This is a serious blog post, even though the title suggests it immediately. As we move forward with our blog, but the social web, is - at least sometimes - spoken rather than written language announced. And when asked how I add an item to the name of the graph search, and the social graph, and searching the social graph and, moreover, is a bit of Open Graph, I was somehow "graph fuss" in the sense . I hope this is understandable.

The main reason for today's article is of course the Facebook Graph Search. Recently, whip them messages through the online media, before this new - search warn of Facebook - and absolutely overdue. Basically on any of the major media portal you will find instructions on how to get and its can protect privacy . Even Facebook itself has its users discreetly, so unobtrusive that you can notice it though, but do not see much of the alert regular users. In public, the graph search is therefore represented primarily as a threat to the private sphere, which of course it is. Still, this is not a satisfactory answer to the question: What is the Graph-Search?

After we have because sometimes "janz asked stupid", we approach the matter from the beginning and make hard: The graph consists of two search words and search graph. We now consider individual times.

And here we are again at the title of the article. With "graph" could mean two things namely Facebook, the social graph and the Open Graph

Already almost a year ago has Christina Liese on reachblog.de delivered a nice explanation for the social graph:

The heart of Facebook is the social graph that represents a conceptual for a sociogram basically. In the context of the Internet sociograms represent the personal connections between people, how they relate to each other and what they are interested.

In the social graph of Facebook so the connections of each user are recorded to each other. Friends, friends of friends, interests, places visited, photos, interactions among themselves, likes etc.. How can the relations that represent the similarities and differences of the users and as appropriate. There is, for example, assumed that the social graph of the essential evaluation criteria in Facebook's so-called EdgeRank algorithm represents the composition of the news streams of user controls.

The Open Graph is again more or less an extension of the Social Graph While the latter namely, everything is limited to what happens on Facebook itself, the Open Graph has now delivered the interface to the outside. That is also an open graph, since Facebook just opened to the outside. Result were numerous interfaces for apps and content on external websites. The most famous are Bestandtteil sicherliche the social plug-ins, making it possible to a page outside of Facebook to "like", ie "like" to press.

But the Open Graph, or rather the Open Graph Protocol also provides represents a standardized protocol for meta element, which are designed specifically for social networks, particularly Facebook. Because Rich Snippets can be created there with important information. For more information on these meta element and how the look closely, there are, among other things here . Then looks like the original one can read beautifully in a linked above (with quote) Article by Christina.

The graph search, Google is now rolling out gradually, as I said a search. But their data base are not like Google, a self-trumpedfound index, but existing information and data from their own users. With the Graph-Search a user can basically his personal social graph, search his own sociogram. So he can more easily find out what other people are doing so that are associated with it. The problem here is that any information that a user publicly shares with the Graph-Search can be found, since public content so basically can be in each "sociogram" is displayed. The more frightening is the information that Facebook has simple and many of these can be accessed. There are no fields but you can check as before at StudiVZ, but simply enters a record with the limitations in the search box. So far, therefore, the search graph is also a step towards a semantic search. Also, the information from the Open Graph, which users have done outside of Facebook, of course, flow into the search results with a.

Calculated as Facebook so from the huge amount of data a reasonably customized for each user on their search result. To sort the results by the way Facebook seems on the TF-IDF formula (for Term Frequency and Inverse Document Frequency) to access . But I have not yet found out for sure to which version of the term frequency is, ie whether it is a match to the WDF-IDF formula or not. In principle, it works in any case beyond the fact that Facebook wants to provide individually for each user the most relevant and most appropriate results and not generally popular.

And to eliminate the confusion last graph: Graph-Search is the original American name of the new Facebook search. The official German name, so the translation of Facebook, why is always: Search the Social Graph That which is often in newspapers, etc. of the Facebook social graph reads search, is an invention of the media, which - if mans exactly takes - is not that wrong. Nevertheless, something of course all adds to the confusion.

About the dangers of new search lot has been written and I do not even want to talk down hereby, on the contrary, I think that's also scary in detail, how to operate abuse here. Nichtsdestotz but there are also benefits, one for the users, who finally something on Facebook (again) can find the other but also for companies and their marketing. Since the search is not really finished it yet so can not say exactly how much and where exactly affects it, but this is definitely a field that is identified and local search, which is also attractive for SMEs. This opens up many possibilities with the graph search. Additionally you could even write your own articles, the Sebastian Socha but has already done in the usual good and complete manner. His article I can therefore highly recommend if you are interested in the topic, especially since he has attached a small tutorial for chain stores.

Overall, SEO, therefore, would actually move a bit more towards Facebook and Social Media Optimization of website n increases again in value.