Volume 1

Issue 3

September 2004

Feature Article Tony Byrne Talks Technology Tony Byrne CMSWatch tbyrne@cmswatch.com While content management is not all about technology, technology is a critical component of any content management solution. After all, according to Tony Byrne, "you can't do a CMS implementation without technology." Tony Byrne has established himself as a leader in understanding content management technologies and their role, so if you're looking for information on how technology supports content management, then CMSWatch is the place to look. CMSWatch.com, founded by Byrne in July 2001, provides an independent source of information, trends, opinion, and analysis about Web Content Management (WCM) solutions. In this issue of The Rockley Report, Tony Byrne discusses the role of technology and provides tips on the technology selection process, with much emphasis on the need to "try before you buy." For more information on CMS technologies and to subscribe to the CMS monthly digest of new articles and findings, visit CMSWatch.com. Q. What is your background? I started my career as a radio reporter and magazine publisher, but got involved in international exchange and technical assistance with the fall of the Berlin Wall and opening of Eastern Europe. Through an international non-profit, we developed some of the first public e-mail networks in Russia, Central Asia, and Eastern Europe, and started an "Internet Peace Corps." We learned some hard lessons about the difference between network availability and network adoption, particularly in cultures with decades-old legacies of strict information control. Some of our most fruitful work entailed putting Internet stations in public libraries and nascent student unions. That program (now funded by the U.S. government) continues today. When the Web hit in the mid-90s, I migrated to the commercial sector and joined a company building e-commerce storefronts. Q. What drew you to content management, and in particular, to content management technology? At our web development firm, our customers naturally wanted to start updating content themselves. We built some homegrown tools, but quickly saw their limitations and were concerned about the overhead required to maintain and enhance them. So we started working with various commercial technologies (early versions of Interwoven, Vignette, Spectra, etc.), and experienced what now seem to be very familiar challenges: the software was under tested and difficult to customize; there were severe usability problems; authors and editors saw little daily improvement in their work, etc. Some of our better accomplishments actually came from working with open-source CMS tools, despite their substantial learning curve for developers and authors. I began to think there was a need for a public resource to review web content management technologies and the practices around them. So I founded CMS Watch (www.cmswatch.com), and wrote The CMS Report to evaluate tools. The report is now in its 6th edition. Q. What is the most common mistake companies make when selecting a tool? How would you recommend companies ensure they make informed decisions when selecting technology? Every consultant will tell you that the most common mistake is a failure to adequately prepare and prioritize user requirements. I agree. Beyond that, I would say one of the most common (and little understood) mistakes is a failure to actually test systems thoroughly to allow all parties to get comfortable with them-warts and all. This can be an informal "bake-off" or a formal proof-of-concept; in either case, you need to do it before you sign a contract. Some vendors will resist this tooth and nail. You need to hold your ground. A little while ago I went into a large company to participate in the implementation of a major Web CMS package they had just licensed. There was a meeting during which the customer's security guru informed the team that the vendor's approach to promoting content to the "live" servers would violate corporate security policies. The project leader didn't exhale for about 3 minutes. She sensed that the security problem-which had slipped through the cracks in their requirements gathering-would resist resolution. Unsuccessful testing of various work-arounds suggested by the vendor proved her right. The company had to write off the cost of the software and begin their search again. Testing products head-to-head is difficult and laborious, but it really could have saved this company a lot of time and money. Testing also forces you to think about use-cases (or "scenarios"), which should drive product selection in any case. Q. What role does technology play in a successful content management implementation? Well, you can't do a CMS implementation without technology. It's necessary but of course not sufficient. An important question to ask, though, is whether you need a content management tool to solve your content management problem, and if so, what kind of tool. A lot of basic design consistency problems can be solved with Dreamweaver templates. Valuable library services can be provided through a simple WebDAV server. Not everyone needs a CMS. Q. Is the acquisition of technology the most important decision to make in a content management implementation, or is it an important one among many? What are the other key areas of content management a company should focus on? The acquisition of technology is important because you can make a really bad mistake here. On the other hand, after doing their homework well, most companies can get down to 3 or 4 solutions that could all work out very well. The other thing I encourage people to do is evaluate the implementation team as carefully and critically as you evaluate the software-even if the implementation team is sitting down the hall. You're almost surely going to spend more money on services than software, and the quality of the implementation will likely have a greater impact on your overall success. So proportion your selection energy accordingly. Finally (and this is no surprise to your readership), without proper content and process analysis-along with some vision about how you are going to improve your business by managing content better-all technology investments are in vain. Q. You have done a lot of work in the field of web-based content management, but you are starting to do more in the area of enterprise content management. What prompted this move? How are these two areas of content management the same or different? I started getting more involved in enterprise content management because there is some convergence happening, although not the kind of convergence that many vendors talk about. Part of what's going on, I think, is that web content managers want to expose more content from deeper in the enterprise-sometimes integral documents, sometimes content chunks or components. At the same time, enterprises are seeing customer and employee demand for data and content from all sorts of different information systems to be made available via the web. So more documents and assets are coming to the web, and vice-versa. Of course, many enterprises also want to reconcile print and electronic publishing processes further upstream (and they are all finding it very hard!). Another big challenge here is combining data and documents in various useful ways. For example, a common scenario in the mutual fund industry is to be able to merge brochure-type content about a particular fund with current data about its performance. Enterprise search engines are starting to offer some interesting capabilities here. The key thing for any organization embarking on an ECM strategy, I think, is to be brutally rigorous and specific about identifying high-value use-cases for information integration, then tracing those to specific repositories and management systems, rather than starting from the premise that you need to be able to make all content and all data within the enterprise manageable inside of one über-dashboard. That's not realistic. No one really knows how this is going to play out from a technology perspective. Analysts keep predicting that major platform vendors (SAP, Microsoft, IBM, Oracle) are going to dominate the CMS space. That hasn't happened yet and won't happen soon, if ever. However, the roll up of various content management products within major ECM vendor "suites" continues apace. The products in these suites are really free-standing tools packaged together for marketing purposes, but usually sold, implemented, and supported by separate internal groups within the vendor. Sometimes people ask why ECM vendors don't just dissolve all these tools into a single package to create a truly "enterprise" solution. After all, document management (DM), records management (RM), digital asset management (DAM), and XYZ management products do fundamentally the same thing: they ingest content, enable repository services, employ metadata, support workflow, allow you to decompose and recompose derivatives, then output or archive content in various formats. Is there another über-dashboard here for all your content? I don't think so. It turns out that these different tools tend to get used by different people within the enterprise, who employ different authoring systems, have varying interface needs, and create diverse downstream products. Sure, these systems often need to work together, but we are all still figuring out how, when, and where. Smart managers will ask "why," too. So I think the product families (DM, RM, DAM, WCM, etc.) will remain distinct, at least in the near-term. ECM suite vendors seem to agree. They are expending much more effort cross-selling these tools than integrating them. I would like to see greater focus on the shared methodologies and disciplines required to make any of these various content management products work effectively, and less focus on the need to buy multiple packages from the same supplier. For example, we could probably all learn a thing or two from the document imaging specialists who may be laboring somewhere around the corporate mailroom. They've been dealing with metadata, indexing, storage, and workflow for decades, and probably lay the most legitimate (and certainly the most longstanding) claim to mantle of "enterprise" content manager. Q. What is your favorite CMS tool? People ask me this a lot. I don't have a favorite. Really. There is a right time and place for nearly every product. The trick is the find the right tool for your circumstances and budget. I don't think magic quadrants and other horserace-type rankings are helpful in this regard, because they don't tell you how a particular solution works, and what it will likely cost over time. All the more reason to try before you buy.