Open Framework, Information Management Strategy & Collaborative Governance | Data & Social Methodology - MIKE2.0 Methodology
Wiki Home
Collapse Expand Close

Members
Collapse Expand Close

To join, please contact us.

Improve MIKE 2.0
Collapse Expand Close
Need somewhere to start? How about the most wanted pages; or the pages we know need more work; or even the stub that somebody else has started, but hasn't been able to finish. Or create a ticket for any issues you have found.

Fusing Enterprise Search and Social Bookmarking

From MIKE2.0 Methodology

Share/Save/Bookmark
Jump to: navigation, search

Contents

Enterprise Search

Since the Industrial Revolution, companies have invested in automating repeatable processes to become more efficient and competitive. As a result, the role of the workforce has evolved from performing once manual, repeatable tasks, to critical thinking and decision making – employees have become knowledge workers. Knowledge workers are dependent on information to be effective at what they do. Over time, organizations accumulate a wealth of knowledge and information assets, some of it structured, most of it unstructured. Often, these organizations fail to provide a means to cohesively and intuitively discover these assets. As a result, knowledge workers waste time searching for information.

Enterprise Search solutions are designed to solve this problem. They allow decision makers to bring together information from federated sources, applying consistent ranking across all assets. This improves the diligence behind the decision making process as decisions are informed by a variety of disparate “perspectives”. Enterprise Search also inherently means transparency, and transparency makes it easier to mitigate potential non-compliance issues before they become public. For the business, these factors mean reduced risk.

Where Search Falls Short

Collaboration, however, is one element that enterprise search does not address directly. While it’s beneficial to enable information discovery, it’s even more beneficial to support the activities that occur after information is discovered. Often knowledge workers look to seek clarification about the content they find and yet have no idea who or where to turn to. It would be beneficial for the knowledge worker to be able to discover other knowledge workers within the organization who have found a particular information asset to be of value. Having demonstrated a shared interest, these knowledge workers could provide additional insight, have access to other relevant resources, or prove to be somebody worth collaborating with on knowledge development.

Social Bookmarking

Social Bookmarking systems have taken hold on the internet in recent years. Initially, the premise behind social bookmarking was simply that users could have a way to create and access their bookmarks (“favorites” in Internet Explorer) from any computer. Users could also dynamically categorize, or tag, their bookmarks to A) help with finding them later and B) correlate or group them. But it’s the progression of the social dimension of this bookmarking activity that is the most intriguing. Social bookmarking systems contain information about all users who’ve bookmarked a given URL, along with the tags they used to categorize it. This is significant for two reasons:

  1. The usefulness of a given document can be determined by counting the number of users who found it interesting enough to bookmark so they could find it later.
  2. A new view on how users perceive, or categorize, a given document surfaces by analyzing the aggregate of the tags associated with it (this is known as a folksonomy). If 50% of the userbase used the word “enterprise 2.0” to categorize a document, we can safely determine the document is about enterprise 2.0. Contrast this to the enterprise, where documents are generally categorized in formal taxonomies (or an “expert’s” opinion of how content should be categorized) which most knowledge workers find less than intuitive.

The best way to illustrate the usefulness of social bookmarking is by walking through the bookmarking process on the internet using del.icio.us (a popular social bookmarking service).

A del.icio.us Walkthrough

Suppose we found a good document about wikis we wanted to find later. We’ll bookmark it on del.icio.us:

Delicious Bookmarking.jpg

del.icio.us intelligently suggests the tags we should reuse from both our existing tag base and the broader, collective tag base. We select the tags we feel are relevant, then add notes about the document to expand on why we’re bookmarking it, what the document is about, etc.

After saving the bookmark we see part of the social dimension of del.icio.us:

Bookmark SavedByCount.jpg

The highlighted text, “saved by 1662 other people”, tells us there are others who also found this document to be of value. Who are they? Maybe they have other content we’d find useful since they’ve already demonstrated a shared interest.

After clicking this “saved by…” text we see a list of the other users that bookmarked this document, their notes, and an aggregate “tag cloud” (folksonomy) to showcase collective perspective on this particular piece of information:

User List, Notes

UserList.jpg

Aggregated Tag Cloud - Folksonomy

Folksonomy.jpg

Suppose, then, that we liked what “dmlgruppen” (third user in the user list) had to say about this document and wanted to see if he had links to other interesting information. Clicking his username brings us to his bookmark page:

Dmlgruppen.jpg

Here we can see “dmlgruppen’s” notes for other documents he’s bookmarked and view the tags he’s used to categorize them. If we found his interests interesting, we could create an RSS feed so that every time “dmlgruppen” adds a new bookmark we get a notification, or we could fine tune the criteria so that we receive a notification for only those bookmarks he adds with the tag “web 2.0”, for example.

Further still, we could create an RSS feed for the tag “enterprise 2.0”, so that any time a del.icio.us user adds a bookmark with this tag, we receive a notification. This allows us to stay up to date with information assets related to a given topic as determined by the community.

In this way, users on the internet not only share information but build communities based on shared interests.

The Value of Weak Ties

In 1973, sociologist Mark Granovetter wrote an article entitled The Strength of Weak Ties, in which he outlines the value of “weak ties” for information dissemination within a social network. Granovetter argues that “…the strength of a tie is a (probably linear) combination of the amount of time, the emotional intensity, the intimacy (mutual confiding), and reciprocal services which characterize the tie” (page 1361). The level of overlap between social networks for person “A” and “B” depends on the strength of the tie between them. If A and B have a strong tie, and B comes up with a new piece of information and tells A, this information will likely be diffused to a largely redundant set of individuals due to overlap within their social networks.

But consider person “C”, with which A has a weak tie and B has no relationship with at all:

WeakTies.gif

A also informs C, and B’s information now reaches a new audience through C’s social network (with which he has minimal overlap).

When discussing the value of weak ties, Granovetter argues that “Intuitively speaking, this means that whatever is to be diffused can reach a larger number of people, and traverse greater social distance, when passed through weak ties rather than strong” (1366).

Applying Business Context to Weak Ties

Harvard Business Professor Andrew McAfee shows that casual relationships (weak ties) within the workplace broaden the diversity of knowledge available to a knowledge worker. McAfee summarizes Granovetter’s findings, stating:

“..strong ties are unlikely to be bridges between networks, while weak ties are good bridges. Bridges help solve problems, gather information, and import unfamiliar ideas. They help get work done quicker and better. The ideal network for a knowledge worker probably consists of a core of strong ties and a large periphery of weak ones. Because weak ties by definition don’t require a lot of effort to maintain, there’s no reason not to form a lot of them (as long as they don’t come at the expense of strong ties).”

Subsequent research on weak ties at Harvard University by Morton Hansen has shown that weak ties help reduce information search costs, and in the previous section we illustrated this point by demonstrating how weak ties with shared interests (i.e. “dmlgruppen”) can expose us to new and likely helpful information assets.

Companies that rely on innovation for economic viability should recognize the value of informal networks (relationships formed outside of an organizational chart) by encouraging the development of weak ties to improve knowledge sharing and collaboration.

Fusing Enterprise Search with Social Bookmarking

Enterprise search solutions tackle relevancy in a number of ways. The most famous of these is Google’s Page Rank algorithm, which contextually rates content based on inbound links, or votes, and the weight (page rank) of the site providing the inbound link. This, combined with sophisticated text matching techniques, produces relevancy.

But what if we harvested the informal network’s opinion of information assets at search-time to allow the searcher to make a more informed decision about what to look at? Remember, users bookmark content they found helpful and plan on revisiting. Integrating social bookmarking “metadata” (including folksonomies) can improve relevancy and encourage the development of weak ties.

The first step is deploying an enterprise social bookmarking tool, and herein an organization has many options. IBM has invested in and internally uses an application called “Dogear”. Connectbeam has come out with an appliance model that combines social bookmarking with social networking and provided one of the first enterprise social bookmarking case studies (coincidently they integrated their software to a Google Search Appliance). On the open source front, Scuttle is a PHP-based solution modelled after del.icio.us. Regardless of what option is selected, it’s important the social bookmarking app have integration capabilities (using document URL as the key).

Enterprise social bookmarking allows the informal network to save and classify corporate information assets. Over time, a corporate folksonomy takes shape, and we begin to see the knowledge worker’s perspective on information assets (which may be very different from how said assets are classified in the corporate taxonomy). We can also start to gauge which information assets are helpful over those that aren’t – nobody wants to revisit unhelpful assets, so they won’t be bookmarked.

The second step is deploying enterprise search. Again there are many options here. FAST, Verity and Google all have compelling solutions. On the open source front, Lucene is a great toolset to base a custom-built search solution on. Regardless of the option, the effectiveness of an enterprise search solution increases the closer it gets to 100% coverage of corporate information assets. Fused with enterprise search, the collective intelligence of the informal network enhances the information discovery experience, as demonstrated here:

Fusion.jpg

In this example, the searcher is immediately exposed to the informal network’s perspective on information assets relevant to his search. The first result has been bookmarked 43 times and tagged, collectively, with “information management” and “wiki”. The searcher can choose to view other documents classified with one of these tags (i.e. “information management”) by clicking the tag hyperlink, or he can view a list of other knowledge workers who’ve bookmarked “Main Page – Information Management Wiki”.

Integration between the enterprise search and social bookmarking applications improve the information discovery experience by incorporating collective intelligence into the search results. Moreover, knowledge workers are able to discover others within the enterprise who have shared interests. It’s through this process that weak ties are established, and this results in better knowledge dissemination (as Granovetter shows us). Through these weak ties, knowledge workers broaden their social networks which increases the pool from which knowledge workers tap into for collaboration and knowledge development.

References

Granovetter, M.S., (1973) ‘The Strength of Weak Ties’, American Journal of Sociology, Volume 78, Issue 6, May: 1360 - 1380

McAfee, A., (2007) ‘The Ties that Find’, The Impact of Information Technology (IT) on Businesses and their Leaders, October

Wiki Contributors
Collapse Expand Close