites but are also very similar.

• Browse as neatly defined by googles define command is; feed as in a meadow or pasture; “the herd was grazing”. This implies a systematic method of acquiring an item
  with regard to the “locating things” problem, in essence, I look in one folder, it’s not there, up a level, next folder, it’s not there; ad nausium.
• Search again defined by google is; the activity of looking thoroughly in order to find something or someone. Sounds pretty similar to browsing right, but, in this case, the computer does the systematic locating of the thing, not us.

This isn’t the only solution, there’s also a slightly more powerful solution and that is to make sure that relationships between items are relevant, that strong and weak links describe and inform, and assist in discovery.
By solving the problem of “locating a thing” with a focus to the relationship between items of interest we’re actually offloading the task from our wrists in the case of browsing, and our memory in the case of searching, to our ancient inbuilt ability to forage for things.
A good example of where filtration and grouping assists in foraging rather than browsing is in the gnome-main-menu produced by Novell. Specifically in the control centre and application browser windows. These tiles have a nice large target size which is always a good start, but they’re also clearly grouped, and those groups are jumpable from the side, filtration is provided but as we’re not starting from a blank canvas we don’t necessarily need to know exactly what it is we’re looking for. This probably the earliest example of utilising the skill of foraging over systematic approaches like simple search and browsing.
In this instance foraging can be defined as; Locating an item of interest by following paths of highest probability utilising experience and understanding to locate the item.
This is a fundamentally different way of addressing the problem. Happily this is the direction that GNOME-Zeitgeist and Tracker VStore are headed in, using strongly described content in a space which allows us to use a probabilistic reasoning to locate items.
With foraging, again we’re exploiting the strength of weak links, allowing us to follow interesting paths to arrive at our target, or targets and possibly find new interesting things along the way.
For more information please read;

• Nexus: Small Worlds and the Groundbreaking Theory of Networks
• Wikipedia on information foraging
• Interaction Design on information foraging theory