LISA Home page [© 2010 • ISSN 1420-3693 • www.localization.org]
© 2010 SMP Marketing • ISSN 1420-3693 • www.localization.org
Making XML Work for You

Dan Dube, Director, Business Analysis and Development, Innodata Isogen

XML has been held out as the wave of the future and billed as a cure-all for content problems, but the reality of XML has often fallen short of the promise for a number of reasons. In this interview, Dan Dube, Vice President of Business Analysis and Development at Innodata Isogen, discusses how to make XML work for you and how to avoid common problems.



Dan Dube

Insider: Obviously, you are a proponent of XML technologies for multilingual solutions. Briefly, what can XML deliver for multilingual publishers, and why should a company look at using XML rather than programs like FrameMaker, Quark XPress, Adobe InDesign, etc.?

Dan Dube: Our fundamental belief is that content is a key corporate asset, and technology should be considered an enabler to support that asset. Technology evolves quickly, and tools will inevitably become obsolete. (Just ask those technical documentation departments that invested heavily in Interleaf in the late 1980s!) It is inevitable that companies will eventually need to upgrade or replace their current publishing technology infrastructure. If the content conforms to a standard (XML), then as long as the new technology supports that standard, your content will be portable across applications. This allows you to preserve the investment you’ve made in that content and relegates the publishing technology to a commodity item.

Content is a key corporate asset, and technology should be considered an enabler to support that asset.

The key advantages of migrating to an XML-based solution for publishing are the following:

  • Reuse of information: XML-based technology allows you to store information at a lower level of granularity than a “document.” Instead of storing chapters or books as files, you can store discrete information components (e.g., sections, warnings/cautions, etc.). These information components can be shared by multiple documents; updating one information component can ripple the changes automatically to all documents that share that piece of information. Furthermore, you’ll only need to localize that piece of information one time; only information that changes between localization cycles needs to be sent to a localization vendor for analysis and translation.
  • Single source publishing: Because XML is format-neutral, you can author the content once and automatically render it to multiple output formats, including (but not limited to) PDF/PostScript, HTML, compiled help, etc.
  • Semantic markup: XML provides the ability to add attributes/metadata to your content, which makes it easier to find and retrieve information that is relevant for your needs.

I’ve yet to see an example of a document in the documentation field that doesn’t have some amount of inherent structure behind it.

All of these factors have enabled our customers to achieve significantly faster time to market, dramatic reduction in publishing and localization costs (averaging between 50-70% cost savings), and the ability to do more work with the same or fewer number of staff.

Insider: An XML solution may be overkill for companies that produce little documentation. At what volume does XML become a viable proposition?

Dan Dube: The cost of commercial XML tools has come down rather dramatically in the last 18 months, and there are also open-source alternatives available. We’ve built ROI justifications for very small production shops to make the investment to migrate to XML. We have customers that have invested less than USD 100,000 for software and services and have achieved positive ROI in under six months. In another instance, one client (a government agency) justified an investment in XML for web and print distribution of press releases releases.

XML enables a dramatic reduction in publishing and localization costs (averaging between 50-70% cost savings).

Insider: In your experience, how long does it take to realize a return on investment from utilizing XML?

Dan Dube: We strive for a goal of helping our customers completely recoup their initial investment within 6-12 months of going into production with an XML-based environment. In every case, we’ve met or exceeded these goals. This is primarily due to our implementation methodology, which emphasizes starting with a production-capable pilot environment that is a representative subset of the overall production environment. By quickly implementing and using a pilot environment in a “live” production capacity, the investment in the pilot environment can be recovered while the system is being scaled up into production with other documents.

In my experience, translators like working with XML.

Insider: What drawbacks are there for XML use at present?

Dan Dube: In my opinion, there aren’t really any drawbacks to the XML family of standards or the technology to support those standards. Any “failed” XML implementation projects usually are due to one of the following:

  • The customer purchased the wrong technology for their needs. This is very common. For example, some well-known content management vendors will present their product as the solution to all of their issues, without even taking the time or initiative to understand the unique intricacies of that customer’s unique production environment. (“Here’s the solution… now what’s your problem?”) Our approach is to analyze a customer’s environment, gain an understanding of their requirements, and then recommend the best mix of technology to meet those requirements.
  • Fear of change. If the end users are not involved in defining their requirements and given the opportunity to get excited about the benefits that the new system may bring, then they won’t see the value of changing their comfortable work environment, and the project will likely fail.
  • Biting off more than you can chew. In some cases, a company will try to do a corporate-wide deployment of a new system environment. This is a recipe for disaster (and the primary reason that we are proponents of the controlled pilot approach to implementation).

Insider: XML obviously works best for structured-document workflows, but even many so-called structured documents are not in fact very structured. These issues may arise when moving to XML since it enforces structural concepts in documents. How can companies gauge the suitability of their documentation for an XML workflow?

Dan Dube: I’ve yet to see an example of a document in the documentation field that doesn’t have some amount of inherent structure behind it. The best recommendation I can give is to have a qualified professional analyze your content suite to find opportunities for structural patterns, information reuse and metadata enrichment. A relatively minimal investment in this level of analysis can help determine the business case for moving that content set into XML.

Insider: Moving to a uniform XML process from a current heterogeneous publishing solution will obviously require adjustment and take time. What are the most critical steps in making the transition, and what steps cause the most problems?

Dan Dube: Here is a short list of my keys to successful change management when migrating to an XML production environment:

  1. Communicate prior to beginning the project
    • make sure that there are clear and measurable business objectives communicated by senior management
    • solicit feedback and promote awareness from users at all levels of the organization
  2. Communicate during the analysis phase of the engagement
    • requirements analysis should include feedback/representation from all stakeholders (e.g., authors, editors, localization managers, production, etc.)
  3. Communicate during the pilot implementation phase
    • publish newsletters to all interested parties throughout the organization
    • set up a lab or demo environment to showcase the new software environment
    • schedule regular participant seminars
  4. Communicate during the production rollout
    • publish newsletters to all interested parties throughout the organization
    • set up a lab or demo environment to showcase the new software environment
    • schedule regular participant seminars
  5. Control the scope of the rollout
    • phased implementation rollout
  6. Consider using an experienced systems integrator
    • they’ve done this many more times than you have and can see the “danger signs” early enough to adjust/redirect project progress to ensure successful outcomes
  7. Remain focused on the business objectives
    • the only way to measure success is to ensure that management’s business objectives have been achieved in a measurable way
  8. High-level sponsorship
    • make sure you have senior management support to help drive change across all stakeholder organizations, or the project is doomed

Insider: What types of documents are most ideal for an XML workflow, and which are least suitable?

Dan Dube: The most suitable document types are those that have a long shelf life (e.g., they will be maintained/updated over a period of several years), and/or are made up of “information components” that can be shared or reused. In addition, documents that need to be distributed to multiple output types (e.g., paper and web delivery) are good candidates for migration to XML.

Documents that are considered to be one-off are less appealing candidates for the investment required to move to XML (e.g., a newsletter).

Insider: What advantages does an XML workflow provide for localization? Are there any disadvantages? In your experience, how do translators like working with XML files?

Dan Dube: The primary advantages that our customers have measured and shared with us fall into these major categories:

  • Lower cost of localization (because only changed content is ever sent to the vendor for translation)
  • Lower/eliminated desktop publishing costs (due to the implementation of automated multilingual composition tools based on XML)
  • Faster time to market (because there is less content to translate and no DTP)
  • Higher revenues (because there is an ability to get product to market sooner in foreign markets)

The primary disadvantage is the up-front investment of time and money to migrate to XML. However, this is usually overcome by the quick recapture and ROI of the investment costs.

In my experience, translators like working with XML. It eliminates the need to do messy file conversions (e.g., FrameMaker to RTF) when using Translation Memory (TM) tools. They receive XML files, run them through a TM system, clean them up and deliver localized XML back to the client. In many cases, it allows the localization vendor to provide a higher level of service to a greater number of customers.

Insider: What resources are available to help companies plan a transition to XML?

Dan Dube: There are a number of service providers that claim to offer these services. They comprise a wide spectrum of companies, from the large traditional management consulting firms and systems integrators to small boutique XML consulting firms. To my knowledge, Innodata Isogen is the only service provider in the industry with the breadth of experience and staff to focus on optimization of a client’s entire content supply chain.


Dan Dube has over 19 years of experience in business process re-engineering and implementation of complex information management systems. In addition to leading the implementation of XML-based systems worldwide, he has also designed and implemented automated localization systems based on XML components for several Global 2000 companies. You can reach him at ddube@innodata-isogen.com.




Contents


LISA Business Data

LISA Publications Catalog

Industry Insights Reports

Best Practice Guides

Surveys

QA Model

Forum Summaries and Presentations

LISA Globalization Consulting Network

Webinars and TouchPoint Advisory Calls


Join LISA

Subscribe


Upcoming Events

LISA Forum USA
(Foster City, California, April 13–16, 2010)

LISA@Chinasoft Fair
(Chengdu, China)

LISA Forum Asia
(Suzhou, June 28–July 1, 2010)

LISA Forum Europe
(Budapest, October, 2010)

LISA Forum India
(New Delhi, December, 2010)


Open StandardsTBXTMX

Terminology SIG

Job and CV Postings