Knowledge management, information management, content management, document management, etc. It’s not always clear what the distinction is between each of these. The difference sometimes depends on whether you are discussing a discipline or an industry. Even in the “industry,” it depends on whether you are dealing with products or services.  Based on how these designations are most often used, it seems as if knowledge management and information management are services, whereas content management and document management are primarily products. In reality, of course, a combination of software products and professional services characterize the industry anywhere along this continuum of knowledge/information/content/document management.

Industry conferences exemplify this combination of products and services. The exhibitors are mostly product vendors (with a smattering of professional/consulting services firms); whereas the conference sessions are primarily about the ideas, methods, and techniques of information or content management, that practitioners and consultants, not vendors, share. The recent Gilbane Conference in Boston, which took place from November 29 – December 1, typified this duality of products and services.

Gilbane, with the tagline “Content, Collaboration, Customers” and a description of “all about helping organizations apply content, web and mobile technologies to communicate,” focuses on “content management” – a sub-industry that is better known by its products of content management systems (CMS), than by its services. Statements by a number of speakers, however, made it clear that professional services were at least as important as the products.

In the opening keynote (a series of seven speakers presented “big ideas” or “bold statements”), the first speaker, Christos Cotsakos, emphasized the need for experts in the social sciences to serve the “attitudinal” side of content management by providing solutions for personalizing and customizing content, with the ultimate goal of turning the “volume of data into value.” When Scott Liewehr stated among his top ten ideas, “if we put as much effort into content as we do in the CMSes, we’d be better off,” he drew applause from the audience.

Content management has evolved to include newer collaboration tools and techniques that comprise what we call social computing or Web 2.0 (such as wikis, blogs, communities, social tagging, etc.) are now being used within the enterprise. This was the focus of several sessions at the Gilbane Conference’s Collaboration track. Even if such tools may seem self-explanatory to employees who have already been using them in their personal lives, effective use for business processes and workplace collaboration often requires planning, guidance, and training—which consulting services offer.

“It’s not just the tool,” were the first words of Lisa Welchmen’s presentation on web standards and governance, “It’s a governance problem.” And governance planning is a typical service of consulting service companies such as PPC. In the conference session on SharePoint adoption strategies, the moderator, Marc Anderson, introduced the topic of SharePoint as what you can do with it, not the issue of the technology. Good adoption requires governance, and governance is not just rules but also guidance, explained one of the speakers, Chris McNulty.  “Guidance” is, of course, the specialty of consultants.

Products may seem to be the bigger driver of an industry than services, but that’s not necessarily the case. This perception is largely due to the fact that software products have a better name recognition than most professional services, often with products that match the company name, logos, and brands.  Product vendors (with the exception of a few big names, such as IBM, Microsoft, and Oracle) tend be very focused in their product offerings, thus confined to a single sub-industry (such as content management, document management, or digital asset management). Consulting companies tend to span these areas of sub-specialization, so are not always so clearly identified with an “industry.” In actuality, to obtain the desired results, an organization should probably invest more in professional services (unless expert knowledge managers, content managers, and taxonomists are on staff) than in content management technology. 

Even if a company has in-house information management expertise, an engagement of an external consultancy, such as PPC, can still be valuable in bringing in a fresh perspective. At Gilbane, Cotsakos stated: “When you are immersed in your own environment, you are less likely to change it.”

After another year of attending Taxonomy Boot Camp, I find myself even more encouraged by the progress made within the field and the more central role taxonomies are playing in an organization’s information management projects.  In past years, much of the conversations at Taxonomy Boot Camp revolved around convincing sponsors or stakeholders of the need for a taxonomy, or even attempting to educate people as to what a taxonomy is or should be.  This year’s Taxonomy Boot Camp served as a clear demonstration of how far we’ve come.  Though these topics still came up, it is clear that most organizations have matured in their understanding and usage of taxonomies.  There are more full time taxonomy positions, more case studies of how business taxonomy is used, and certainly more educated conversations regarding the practical application of taxonomies.  All of this is good news for organizations in their quest to improve findability, usability, and overall information management and reuse.

I also took note of the central role that taxonomy tools played in the discussions this year.  I’m sure part of this is that Synaptica, Smartlogic, and Concept Searching were among the sponsors for the event.  Moreover, though, the topic of taxonomy tools was raised by event attendees.  This tracks with what I’ve been seeing in procurements as well as the overall trend for taxonomy design.  Organizations are learning to leverage enterprise taxonomies consistently across their various applications.  Organizations are also designing more complex designs that often include multiple metadata fields populated by multiple taxonomies to enable faceting and serve a myriad of different audience needs.  Because of these trends, taxonomy tools are becoming more critical to the success of these organization’s efforts.

Finally, I echo my colleague Heather Hedden''s observations regarding taxonomy governance. This was intended a central theme and I take a personal satisfaction in that--as taxonomy and content governance have been central to PPC''s efforts for many years. Successful taxonomy governance can counteract the natural trends of taxonomies "breaking" over time. Instead, taxonomy governance can guide the evolution of a taxonomy from good to great. If applied successfully, governance will empower business users to leverage taxonomy and ensure more consistent tagging of content. In addition, taxonomy governance can serve as a critical change management and communications component, tracking back to my first point: to better educate organizations on how taxonomy can and should be leveraged to enable the business.

A taxonomy is a very useful tool for organizing and finding content, but it is only useful if it is properly managed, maintained, and followed. Over time, a taxonomy could become out of date, or it could morph into a style and format different than how it started, and internal inconsistencies could appear. Even if you have the best-designed taxonomy, you cannot simply walk away from it once it is implemented. Policies, procedures, and documentation for the ongoing management of taxonomy are a crucial component of a taxonomy implementation or revision.  Collectively, this is known as the governance of the taxonomy.

At this month’s Taxonomy Boot Camp conference, I noticed that taxonomy governance was not a dedicated presentation topic, as it often has been in the past. Rather, governance issues were discussed within numerous presentations (e.g., with a slide on the subject). This seems to indicate that governance is no longer seen as new and different, but rather has become an accepted, standard part of any taxonomy activity. It’s helpful that individuals both new and experienced with taxonomies are reminded of the necessity of a governance plan.

Some of the presentations that mentioned governance included “Enabling Social Media Through Metadata,” “Successfully Advocating for Taxonomy in the Corporate World,” a vendor-sponsored luncheon keynote “Semantic Technology: Why?” and  a case study on “Microsoft’s Experience with Modeling Terms Sets in SharePoint 2010.” This final presentation had multiple slides on governance topics, including a bulleted list of the slide of “Business Benefits of the Taxonomy Governance.” The first day’s closing session, an update of taxonomy standards, also mentioned how standards support governance.

In some presentations, governance priorities were mentioned without even reference to the “g” word, which is a good thing. Some technical or knowledge practitioners may shy away from learning “governance,” thinking it should be the concern of only managers.  While managers have a responsibility for the policies, procedures, and documentation that comprise governance, everyone involved in a taxonomy (e.g., taxonomists and taxonomy editors, subject matter experts, user interface designers, indexers, system implementers) has a responsibility for carrying out the taxonomy governance plan. And if you call it something else, that’s fine. I remember that when I first heard of “governance” several years ago, my initial reaction was: “I’m not sure what that is, but it doesn’t sound very interesting,” even though I had already had experience in taxonomy governance without using the term.

Another name for a taxonomy is a “controlled vocabulary.” More precisely, a taxonomy is a kind of controlled vocabulary in which the terms are structured into hierarchies or facets. But if we return to taxonomy’s roots as a controlled vocabulary, then we are reminded that we need a plan for whom and how that control is carried out.

If you would like to learn more about taxonomy governance and information management governance in general, read PPC’s recent white paper Best Practices for Information Governance.

In an earlier blog post, I listed ways in which the design and structure (in addition to the terms) of a taxonomy can vary, and thus the fact that a taxonomy needs to be customized for each client. This is, of course, is what we do as consultants: serve clients’ specific needs. But in order for a taxonomy to function properly, it must also be designed according to industry best practices. Thus, a key task for us is to balance client needs and requests with standard taxonomy design practices.

Following Industry Standards

The industry standards for taxonomy are contained within ANSI/NISO Z.39.19: Guidelines for the Construction, Format, and Management of Monolingual Controlled Vocabularies. (“Taxonomy” and “controlled vocabulary” may be used interchangeably.) Topics covered in this book-length document include, among others: taxonomy term choice and format, term relationships, taxonomy presentation and display, taxonomy construction and testing, and taxonomy maintenance. Although some taxonomies are quite simple, lacking some features described in the guidelines, even these simple taxonomies should adhere to these best practices to the extent relevant so that they are useful and usable.

It is worth noting that the word in the title of the document is “Guidelines,” not “Standards,” implying some openness to variation. As consultants, we are not only familiar with the ANSI/NISO guidelines but also know when it is okay to diverge from them for a particular client and to what extent that is acceptable. For example, the guidelines describe the required characteristics of hierarchical (broader or narrower) relationships. However, in a taxonomy that has no associative relationships, it may be permissible to have a type of related term treated as a narrower term, such as an associated service listed under (as if narrower to) a product category.

Responding to Client Needs

We respond to client specific needs in a number of ways when we design (or re-design) a taxonomy. A PPC project usually kicks off with an onsite workshop that engages key stakeholders of a taxonomy. In this interactive workshop, while giving introduction to a taxonomy’s purpose and possible structures, we also learn from the participants the particular needs of an organization: what this taxonomy aims to achieve, who will use the taxonomy, what its scope of coverage is, and what the current search and navigation problems are.

We also engage in one-on-one or small group stakeholder interviews to obtain more insight into the current issues of information management and content retrieval for varied users with different roles. Also as part of the process of obtaining information of a client’s unique needs, we gather and analyze sample content within the scope of the taxonomy.

Another way that we respond to the specific client needs in the taxonomy design process is through facilitating taxonomy testing and validation. A taxonomy comprising several multi-level hierarchies is often a good candidate for a user card-sorting test during the design process. Validation of a taxonomy, typically performed near the end of all taxonomy projects, involves testing the taxonomy to index (tag) and/or retrieve actual content in the client’s repository.

Balancing Both

The industry guidelines provide the taxonomy framework, and from the client we gather information for the details needed for the taxonomy design. Usually the guidelines provide the flexibility to adapt to specific client needs. We often must balance responses to client requests with adherence to taxonomy design best practices. But that’s what keeps the job of a taxonomy consultant interesting!

We consultants provide solutions to our clients’ problems, and while there are often similarities in various clients’ information management problems, the solutions are never identical. There is no “one size fits all” answer, and this is especially the case in the highly customized nature of a taxonomy project.

A taxonomy project naturally involves customized terms and categories to reflect the scope and nature of the content and also respond to the needs of the target users. This is especially true for internal and enterprise taxonomies, but it is also the case for public-facing web taxonomies used for ecommerce or information services, since every vendor has its own unique strengths and content focus. Occasionally, a very standardized term-set (such as geographic places or industry types) can be licensed from a taxonomy publisher as a part of larger taxonomy solution. Taxonomy customization goes far beyond the terms included, though. The overall design and structure of the taxonomy varies in each case.  Taxonomy design and structure may differ in all the following ways (and perhaps more):

• Size, Growth and Change: Tens, hundreds, thousands, or tens of thousands of terms; stable in size or growing; relatively static or frequently changing
• Design Type: Primarily hierarchical, primarily faceted, hierarchical and then faceted at the lowest level of the hierarchy, faceted with limited hierarchies within one or more facet, or a traditional thesaurus design (term-focused with various relationships but no dominating structure)
• Depth: Typical number of hierarchical levels deep
• Breadth: Typical number of terms at the same level sharing the same broader parent category
• User Display: Hierarchical or faceted browse, alphabetical browse with or without variants included, search box with or without auto-complete matching, or any combination of these; and whether there are different displays for different users—content taggers vs. end-users
• Polyhierarchies: Whether terms can have more than one broader parent category and thus appear in more than one hierarchy and how extensive (common) it is to have polyhierarchies
• Related Terms: Whether there are additionally non-hierarchical relationships between related terms in different hierarchies and how extensive these relationships are
• Variants: Whether terms have “synonyms” (technically called non-preferred terms) as variants to support alternative searches by users and how extensive the presence of non-preferred terms is.
• Additional Term Content: Whether taxonomy terms include notes or definitions, URLs, or other metadata common for names or organization-type terms
• Search Engine Optimization (SEO): Whether taxonomy terms should be optimized for web search engine retrieval and to what extent (depth within the taxonomy)
• Translations: Whether to have foreign language or locale versions of each taxonomy term
• System Constraints: Whether the system has restrictions on term length or the use of special characters such as ampersands

The chance for contractors to bid for the recompete for the management and development of the General Services Administration’s (GSA) carbon footprint tool ended on August 26^th. While this is an important piece of work, ultimately providing an online greenhouse gas (GHG) inventory tool for the federal government, the most interesting thing about this particular procurement is that for the first time, part of the scoring was based on the winning bidder’s environmental credentials.

Specifically, GSA asked bidding contractors to declare their commitment to accounting for their GHG emissions and the degree to which environmentally sustainable products would be used in carrying out this work. What’s interesting here are the links between these evaluation criteria and the federal government’s aspirations to bring its supply chain into its own sustainability targets. It seems unlikely to be coincidence that the evaluation criteria in this RFP map directly onto the targets set out in Executive Order 13514, namely for federal agencies to reduce their scope 3 GHGs (Sec. 2, b, i) , and to ensure that 95% of all procurements meet the very same sustainability criteria (Sec. 2 h) set out in the RFP.

This starts to illuminate the ways in which the federal government is expecting federal agencies to meet the sustainable procurement objectives in the Executive Order, yet it represents a fairly light touch for now. GSA is not asking bidders to demonstrate a reduction in GHGs; rather, they are asking for evidence that potential contractors are sufficiently aware of government objectives to be, at the least, shadowing the requirements placed on federal agencies.

What this means for federal contractors is more significant. It marks the first steps towards a world where measuring and managing your sustainability is part of the license for doing business with the federal government. That is, of course, unless your business strategy is going to focus on the 5% of federal acquisitions that will not require evidence of the vendor’s sustainability practices.

As we’ve worked with countless organizations over the years to implement a wide array of information management technologies including websites, intranets, portals, content management, social computing, and document management tools, I’ve seen many of our clients focus on the technology itself.  As I’ve often noted, however, when it comes to successful information management, and most certainly successful knowledge management, the technology is simply the enabling component, not the solution itself.  Softer considerations including change management, governance, and communications are critical success factors.  At the core of the issue, really, is the content itself. 

The phrase “content is king” has been used and reused, but I prefer my NERD acronym.  To successfully implement and maintain an information management technology, you must ensure the content within it will initially be, and continue to be, NERDy (New, Essential, Reliable, Dynamic).  Users need New content, meaning as information is generated, they must have access to it immediately.  Users need Essential content, meaning they must be able to find what need to do their jobs or complete their missions.  Users need Reliable content, meaning they must be able to trust what they are seeing as accurate and up-to-date information.  And users need Dynamic information, meaning as the business changes, the information available to them must keep pace.  If these four simple criteria can be met (along with a host of design criteria), information system projects stand a good chance of success.

Many organizations mistakenly attempt to place any piece of content, that has ever existed within the organization, within their information management systems.  Most of that content, however, exists on a very long and very thin tail that will most likely never be sought after or useful to the users of the system.  As a result, it will serve no purpose but to be a burden to maintaining the NERDiness of content and, worse yet, a diluting factor to the ability to find the content users need most.  A simple rule when designing your information management system and the accompanying governance plan and content migration strategy is to first understand the level of effort around truly maintaining content.  Only populate your systems with content that is truly NERDy and you will find you are meeting the needs of your users in a way that is truly sustainable and efficient.

The federal government has been tracking its ‘carbon footprint’ since President Obama signed Executive Order 13514 way back in October 2009. However, only very recently have we been given an insight into what the federal government’s carbon footprint actually looks like and been able to start to take stock of the scale of it, the different types of sources contained within it, which agencies are important contributors to it, and most importantly—in the light of this new information—what strategies might the federal government deploy to reduce emissions in order to meet its targets.

The White House’s Council on Environmental Quality reported total GHG emissions from federal activities of 121 million metric tonnes of CO2e (mtCO2e) for FY2010 by. While this is only 0.2% of total annual US emissions, if the federal government were a country it would nestle between the Czech Republic and Uzbekistan as the world’s 35th largest emitter. If the federal government were a corporation its emissions would be larger than BP (65 million mtCO2e), Dow (42 million mtCO2e), and Ford Motor Company (5 million mtCO2e) combined. And, for those of you that like these types of analogies, if the federal government was a sporting event, let’s say the Super Bowl, the federal government’s annual emissions are the same as 242 million Super Bowls.

So where are all these emissions coming from and who’s producing them?

The Department of Defense (DOD) is significantly the largest federal emitter producing some 86,500,000 mtCO2e, the US Postal Service (USPS) rolls in at second place with 7,300,000, and the Department of Energy (DOE) comes in third with emissions of just under 5 million mtCO2e. In fact, 96% of all federal emissions come from just 11 federal agencies, all emitting more than 1 million mtCO2e in FY2010.

As far as the activities that create these emissions, the single largest source (40%) is from non-highway vehicles, aircraft, ships, and equipment – primarily from DoD’s activity, plus the fuel in the trucks that the Department of Homeland Security (DHS) uses. Next comes electricity consumption at 32.8 million mtCO2e (27%) then on-site fuel combustion in stationary assets (boilers and furnaces etc) at 10.8 million mtCO2e (9%).

That’s a lot of emissions, but it’s ok , because we’re going to reduce these emissions by 28% of 2008 levels by 2020… aren’t we?

Well, sort of. The 28% target only applies to some of the emissions. Because there are a series of potential exemptions listed in the EO itself, DOD and a number of other agencies have been able to exclude some 45% of total emissions from the targets meaning that the 28% target is only actually applied to 66 million mtCO2e. However, this does mean that the federal government is due to deliver annual savings of just under 20 million mtCO2e by 2020. This is equivalent to the emissions of nations such as Jordan, Estonia, and the Dominican Republic – or the entire State of Maine.

So what do the numbers tell us about ways in which the federal government can meet its target?

That’s the good news – by far the largest element (84%) of Scope 1 & 2 target subject emissions are associated with the operation of federal facilities, basically buildings of one sort or another. This means that tried and tested energy efficiency technologies, such as high performance lighting and heating and air-conditioning systems are available to dramatically improve this performance. Add in weatherization programs and the use of better building management systems and it’s reasonable to believe that with the application of a strategic approach to identifying, prioritizing, and funding these sorts of measures this target is very achievable at a positive net present value. This means that in the long run this program will save, not cost, the government money.

Meeting the Scope 3 target may be more challenging. Collectively Federal Agencies must reduce the six scope 3 categories by 13% from 2008 levels by 2020. 25% (600,000 mtCO2e) of this target will be delivered through reductions in transmission and distribution losses if the government meets the 28% Scope 1 & 2 target, which leaves a further 1.8 mtCO2e to come from the remaining categories. What these categories have in common is the need for a sustained and successful behavior change program as the federal government is going to need the willing cooperation of its employees in order to reduce emissions from employee commute (56% of Scope 3 emissions), business travel (25%), and off-site waste disposal (7%).

So, through a combination of technology and culture change the federal government is poised to tackle their GHG emissions and demonstrate leadership not only within federal circles but nationally and internationally as well. Federal leaders charged with delivering these environmental improvements will face challenges, especially given current budgetary restrictions, but they must always remember that investments made can save money over their lifetime. As such, these executives are leading the charge to cut the deficit as well as improving federal environmental performance.

Welcome to the PPC Energy and Environment blog, where my colleagues and I will be writing about pressing topics in the field of energy and environmental management.   

During my time here at PPC, I’ve helped our clients manage and respond to a multitude of environmental issues ranging from the cleanup of nuclear waste, to the emergence of new contaminants in surface and drinking water, to organizational carbon management.  I also have the pleasure of working with a team of highly talented energy and environmental consultants who have taught me that effective environmental management is about striking a balance between organizational goals, budget realities, and external drivers (whether regulatory or voluntary).  They have also taught me that as soon as you find that balancing point, it will begin to move in response to an ever changing landscape shaped by the economy, technology, scientific progress, public opinion, and the political climate. 

Through this blog, we will bring you the analysis of our top experts who will discuss current issues and future trends in key energy and environmental topic areas:

• Energy and Carbon Management
• Climate Change Adaptation
• Sustainable Transportation
• Sustainability
• Supply Chain Management
• Waste and Resource Management

In our first blog entry, Daniel Waller (PPC’s Knowledge Leader for Carbon Management) will present an analysis of the U.S. Government’s first ever comprehensive Federal Agency GHG emissions inventory. 

I think you will find the analysis and insight of our experts as intriguing as I do, and I hope that you take away from this blog some practical knowledge that helps you to navigate the energy and environmental challenges facing your organization and the ways these challenges intersect with information technology and Sustainable IT.  Finally, this blog is not intended to be a one-way flow of information, and to that end, I encourage you to share your questions, thoughts, and reactions to the topics that we cover.

Migration to the cloud promises to yield significant cost savings. These savings have eluded many organizations as the promise of virtualization and cloud computing have not lived up to the hype. A recent survey of 3,700 IT professionals conducted by Symantec showed that of the approximately 25% of IT managers who implemented some form of virtualization, many expressed disappointment that the expectations and benefits in performance and operational cost reductions were not achieved.

While respondents from the Symantec survey expressed general satisfaction with results from server virtualization, a third were disappointed in storage virtualization, private storage, and private or hybrid cloud computing. At the center of these complaints were issues related to scalability, security, and time required to deploy new resources to the cloud - issues similar to those originally cited as reasons to move to the cloud.

Gartner has conducted a great deal of research on cloud adoption. Based on their research, Gartner advocates moving traditional computing into virtualization to build individual private clouds, either operated by an internal IT department or with the assistance of private cloud providers. They emphasize that building a private cloud is more than simply adding virtual machines to physical servers. They estimate that almost half the x86-based servers already carry virtual-machine workloads. With the assistance of management tools like VMware and Citrix, projections are that this number could reach 80% by 2015.

IT challenges are not simply technical. Lessons learned so far indicate that while cloud computing offers the potential or lowering IT costs, user satisfaction must not take a back seat. It is important to work with the user community to develop key success factors including:

• Developing a comprehensive service management that delivers reliability, savings, security, and service;
• Working with functional groups to automate services, when appropriate;
• Defining service levels that are measured and reported to user community; and
• Applying sound project management and customer relationship management to ensure solid transition planning and implementation.

To ensure that an organization is prepared for the changes associated with cloud adoption, a cloud transition plan should be developed. Agencies need to assign a cloud adoption point person - someone in charge of coordination. Traditional mechanisms should be leveraged, such as a business case with alternatives analyzed, a staff plan with authorities and controls, and finally a clear definition of metrics that will both articulate the target and measure progress towards that target. The enduring lesson for me has been that as technology innovations are created, success depends on the core principles of sound planning, project management, and communication. Try as we may, we can’t run from these things.

Everyone is discussing moving to the Cloud in Federal IT. Identifying the need for a consistent federal strategy, the U.S. Office of the Chief Information Officer (OCIO) released its Federal Cloud Computing Strategy in February. The declared focus of this strategy is a Cloud first policy. Federal agencies will be accessed on how quickly they adopt cloud computing options and incorporate cloud assets into their IT portfolios.

Agencies must address the challenge of defining a formal approach to cloud computing. In absence of this governance, agencies will not achieve the promise of commodity computing and delivery of core services may suffer. In some ways, this cloud migration reflects the challenges agencies faced when services were first proposed to be delivered via the Internet. Most agencies struggled with the strategic planning aspect of migrating services to the web, and some agencies struggled to make their web presence and service offering relevant.

Developing an agency cloud strategy should not be completed by the OCIO alone, but should include Communications, Public Affairs, and representatives of the core mission areas.  Not all services should be migrated to the cloud. Many cloud services have experienced high profile outages, including those provided by Microsoft, Google Mail, Rackspace, Salesforce.com, and Amazon S3. While these outages were not overly long in duration, each lasting but a few hours, it underscores the need for planning and comprehensive governance to manage cloud computing for each agency. Potential IT outages will reflect on your agency performance so these decisions about which services are cloud capable should considered while defining an agency strategy.

In development of your agency’s cloud computing strategy, you should address these cloud strategy questions:

1. What are our agency’s target costs savings?
2. Has our agency rated our applications for complexity?
3. Has our agency assessed the legal issues?
4. Does our agency have a comprehensive data management strategy?
5. How accessible are our IT assets?
6. How secure do our applications need to be?

Before assuming what may be best for their agency, IT leaders need to clearly address these questions and, as the reality is, that they’ll learn that moving to the cloud isn’t a simple undertaking. It is an evolution and brings elements of trust and control into sharp focus for all involved.

Agency cloud governance should leverage existing structure but will likely prove inadequate to answer the questions the migration to cloud will introduce. Defining a decision rubric to address these questions and enable agency decision making for which cloud deployment model and cloud service models should be considered is vital for the near-term. Longer term for cloud incorporation into executive reporting and dashboards will allow agency management to assess cloud performance and make recommendations for improvements.

In addition to the goals of your cloud strategy, some new questions will need to be analyzed as part of the governance process. Perhaps augmenting the Exhibit 300, these questions reveal your applications suitability to migrate to the cloud. Proposed new questions are:

1. Which cloud models does your application require (private, public, hybrid)?
2. Does your application have a proposed SLA approved by management?
3. What protection is required for your data?
4. Have you completed a RACI matrix for continued application operations?
5. Does your application have a contingency plan in the event of a data breech?
6. Do you have a reverse migration plan in place in the event one is needed?

As is the case with many IT innovations, proper planning and governance increases the value to the agency. Having clear goals, a detailed strategic plan, and adequate controls in place will focus cloud migration efforts to agency benefit. Cloud computing has tremendous potential to provide essential government services at a much lower cost.  

Source: http://www.cio.gov/documents/Federal-Cloud-Computing-Strategy.pdf, Federal Cloud Computing Strategy, February 2011

Enterprise search refers to the technologies, software, and techniques for searching the internal content of an enterprise, as opposed to the public Web, but also encompassing disparate content from throughout an enterprise, not just from a single repository. As enterprise content inevitably grows in volume, the importance of effective search is also growing. Furthermore, it’s not just the total volume of documents, data records, and other files that is a challenge -- but there are also additional issues, such as retrieving non-text files, discerning the correct document among multiple versions, ensuring that legally required records can be found and that outdated records are removed, and discovering useful information not known and thus would not been directly entered in a search, etc. Whether to stay competitive, stay compliant, or just stay on top of the growing body of enterprise content, effective search is crucial.  

Technologies have improved, but technology alone cannot provide all the solutions to successful search. This became quite clear at last week’s premier conference in the field, Enterprise Search Summit, held May 9-11 in New York, where the recurring themes were expertise, user experience, social search, and taxonomy.  In the opening keynote, Eric Reiss declared that “enterprise search is about user experience” and that users should be taught how to search better. A subsequent keynote, presented by Lisa Welchman was all about web governance and emphasized following standards, and Thomas Vander Wal’s keynote was about social search. These may not be new ideas, but they are newer to search.

One attendee blogged after the conference that “recurrent topics that have emerged from the various presentations were: semantic search, social search, taxonomy, facets, and analytics.” Another attendee summed it up well in a Tweet: “Interesting to see that enterprise search is about text extraction, facets, expert knowledge (taxonomy), user feedback, UI.” In a recap session at the end of the first day, industry analyst Avi Rappoport went so far as to say that the technology today is pretty much the same as it was back in 2004, the first year of the summit, and the search challenges are the same, but the real savior is user experience improvements. These include matching a term in context, faceted metadata search/browse, taxonomy and synonyms, and “curated” answers (best bets).

As a taxonomist, I was pleasantly surprised at how many times I heard taxonomy, even thesauri, mentioned as an important enhancement to search (and I’m not counting my own pre-conference workshop on the subject).  In the past, taxonomy had often been seen as the antithesis of search (not needed if search works well). Fortunately, search experts realize now that this is not the case. In addition to the mention of taxonomy in several of the keynote sessions I attended, during one time slot two concurrent sessions were actually dedicated taxonomy case studies, making it difficult to choose which to attend: “A Practitioner’s Perspective on Taxonomy, Ontology, and Findability” and “The Science, Practice, and Art of Classification.” Even in the Tweets on the conference, I counted the word “taxonomy” mentioned many more times this year than with last year’s conference.

While the sessions were valuable and well-received, the exhibit was still smaller than it had been in past years. There were just 20 exhibitors this time and 19 in Spring of 2010, down from a high of 41 exhibitors in May 2007. I will interpret this as a lingering effect of the recession. It was also noticeable that some of the bigger players in enterprise search who had exhibited in the past, did not show up (Autonomy, IBM Microsoft and its FAST subsidiary, Oracle, SAP, and SAS).  This may also be due to vendor perception that the increasingly attendees of Enterprise Search Summit already have an installed search product, and now they are looking for ways to optimize it instead. Indeed it was common to overhear attendees asking each other which search software they use. That does not mean that the market for search software is saturated yet. Far from it. Rather, it’s a higher level user that attends Enterprise Search Summit. The fact that approximately 25% were from overseas (by a show of hands), indicates the keen interest in this conference by search experts worldwide.

What all this means for the industry is, while technology will still advance, there is also a growing need for search-related services in implementation design, user interface design, taxonomy design, etc. Enterprise Search Summit is not just about bringing together search software vendors and customers. Rather, implementation specialists and consultants, such as Project Performance Corporation, have an increasingly important role to play in the search area.

Over the last several months, PPC has been actively blogging on some of the Information Management topics that are core to our business and yours.  I am now pleased to begin complementing these Information Management topics with accompanying topics around the equally core Enterprise Services topics. The blog will focus on strategic management of IT resources with special focus on emerging technologies (e.g., Cloud Computing, Semantic Web), IT performance (e.g., Governance, Value Generation) and IT Trends (e.g., Social Networking, Consumer Technologies).

PPC has been providing leading edge enterprise services for over 20 years.  My role as Chief Enterprise Architect for PPC has provided the opportunity to work with many Federal and commercial clients spanning a myriad of organizational constructs and maturity. This is the perspective I hope to bring each blog entry:  to look across the IT landscape and discuss relevant trends for our industry.

The first blog entry will focus on an emerging technology, Developing Cloud Computing Governance. The enthusiasm around “moving to the cloud” is all around us. What are the implications for an organization, what are inherent risks, and how would an effective governance framework ensure that this move is successful.

I look forward to coming months and the exciting topics we will discuss. I especially look forward to getting your thoughts and feedback on the topics we cover here.

Interest in taxonomies is popping up in more and more places: different industries, organizations, agencies, and also in books, articles, and conferences.  One conference that now includes sessions on taxonomies, which previously did not, is the annual meeting of the American Society of Indexing (ASI).

I attended the ASI annual conference April 28-30 in Providence, RI, and was quite impressed with the taxonomy-focused sessions. These comprised a full-day pre-conference workshop on taxonomy and thesaurus creation, a poster session comparing taxonomies with thesauri, and two 1 ½ - hour sessions each with four speakers:  “Controlled Vocabularies & Indexing in Large Operations” and “Shaking the Tree: Case Studies of Taxonomies in Action,” both of which were well-attended and well-received. Due to space limitations, I will describe just the latter.

Although “Shaking the Tree” was aimed at an audience new to taxonomies, as it focused on examples, it was also very informative to those of use with more experience. It comprised four quite different but complementary presentations, including my own on retail website taxonomies.

A taxonomy for a retail website may seem simple, because products fall naturally into categories and it does not take a lot of subject matter expertise, since we all have some online retail buying experience. Nevertheless, creating a large retail website taxonomy can entail a number, as I explained based on my client work with PPC.  In other words, the complexity of the final outcome (i.e. simple hierarchical taxonomy vs. a complex thesaurus) does not always reflect to the complexity of the task to build it.

Jan Wright of Wright Information Indexing Services presented two very different case studies. In one case, she had to develop a much more highly specific cataloging system based on Library of Congress classification in order to catalog with resources of the Navajo Nation Library. It was refreshing to see a connection with the cataloging world, which we “taxonomists” tend to ignore and leave to “catalogers,” even though the skills required are essentially the same, as this presentation illustrated. Jan’s second case study recounted building a corporate product taxonomy for computer hardware from scratch.

Although we usually think of taxonomies as just for search and retrieval, Ceilyn Boyd presented a case study on a project to develop a taxonomy used to manage metadata for a survey of audiovisual resources at Harvard University library archives. The Harvard SAVE (Support for Audio Visual Evaluation) Taxonomy was designed to support "a multi-user, item-level, Web-based physical condition assessment survey of all audiovisual materials at Harvard University." A taxonomy for a “survey” may sound strange at first, but makes perfect sense when we learned that the taxonomy was supporting metadata for asset management.  Although Ceilyn did not use the word “metadata,” it became clear that sometimes different terminology is used in different situations, even though task and goal may be similar.

The session concluded with a presentation by Christine Connors entitled “Taxonomies – A foundation for more,” which took a look at ontologies. Not an actual case study, the presentation was a good introduction to ontologies with both its visualizations and web site application examples. These included DBPedia.org and RADiFy, a bookmark for of annotating a web page with RDFa. One of the points that Christine made was that taxonomies or ontologies can help make the distinction between “aboutness” and mere “occurrence” of a topic in document or content asset to be tagged. This is a principle I am familiar with from my background in indexing, but am reminded now that it is a point to make when communicating about taxonomies as well.

The ASI conference is small enough (about 160 registrants) that it is easy to meet people, and the attendees are in fact very open introducing themselves, as many are independent contractors. At the same time, the conference is large enough to support three or four simultaneous track sessions to appeal to different interests, and single day registration options are available. So if you are interested in taxonomies and content management, rather than back of the book indexing, you may still find it worthwhile to attend part of the conference. More information is at www.asindexing.org.

In recognition of Earth Day, I’d like to pause on the topic of Green IT.  Though I didn’t think of this entry in a rain-water shower, type it on a solar-powered keyboard, or proof-read it on 100% recycled paper, we are a company that “lives green” with a sustainability plan and carbon foot-print monitoring.  With our deep expertise in energy and environmental issues combined with our comprehensive IT services, we’ve been doing Green IT long before that term ever existed.  My sense today is that many people confuse Green IT with a limited view of server management or consolidation.   We take a broader view that considers environmental impact, process improvement, and IT value as three interrelated factors.  We’ve coined the term Sustainable IT to represent the combination of these three components and the business value that can result from a cohesive strategy that integrates all three.  In considering the idea of Sustainable IT and the specific services within that concept, I often focus in on Knowledge Management.  KM has been around for a long time, often with mixed results as I’ve critiqued in the past.  Our work around practical KM, that which focuses on business value for the end user, centered on the concepts, practices, and technologies used within and between organizations to capture, manage, and distribute knowledge is inherently green.

In other words, good KM, practical KM, is Green KM.  If an organization focuses on capturing, managing, and sharing information more effectively, work will become more efficient.  There will be less unnecessary effort, less rework, and therefore less waste.  If people are able to connect and collaborate more effectively, there will be less need for travel.  We’ve already seen significant benefits from allowing employees to telecommute, something that wouldn’t be possible without our Information Sharing Portal and other collaborative/KM technologies.  Equally, our HD video conferencing system has drastically reduced our need for international flights while creating new avenues to capture and document tacit knowledge.  The common theme to all of these initiatives is that they address business needs directly.  They not only benefit the environment, they benefit our knowledge retention, efficiency, and customer satisfaction.  That, in essence, is our concept of Sustainable IT -- Environment, Process, and Information Technology -- considered together to make the right decisions for your organization, your people, and your planet.  

This week, I was in Scottsdale, Arizona to attend and present at the SharePoint Strategy Summit.  Even though I have been designing and deploying information strategies and business objectives within the SharePoint platform for the last five years, there is always so much more to learn and know when it comes to a product like SharePoint.  

Ultimately, the intent for this conference was a bit different from others I’ve presented at.  Whereas other conferences tend to discuss technical SharePoint, how to develop workflows in Designer or leverage InfoPath—this conference was a true strategy summit.  It intended to arm conference attendees with the necessary information to make educated decisions and plans as they develop their own strategy to deploy SharePoint. 

SharePoint has matured quite a bit since its inception, and my colleagues and I have seen and experienced many successes and failures with the product in the span of its evolution.  The themes that emerged from the conference really are a reflection of SharePoint as it is today, about a year and half since the release of SharePoint 2010, and on the cusp of a potential new release of SharePoint in the next few years.  These themes are more like lessons learned and they clearly communicate that this is commentary from a group that has been working with SharePoint for quite some time:

• Breadth of SharePoint:  No one knows everything about SharePoint.  It’s such a diverse and complex platform, and there are always new things to learn. It can handle any number of business objectives being thrown at it. 
• The SharePoint Eco-system:   The popularity and market share of SharePoint is undeniable.  Whether it is the right solution for you or not, it’s unquestionable that there is a large support network of consultants, development add-on products, functionality enhancement add-on products, expert blogs, wikis, and open source code that can support you in doing whatever it is that SharePoint can help you accomplish. 
• Governance is the Key to Success and Future Proofing SharePoint:  Every consultant had the same message: governance is the key to success.  Not just any governance, though. Realistic governance.  An attendee came up to me with a startling revelation about governance. He originally thought of governance as being egalitarian, driven by rules and policies. However, it’s more about transparency.  The rules don’t have to be equal, and if you are transparent about why the rules are unequal, this actually promotes appropriate management of the content and sites within SharePoint.
• Education vs. Training:  While training is very important, it is very different from educating your users.  The difference? With education, your users don’t even have to know that the platform for their new intranet is on SharePoint, but they have to know what’s in it for them.  Give your users a simple compare and contrast, demonstrate how business is done today, and then show them how that same business can be improved with technology. 
• Measuring ROI:  There were a number of different arguments for measuring return on investment (ROI) with SharePoint.  Some don’t believe you can measure ROI, while others believe you unequivocally can.  With often poorly-defined objectives that are too high level, it’s hard to tangibly equate improvement with dollars or measure how SharePoint has drastically improved time savings.   I do believe ROI can be measured against an installation of SharePoint, but it requires that objectives be well-defined and quantifiable.  Then create a baseline by measuring time or dollars prior to deploying SharePoint.  You need to know what you’re trying to improve before you can acknowledge that improvement has been made. 
• Taxonomy and Metadata are Extremely Important:   This one was a personal win for me.  Despite its reputation in the highly dynamic and exciting SharePoint and IT world as being ‘boring,’ true technical experts in the SharePoint domain are increasingly recognizing that a key differentiator to a successful implementation of SharePoint is a well-developed taxonomy and metadata schema to manage information within a site collection, site, library, or list. 
• How Do You Measure Success or Failure of SharePoint?  This is a very subjective assessment, but most examples of failed SharePoint installations were a result of poor project management, poorly defined requirements, or little to no governance.  Success can perhaps be measured or quantified by metrics to determine if any ROI has been attained.

What was truly remarkable regarding these workshops was the universality of topics.  I’ve long championed the power of subjects or topics as the key building blocks for core taxonomies.  They help to avoid the use of confusing jargon or complex organizational structures.  Subjects are also relatively easy to translate in order to enable multi-lingual taxonomies, which is of course especially important when dealing with a global organization.  As we ran these workshops, the various business representatives consistently noted how they were different from their colleagues around the globe.  They expounded upon their unique business and were intent on the differences in how their unit or office differed from their colleagues.  However, as soon as we began identifying the topics in which they wished to categorize their content, we discovered it was closely aligned with their other offices and colleagues.  In short, they discovered their business and the information it contains possessed a lot more similarities with their colleagues than differences.  As a result, we’re well on our way to designing a single global intranet powered by one taxonomy that will support them all.  For them, the ramifications of these workshops go well beyond the intranet.  Many of them noted their pleasant surprise in how closely they were aligned with their colleagues and believe it will help them be more effective in their business.  In other words, as I’ve long stated, good taxonomy design can make an organization more efficient and effective.

With our current design efforts nearing completion, our next task is to develop a taxonomy governance model to ensure the long-term viability of this taxonomy.  We will use our taxonomy governance process to empower these same users to make ongoing design decisions that will benefit the enterprise and keep this taxonomy trending in the right direction.  More on that to come.

There is a common set of facets or metadata fields that occur again and again when we design taxonomic metadata to describe content items. These universal facets include:

• Type of content,
• Geographic location,
• Organizational entity, and
• Product or service.

The function or purpose of the content item is another universal facet, but it is one of the least understood and most difficult to build and use. In an ideal world, the function or purpose of each content item would be specified when it is created, e.g., “financial reporting,” “human resources management,” “marketing,” “sales,” etc.  In reality, we need to define and agree on a broad set of business activities to be able to answer the question: “The purpose of this content is _____.”

Controlled vocabularies should be based as much as possible on existing sources which should then be customized for use in the particular application that is being developed. The benefit of using existing sources is that the terminology has already been shown to be useful, and it will already be familiar to employees and other staff, saving development time and training effort. Common sources for business activities terminology are industry standards and records retention schedules. Records retention schedules are typically based on a combination of business function (purpose) and document type, for example:

An example of an industry standard is the Federal Enterprise Architecture Consolidated Reference Model (FEA CRM) which provides measurement categories for citizen services such as such as community and social services, defense and national security, disaster management, economic development, etc. and other customer results, processes and activities, and technology measurement areas. Another useful industry standard is XBRL (Extensible Business Reporting Language) taxonomies.

Here’s an example of a small but typical list of business activities that could be used as a starter set for developing a controlled list of document purposes:

• Accounting
• Community Outreach
• Company Safety
• Customer Service
• Emergency Services
• Environmental Programs
• Facilities Management
• Financials & Investor Relations
• Human Resources
• Information Technology
• Legal & Ethics
• Maintenance
• Media & Communications
• Sales
• Supply Chain Management
• Training
• Travel

In document management applications, the purpose of the document is sometimes pre-determined by the template that is used to create it. In some information management applications, the purpose of the content item can be pre-determined based on the user is who is logged on creating it. In these cases, business rules need to be implemented that automate the tagging of a new document or other content item with the default business purpose.  But in most cases, the purpose of a document can only be determined by a person who is explicitly identifying it. It is a challenge not only to build a concise and comprehensive set of business purpose categories, but to build it in a way that is intuitive for people to use consistently. In our experience, determining the business purpose of a document is not a completely natural activity for people to do—it should be intuitive, but unfortunately it isn’t.

Here’s a thought experiment. What was the last document you produced? Probably it was an email message. My last message was a message related to interviewing a prospective new employee. The purpose of that document was “Human Resources.” What was the message you produced before that? It was a response to a manager in our contracts department. The purpose of that document was “Accounting.” Even though both documents were email messages, they had completely different business purposes. Getting staff to identify the business purpose of their documents is difficult, but it’s an important part of identifying what documents need to be retained and how to schedule or categorize them. It also provides a basis for measuring business activities. It should be possible to decide what the purpose of every document you create is, and by doing this you will improve the measurability of what you are doing and the findability of your work product.

In the name of user-centered taxonomy design, I found myself in Bratislava, Slovakia this week to design a taxonomy for a global company. What’s in Bratislava you might wonder? How does a taxonomy designed from Slovakia fit in with a global taxonomy schema? The city boasts the Bratislava Castle, a former seat for the Austrian-Hungarian Empire, and the hot chocolate is phenomenal. It also happens to be one of the European hubs for our customer. The taxonomy workshop I facilitated for them will be one of six that will be sponsored around the world in the next few weeks.

When you’re a global company, having a unified intranet is a constant challenge. Information management is a struggle between different languages, terminologies, and cultural perspectives. User-centered design assists in bridging gaps between these differences by trying to understand the common vocabulary used between geographic hubs and how users approach information sharing. This week I worked with business representatives from Germany, Bulgaria, Italy, and of course, Slovakia. Together we worked to design a business taxonomy that provides capability to categorize and search for information that they find valuable.

The interesting part about my trip to Bratislava and the other upcoming workshops will be how the results come together. Similarities between the taxonomies and metadata will demonstrate common subjects that are important to all audiences. Differences may help provide a broader and fuller range of subjects that only a larger and diverse group of users can provide. The differences may also be linguistic, the subjects may be the same, but are labeled under different names. The workshop exercises will help determine the overall design strategy for the global intranet. That strategy may emphasize the importance of audience targeted information, localized content, personalization, and the need for faceted navigation.

Collaboration and social computing are popular topics of conversation. The workshop participants are interested in social computing, reinforcing the need within global organizations for informal communication channels to share ideas and facilitate global dialogue. It has been a fun assignment, being able to wear both hats of the “tourist” and the “taxonomist.” The workshop participants were happy to offer their local expertise on favorite Slovak cuisine and the best restaurants in town.

Last week, I heard a myriad of experts discuss ”knowledge to action” at KMWorld 2010. Everyone had his/her own spin on it. This year’s KMWorld was structured with a variety of different types of sessions. The number of panels and case study sessions was equal to the number of typical lecture-style expert-led sessions. The panels kept the pace interesting; experts were there to question one another and to provide attendees with a variety of opinions. The KM version of the popular ESPN TV show “Pardon the Interruption” became a thoughtful debate.  Case study sessions allowed up to four different organizations to share their success stories and lessons learned.  We heard from many that the greatest take away from the conference was hearing the real examples of ”knowledge to action.”

The high energy and enthusiasm at KMWorld was undeniable. I’m proud that PPC could be a part of that dialogue and lead some of the discussions. Between KMWorld and the SharePoint Symposium and Taxonomy Bootcamp that ran alongside the main conference, PPC spoke at a total of nine sessions, including a few of the popular panel and case study sessions. 

For me, the interesting take away from KMWorld is the discussion on the what ‘knowledge to action’ can mean. The underlying principles and logic behind information management are shared at KM World every year, but what distinguished this year from others are the creativity and the social journey that was discussed in many of the sessions, along with examples of real life implementation. There were a few themes indicating growth in the industry as well as areas where knowledge management continues to struggle:

• Social computing has proven to be more than just another popular trend. Companies are finding real value in the knowledge sharing capabilities of social computing.
• Knowledge and information governance are proving that the right balance of governance is still difficult to put in place, and even harder to master.
  
• SharePoint is incredibly popular, and this is not news. However, there is still much discussion on its strengths and weaknesses as a web content management solution.
• User-centered design for taxonomy and metadata has emerged as a strong theme from many industry-leading taxonomists. Taxonomy and metadata are true drivers in developing semantic web within the enterprise.

This past week I’ve been attending Taxonomy Bootcamp as a precursor to KMWorld.  In addition to attending any number of fascinating sessions, I was proud that PPC presented during three separate sessions between myself, Joseph Busch, and Tatiana Baquero. 

The Bootcamp overall was the best I’ve attended.  Kudos to the organizers and the attendees.  Not only was there active participation and strong attendance, but the overall understanding of where taxonomy fits into the business seems to have increased greatly.  I consider this as a great sign of where the industry is going and the place that knowledge management is playing in today’s dynamic businesses.  A few of my key take-aways from the last couple days are:

1. Taxonomy management and auto-categorization technologies are increasingly at the forefront of information management projects.  Denise Bedford of Kent State University made a key point that organizations who aren’t, at least, planning for auto-categorization as part of their information management strategy will quickly be overwhelmed with the mass of content being generated in today’s businesses.
2. The role of the end user is increasingly being recognized as critical to the taxonomy design and management process.  This is what I refer to as the Democratization of Information Management, the idea that laypeople end users are now central to the capture, management, presentation, location, and acquisition of information.  Many of the world’s experts on taxonomy design reiterated this point as a central theme in their discussions.  In short, taxonomy is for those who need it, and increasingly that means the business users themselves instead of the librarians or information scientists.
3. Taxonomy was identified as the key technology underpinning the semantic web through linked data. In this context, taxonomies identify the names of things (people, organizations, places, topics, etc.), unique and persistent identifiers for them, as well as the relationships among them. Sunny Yoon, Goldman Sachs’ taxonomist, said that there is tremendous business value in agreeing on what something means. Michael Uschold, an independent consultant, described the semantic web as “wearing meaning on your sleeve.” Karin Michel, SNC-Lavalin’s taxonomist, said that taxonomy is not artificial intelligence, but it is the basis for searching more intelligently.

As a lead up to KMWorld, PPC is hosting a Thought Leadership website counting down thirty days of Knowledge Management content.  Every day, we make available a new piece of content on a knowledge management (KM) topic.  Most of this comes from conference presentations and white papers we’ve presented over the last couple of years.  It was a fun exercise to pour through these materials and recognize the themes in them.  We’ve made some shifts in methodologies and approaches over the years, but overall, I’m quite proud of the consistency of our practices and the case studies that support them.  I invite you to visit the site each day and seek your own themes from the content.  In advance of that, I offer the following as the overarching themes to successful Knowledge Management and our approach to it.

• Focus on the End User – KM, more than anything else, is about people.  It is the focus on capturing, managing, and presenting information from the people, for the people, and by the people.  Any system, any approach, and any decision must be made with the consideration for the end user.  That is why user-centered design and testing, iterative approaches that engage the end user in multiple feedback cycles, and effective governance that gives the user a voice are critical to successful KM.
• Business Value – For years, there have been those in the Knowledge Management field that talk a lot and do very little (you know who you are).  Lovely papers and floral concepts get discussed at length, but few decisions get made and little business value gets generated.  As a result, KM as a field has often struggled to find respect and validity.  The KM we’ve done has, on the other hand, always focused on concrete results and measurable business returns.  Saying people will share information or reuse information more effectively is not enough.  Serious thought should go into the business value of any KM initiative (and indeed, any project at all). 
  
• Iterative Methodology – Many Knowledge Management initiatives, be they tied to a particular technology or not, tend to be somewhat esoteric to users and decision-makers in their early stages.  Sure, everyone says they want to collaborate or share information, but we know from experience that doesn’t always translate to a successful project or a true mandate.  One key to successful KM projects is to iterate both design and development in order to ensure you’re giving your users something solid to respond to.  Equally, an iterative design allows you to ensure you’re hearing the users’ needs correctly (mapping back to the above point about end user focus).
• Practical Knowledge Management – The final theme for the day focuses on practicality.  Everyone might dream of the Enterprise Information Management system powered by an integrated stack of technologies and supported by cohesive senior mandate.  We can strive for that, and in many cases achieve it, but each organization must also be honest about what they’re ready for and what they can achieve.  Sure, building “Facebook for my company” or enabling geospatial tracking for your sales force is sexy, but all you may need is a simple threaded discussion.  Don’t get sucked into the trend game.  Design what your end users need; your company will benefit from this and you will be on the road to KM success.

How much training is needed to shop for shoes on Zappos, or a home on Zillow, or to shop in a bricks and mortar supermarket or department store? Most of us are wired for shopping, or, more accurately, have learned how to navigate stores by reading signs and responding to the explicit ways that products are typically merchandised – chips are on the end caps, and also in the snack food section somewhere in the middle of the supermarket; fresh foods are around the edges. So we should expect informational websites and intranets to be just as easy to navigate and use, or just as easy to learn to navigate and use – without training. Looking for products or information is “natural,” it’s something we do every day; and we expect to find them in “natural” ways, or least in ways that are easy to learn.

There has been a movement against arranging intranets by organizational structure even though this is a standard, predictable scheme that is easy to learn over time (although it is true that new employees won’t know it at first). We think organizational structure is useful as long as it is not the only way that the intranet is organized. Organizing content by products and services has some of the same problems as organizational structure, but organizing content by “what you make” and “what you do” is important, especially on public websites. One solution is to group products and services by broad categories such as line of business, market (consumer, business, government), industry (financial services, manufacturing, oil and gas, retail), and/or location (U.S., EMEA, Asia Pac).

The goal is the same as shopping for shoes – using a few broad divisions or facets to quickly narrow the browse tree down to the group of items that you are interested in finding. Whatever method is used, it must be easy to learn and natural use, as easy as shopping for shoes on Zappos or a home on Zillow.

I built an entire high school career around mix tapes.  Of all the Breakfast Club stereotypes, I was neither the jock nor the misfit, and I’d like to think that I at least avoided the geek status.  I was, however, a big fan of music.  The amount of time I put into this pursuit was immeasurable.  Reading college papers and fanzines, “bootlegging” concert tapes, showing up early to concerts in order to discover the opening act for the opening act, and picking the brain of anyone with a modicum of musical interest.  I did because, aside from my continued love of new music, it was cool.  It was fun to be known for a good mix tape.  It was, for all of you knowledge managers and requirements gatherers, my persona.  If I were in high school today, however, my persona would be kaput.  The reason is simple.  Anyone with an internet connection can be a musical genius.  The Music Genome Project, Pandora, and iTunes’ Genius use a variety of “content analysis” and user analytics to recommend songs and musicians based on what one has purchased or listened to in the past.

Amazon and a myriad of other retailers, of course, do the same thing for virtually every project under the sun.  Often times, with humorously mixed results.  I, for one, am currently receiving product promotions for My Little Pony and Disney Princess, serving only to remind me I’m a fantastic Uncle to my nieces rather than represent my own personal product interests.

We are now at a place where we can do the same thing for an organization’s own information as is currently being done for music and retail products on the web.  A well-designed and managed tagging scheme powered by business taxonomy, coupled with any number of search technologies, user analytics, and basic business intelligence can yield a recommendations engine for a corporate intranet or knowledge base.  The result would be that users who found value in a particular document would receive recommendations to view other documents on a similar topic, from the same author, or perhaps including the same phrases or concepts.  Though this is eminently feasible today, it isn’t yet a mainstream feature of any information management tool.  I predict that will change quickly.  In the mean time, we’ve started down this path with several of our clients.  Stay tuned for updates, and please be in touch if you have a case study you’d like to share regarding how your own organization has begun the path toward dynamic content recommendations.

Taxonomy as a topic (and as a buzzword) has experience several cycles of popularity.  It comes into vogue and passes by, but managing information so it can be used and re-used is a real requirement for every organization and every individual therein. To begin our conversation on taxonomy, these are some of the frequently asked questions we often get within the realm of taxonomy (and its partner metadata).    

For years PPC has been working with organizations to design, build, and manage taxonomies.  We developed the Business Taxonomy Design methodology, which is focused on business user workshop involvement to ensure business-centric and user-centric taxonomy designs.  To date, we’ve helped over 200 organizations around the world with their taxonomy efforts.  Recently, I was observing a colleague deliver one of our workshops for a client and it struck me that one of the reasons so many organizations struggle with taxonomy design is that they treat it as a purely academic pursuit.  In short, they try to treat as math and science something which is much less about 0s and 1s than they care to believe.  Sadly, I think a number of taxonomy practitioners have fooled themselves into thinking the same.

There is no question that taxonomy design does have a strong science component.  From Zipf curves to card sorting analytics, there are clear “scientific” components to successful taxonomy design.   In the early stages of a taxonomy design (or redesign) effort, it is critical to understand a number of measures, including the distribution of tags to content, the percentage of content tagged, the most searched upon and browsed upon terms and tags, and the frequency of content being accessed.  It is also important to quantify user satisfaction with their existing capabilities and processes regarding taxonomy, tagging, and related areas.  When woven together, these assorted metrics provide a strong baseline and starting point for taxonomy design.  They also help to ensure a project focuses on addressing particular challenges and provides a means to measure project success.  As a project proceeds, these same measurements are important checks against the value of the taxonomy you’re designing.  The science, however, is only the first part of the story.

Taxonomy design is also in large part an artistic pursuit.  What I mean by this is that a taxonomy needs to look and feel right in ways that are difficult to quantify.  The words and concepts that comprise a taxonomy must evoke the proper responses in a wide array of potential users.  I’ve co-opted the word “Gnosis” to represent this during our taxonomy workshops.  What it means, in this context, is that we want the users of our taxonomies to “just get it.”  The pursuit of Gnosis within taxonomies is largely an art as it requires a great deal of flexibility and massaging of the terms you consider in order to arrive at those that will make the most sense to the most people.  It also requires a level of compromise and acceptance of the fact that we may sacrifice one person’s definition of the perfect word to arrive at a term or terms that is at least acceptable to the larger group.  For instance, one exercise we employ in our taxonomy design workshops is to ask all participants to individually write the “child” words that belong within a “parent” category.  We record all of these words together on a white-board or screen and then lead a discussion to find the common themes and ideas that best represent what we want to represent, but will also serve the lowest common denominators of our potential end users.

Finally, the dirty secret of taxonomy design is the consideration of politics.  Quite simply, the passions of senior decision-makers and the needs to present an organization in a favourable light may trump both science and art.  Politics also comes into play through the management of multiple different interests and personalities.  When we run our workshops, we’re often facilitating a balance between different business lines, different geographic centers, or different personality types.  The most effective taxonomies are those that incorporate both the majority and minority opinions and provide a harmony between the myriad “things” that any taxonomy will cover.  Finally, the greatest taxonomy design success will come when Art, Science, and Politics are all considered and brought into harmony.

• Information Governance,
• Social Computing for Business, and
• Business Taxonomy Design.