Gilbane Report logo

March 2005

It would be difficult to find anyone who spends time on the Internet, or indeed who reads newspapers, who has not heard of blogs. Wikis are less well known, though Wikipedia, the free online collaborative encyclopedia is helping to change that. The vast majority of blogs are individual personal journals, many of which have some technical content, but most of which are made up of individual opinions about politics or hobbies. Most of the discussion about blogs is centered around their affect on mainstream journalism, their power as a new communication channel and voice of the people, and how this will impact society. All this is interesting, but what does it have to do with implementing content or knowledge management, or enterprise collaboration applications? IT, business managers, and even analysts can be forgiven for thinking “not much”. In fact, we have been skeptical ourselves.

But, being dismissive of blogs and wikis because of how they are most often used, and talked about, today is a mistake (PCs and web browsers weren’t considered as serious enterprise tools at first either). What is important is how they could be used. They are simply tools, and many of you will be surprised to find how much they are already being utilized in business environments. For this issue, Contributor Lauren Wood provides a straightforward explanation of what they are, describes how they compare with content management systems, and reports on some telling examples of how blogs and wikis are currently being successfully used in enterprises.

Frank Gilbane


Blogs & Wikis: Technologies for Enterprise Applications?

Executive Summary

Blogs and wikis are flexible practices and technologies that are increasingly being used within companies and organizations to ease the creation and dissemination of information, as well as making it easier for companies to communicate effectively with customers, partners, and the public. This article discusses some of the salient features of blogs and wikis, and give examples of companies who already have implemented one or more of these systems.

On speaking with several vendors and users, it is obvious that many companies are implementing blogs and/or wikis or hybrid systems successfully and I expect this to continue as the technologies and practices mature.

What are Blogs and Wikis?

This article will discuss some corporate uses of software systems that fall under the generic title of blogs or wikis. Although they are often talked about together, blogs and wikis are quite different, with different users and different workflows. Blogs, or weblogs, are the latest in hyped technologies. Like any new technology (although blogs have been around for more than a few years), they have the potential to change the technology landscape in ways that are not yet clear. Blogs are best thought of as a way to present information to the world or to a select group. The traditional blog is written in the form of an online diary and includes the writer’s thoughts on a subject, links to interesting information, and often pictures. The writer may post a new item several times a day, or a few times a year. There are blogs on every conceivable subject and in most human languages. Traditionally blogs have been created by one author and represent one author’s views, although there are some group blogs. Wikis are a different method of publishing and presenting online information. The most famous wiki is the Wikipedia, an online collaborative encyclopedia that illustrates the common wiki features, namely collaborative authoring with lightweight content management features such as lists of changed pages, author tracking, and locking. Some wiki systems also have version control and rollback. All wikis make it easy to add new pages and create links.

A technology that is related to blogs and an expected part of any blog system is syndication (following the standard convention, I’ll call it RSS in this article, even though RSS is just the name of one common protocol). RSS is an information gathering system where software either on your desktop, or via a web site, polls websites looking for updates on published information. The updates are marked as new in your RSS reader so you don’t have to waste time looking at articles you’ve already read. Some of the biggest generators of RSS are online newspapers, which typically have several RSS “feeds” or lists of pages, corresponding to the different categories of information they publish. And of course blog systems automatically generate RSS feeds as well.

Using Blogging Software and Systems

Blogging systems are extremely flexible, as befits a relatively new technology. They are almost always architected as a core with plugins that add extra functionality to the core system. They typically use templates for the HTML and CSS for styling, so that changing the appearance of the pages is a simple matter of changing the style sheet. For this reason blogging software and systems are now being used for many web publishing applications, and not only blogs in the traditional sense. The striking thing about blogging systems is how easy the blogs are to update, removing most or all of the pain of writing HTML pages. Even relative newcomers to technology find adding information to a blog (once the system has been installed and configured) easier than creating a word processing document. And it is far easier than adding information to many content management systems, though also far less powerful.

In a similar way, wiki systems are flexible, come with a number of built-in features that depend on the precise system, and can be extended as needed. There are a large number of blog and wiki systems available. Some are open source, some are commercial, and some are commercial but free for certain classes of users. There are hosted services and value-added services. Many of the blogging systems are designed for the personal blogging market, but there are also some that are intended for corporate use, with a different feature set and different expectations of the intended user.

Using blogs and wikis in the corporate world

There are undoubtedly more uses for blogs and wikis in the corporate world than I have room for in this article. I have chosen a range of illustrative uses after talking to both vendors and users that illustrate both some of the advantages of using these new technologies, and issues that need to be considered when contemplating installing such a system. There are a number of different systems available, both open source and commercial, and the fact I mention one system or another doesn’t mean I recommend it, just that it has some interesting features that illustrate a point I wish to make.

It is important to bear in mind that using a blogging system doesn’t mean the pages produced need to look like a blog. The traditional blog has dated entries arranged in a reverse chronological order, uses categories sparingly, and is written by one person to reflect a personal view. These are conventions that are supported but not mandated by blogging systems in general, although of course less flexible blogging systems may not support other uses in a graceful way.

Outward-facing Blogs and Wikis

The most obvious use of blogs in the corporate world is where an employee of a company writes their thoughts on issues, just like with the traditional personal blog, but with the added twist that they also write about their work and the company they work for. The Economist wrote about Robert Scoble of Microsoft in their February 10, 2005 issue, heralding the death of traditional PR and calling him a celebrity blogger. There are other companies where the CEO or President writes blogs to explain their thinking to the world, and I expect the trend to continue. Companies have realized that they need to explain what they do to the world, and they also need a forum to find out what their customers, partners, and investors think. Blogs provide this forum; at the recent Northern Voice blogging conference the keynote speakers Tim Bray and Robert Scoble both talked about the increased listening power that blogs bring. The flip side to this, of course, is that people who write in response to a blog posting expect to be listened to, and they are very quick to pick up on inauthentic, exaggerated, or dismissive responses. To ensure that blogs are real, bloggers who work for a company should write about what interests them and what they know. For example, customers trying to use a software system will appreciate postings about tips and tricks written by engineers who developed the system. People selling a house will appreciate postings written by a realtor about the development of house prices in the area, or movements in interest rates, or other things related to the housing market. The readers also get some sense of the human behind the writings; a blog will typically cover a range of subjects with the postings about software development being interspersed with pictures of the writer’s cats or musings on knitting patterns. It’s important also that people who are in senior positions at companies only write blogs if they actually can write in an interesting way; blog postings that reek of “marketing speak” will be largely ignored or, worse, ridiculed.

Wikis are also useful for providing information and gathering feedback. The “tips and tricks” noted above could also be kept in a wiki, where they would be edited not by one engineer but by a group of people, which could include the public. Unlike blogs, wikis are designed for repeated editing of a set of documents, making them more suitable if the document is expected to have a longer life and should be easy to find throughout its life cycle. As an example, Microsoft uses wikis to gather customer input and ideas.

Easy Web Publishing

Blogging systems are so easy to use that many companies use them as a web publishing system. It’s easy to take a blogging template, rework it to remove the “blog” features such as date and author listings, and set up a reasonable stylesheet. The result looks like a regular web site, but it’s much easier to use than traditional HTML editors. An example is the Seattle Children’s Hospital Events Calendar at http://www.seattlechildrens.org/home/calendar.asp. The blogging system allows the easy addition of new items and automatically creates the RSS feed, and the result doesn’t look like a traditional blog.

Internal Uses for Blogs and Wikis

Many companies use blogs and wikis internally for a variety of uses. This is where the different feature sets of different systems start to become more important. The outward-facing uses of blogs typically need few features whereas using blogs or wikis internally (whether on an intranet or extranet) often requires a different feature set. We’ll see why as I go through some examples of use.

  • Probably the easiest example of using a blog within a company is as a company notice board. Whether the published item is about the office party, an interesting link, or the latest sales success, a blog can help keep people informed of the small items that make a company’s culture more vital. Telecommuters can keep up to date as well as part-time workers or frequent travelers. The blog helps cut down on email traffic and nobody is inadvertently left off the mailing list.
  • Navarik in Vancouver, Canada, writes software for marine shipping companies. They use a blog internally to record when an update to some software has been uploaded to the server for the customer to download. The customer can then put in comments on the latest version, such as if there are any problems found. Thus the blog serves as a log file; since the latest updates are typically the ones people are most concerned with, the reverse chronological filing order and the fact that older items fall below the fold are advantages.
  • CommSecure in Australia makes e-business solutions that are installed over much of the world, with 24×7 support. They use a wiki internally to track the current status of each installation, as well as to document procedures for handling alerts, solutions to new problems, changes in contact information, etc. The wiki is easily updatable and everyone is encouraged to contribute if they have new information. If the answers are in more formal documentation, the wiki serves as the index to that documentation, which saves people in an emergency having to wade through several different sets of documentation provided by third party organizations trying to find the one vital piece of information to solve this particular problem
  • Seattle Children’s Hospital publishes intranet pages in the same way as the external pages mentioned earlier. The same blogging system (a highly modified version of MovableType) is used as the basis of the departmental web sites. Pages are created that serve as gateways to necessary manuals, making it easy for clinical staff to find the most recent version of whichever document they need. And because it’s easy for the department staff to add information about new forms or procedures, the quantity of email has markedly decreased, and the probability of a clinical staff member missing something has also decreased. The system also has plugins that enable pulling information from one blog and publishing it on another (news items, for example), which further decreases the time and effort required to disseminate information across the departments.
  • Software vendors in this space have some interesting success stories to relate; two that approach the information dissemination differently are SocialText, which has more of a wiki feel (case studies available at http://www.socialtext.com/customers/) and Traction Software, which is more like a blog system (case studies available at http://www.tractionsoftware.com/solutions.htm). Both systems have aspects of both blogs and wikis, with the proportions varying according to the perceived customer requirements. I’ll discuss these further in the section on Implementing a System.

As requirements from companies become more complex, for example as the size of the company grows, or in highly competitive or heavily regulated industries, the requirements placed on the systems also grow. The trend in commercial products is towards combined systems that have features from both blogging systems and wikis as well as full audit trails and version control. What is noteworthy about these systems is that they are using the functionality developed for personal online diaries and turning them into systems for information sharing where the individual voice and personality is less important than the information that is being imparted. This is where blogs shift focus from the sometimes hubristic to the collaborative, from the individual to the group. And thus many other types of systems that work to support collaborative efforts are looking to add blogging or wiki-like capabilities, forming hybrid systems. I foresee this trend continuing, and that just as content management systems now are expected to provide ways to take advantage of XML documents, so will enterprise systems be expected to provide blog-like capabilities and/or RSS feeds.

There are good reasons for demanding these capabilities from enterprise systems. Currently much information flows via email. This leads to a number of problems, including the difficulty of “occupational spam” where people are copied on email messages they don’t need to read, and the missing email where people are left off the list that do need to read it. New people don’t have the necessary background, and people who’ve been around for a while have hundreds of unnecessary emails languishing unread in folders. Over-eager spam filters (including a human who is overwhelmed and clicks the delete button too quickly) are a big problem, as is the problem of simply missing a message in the hundreds that people receive each day. Having centralized information sources with RSS feeds solves this problem. The reader subscribes to what they want to subscribe to (or, as befits the enterprise context, what they are allowed or required to subscribe to) and is automatically notified of new content. RSS readers can download all the updates to the local machine for offline reading, just as for email. And spam is taken care of at the content provider end, not the reader end. A further advantage is saving time. It is much easier and quicker to read an RSS feed from 50 projects than to go to the websites of even 20 projects to see if anything new has happened. Since RSS readers flag the new items, the reader doesn’t even need to wonder whether they’ve read this item before or not, the technology takes care of that for them.

This is not to say that blogs and wikis will replace other systems. Proper project management requires to-do lists and milestones, for example, but there is no reason that these can’t be developed within a blog context that allows for RSS feeds and email notification, and allows weekly reports to be summarized to the blog. I use Basecamp to organize the logistics involved with the XML Conference I chair, for example. It has the to-do lists, milestones, categorization of messages, and permission-based editing that is required for a project with a small number of participants, along with ease of use, RSS feeds, and email-based notification that doubles as notification of meeting times and agendas.

Wikis will never replace fully featured content management systems (and any that claim to will be as complicated to use as a full CMS), though I wouldn’t be surprised to see steps being taken down that path that result in wikis being able to do a large proportion of the useful functionality of CMS Some of the differences I see are

Wiki CMS
Creates and edits HTML pagesStores and edits any type of document, including word-processing, XML, multi-media
No granularityFull granularity for XML documents
Documents can be “lost” if not linked toDocuments are never “lost”
Quality of audit trail depends on the wiki softwareAll good CMS have complete audit trails
Often minimal or non-existent locking and version controlFull version control and locking

Some of the items I wrote about in my article on The Role of XML in Content Management would not be possible using today’s wiki packages due to the complexity of the required solution. I expect, however, that some of the ease of use aspects of wikis will be available in future CMS, while wikis in the future should be able to edit XML.

Implementing a System

As with implementing any system, you have to think about what you want to achieve. I hear of many companies who have good results with wikis, whether the usage is sporadic (for organizing ideas and information about a once-a-year company meeting) or continual (the CommSecure example given above). There are companies where wikis are heavily used in the requirements gathering phase of a new product or project but seldom thereafter. Since wikis are easy to set up and use, the rewards in general justify the effort, even if only one person uses the wiki to organize their research notes and thoughts. Using blogging systems as easy internal publishing systems, as for Seattle Children’s Hospital, is also easy to justify. Everyone in companies needs to find internal manuals, forms, and phone directories, and they should be published and maintained using the easiest technology possible.

In other cases, a little more thought has to be given to what should be achieved. Outward-facing blogs and wikis have to fairly represent the company. Employees who blog need to be aware of what they can say and can’t say (the Sun blogging policy is an example of the sort of policy that employed bloggers should follow). Managers need to avoid telling people to blog; forced postings or forced bloggers are seldom of a high enough quality to be worth the time spent writing them. Bloggers also need to be responsive to people who comment on their public postings, not defensive. Wikis can be a great source of public input, but someone needs to act as “editor”, ensuring that all edits made are on the right topic, and wiping any spam or abusive material immediately.

If the idea is to create even a simple knowledge management system, then more thought will have to go into figuring out the requirements and how to motivate people to use the system. For these more encompassing solutions, the grass-roots approach where people can choose to not use the system is seldom as effective as the top-down approach or not giving them a choice. Some of the factors to consider are:

  • Ease of use
  • Company culture
  • Company size
  • Categorization (metadata or labels) and/or indices so the information can be found again
  • Long-term versus short-term information – are some documents edited a lot over their lifetime? Or are completely new versions created, such as for meeting minutes? Should the information be archived, for example for competitive intelligence where you want to assess the likelihood of a competitor’s actions based on their historical actions?

These last two are related, of course. “Pure” blog and wiki systems both make it easy to find the last edited or last created page; blogs because of the dating system and wikis because of the “recent changes” page. To find relevant older information requires some sort of metadata or suitable search engine, or a link from a recently updated document. Wikis suffer from the particular problem that it is easy to “lose” pages that have been written if nobody has linked to them. It is also extremely easy, given even a small group of authors, to end up with a nest of linked pages without being quite sure what is in any of them, so that newcomers to the group have to spend a long time following links to find anything. And without constant maintenance, the links and pages in a wiki have a tendency to go stale. A fuller exposition of ways to help avoid these problems is given in Leigh Dodd’s blog article on the subject.

The Socialtext commercial wiki solves this problem by integrating blogs and wikis in such a way that when a page (wiki page) is edited, it goes to the top of the list in the same way as a blog posting would. And it’s easy to find pages with lots of links to them. This system does rely on having the right conventions set up so that people understand the way a given company or department has structured their information. When used correctly, it can be powerful. Ziff-Davis’ Gaming division cut down on email and increased productivity significantly by using this system for day-to-day coordination, scheduling and requests. It also helps create a “group memory” as more useful documents tend to be linked to or edited more often than less useful documents.

Traction Software, as an example of a different way of structuring the information, uses more of a project and label metaphor on the blog postings so that people can search for information about specific projects or with particular metadata and have a list of relevant postings returned. The postings are often gateways to documents, such as meeting minutes which could be written in some other tool and linked to from the blog posting. The entire system is permission-based to prevent people seeing or editing information they’re not meant to. It can also pull information from different sources to provide a summary for project managers and executives.

As blogs and wikis and hybrids become more widely used in the corporate world, I expect to see more emphasis on permission-based editing and viewing as well as security. Blogging systems always have permission-based editing and often permission-based viewing, even for personal blogging, but wiki systems typically are designed for open, public editing so that restricting who can see or edit which page is not an integral part of the system. We’re just at the beginning of the blog and wiki era in corporate use and it won’t take long before these tools are used as often for collaboration as email is today.

Acknowledgments

I received a lot of input for this article. Some came in the form of comments or email after I posted an article on my blog, asking for input. Others gave freely of their time on phone calls or meetings. I very much appreciate all the input and wish to reiterate that any errors in this article are mine. I received information from: Bob DuCharme, LexisNexis; Christian Watson, Seattle Children’s Hospital; Christopher Mahan, Health Net Inc.; Dave Pawson, RNIB; Derek Miller, Navarik; Jordan Franks, Traction Software; Leigh Dodds, Ingenta; Malcolm Tredinnick, CommSecure; Norm Walsh, Sun Microsystems; Richard Tallent, Environmental Resources Management; Robert Scoble, Microsoft; Ross Mayfield, Socialtext; Tim Bray, Sun Microsystems; and Tony Coates, London Market Systems.