The Dictionary Unit for South African English (DSAE) was established in 1969, as a project of the Institute for the Study of English in Africa at Rhodes University, to collect and record English as it is used in South Africa. Since then, it has compiled extensive archives which reflect the diverse influences which shape South African English.

Which problem did your company have?

Our institution had a history of working in SGML documents and when I arrived in my current post it was clear to me that we would have to convert existing data to XML, and that data capture mechanisms for new projects should output XML.

To support this move we needed XML-editing and authoring software which would make this task painless for non-technical users, and it was preferable to find an XML editing application which would allow us to benefit from existing and emerging XML specifications at the time. Serna also promises to integrate with other applications, so we are not limited in what we do with it in the longer term.

How has Serna helped to solve the problem?

By making XML easier to edit by non-technical users.

How is Serna used in your company?

Mainly to edit dictionary-related XML data, including a quotations database and associated bibliographical records. In some cases also project-related workflow data encoded as simple XML records.

I'll give one example of our workflow. In the past, quotations appearing under dictionary entries would have had to be edited in individual files containing the whole entry as well as quotations related to it. Complex styling rules had to be implemented manually, with much room for error. Now, through the combination of XML mechanisms and Serna's ability to implement them, we can store the quotations in an XML database. They are accessed by searching a Web interface, opened in Serna from a Web link, and presented in different views to the database editor.

Almost all the styling rules are now handled automatically by the stylesheet in the background - previously editors had to spend months learning these rules - and we can now avoid storing duplicates in different locations.

Moreover, bibliographical information such as author names and book titles is now entered through predefined lists displayed as pull-down menus in Serna combo-boxes, thanks to Serna's ability to parse our separately-maintained, external bibliographic documents dynamically via XSLT. With several thousand authors and work titles, this reduces the possibility of error and inconsistency greatly.

This quotations database is then searched for its own sake when researching the historical usage of words (again through the Web interface), and when we begin revising other parts of the dictionary, it will be trivial to import them into other records and publish them online.

What is the most useful feature of Serna?

For me, it is XSL support and the way this allows the developer to package data for users and make the interface interactive without a lot of complicated GUI programming.

How did you learn about Serna?

Via Google. I spent a few weeks researching XML editors in depth and trying to match them to our needs, before choosing Serna.

What are the main reasons you decided to use Serna?

It is technically very sophisticated and highly customisable, but the interface can present a very simple view of the product and the XML document to users.

Why did you choose Serna rather than a competing tool?

On the technical side, the design demonstrates a commitment to the XML standards which many XML proponents write about, but few XML authoring products actually implement: e.g. XSLT and XSL-FO. The explicit support for XSLT was one thing that made me choose it over XMLSpy at the time. The Python plugin functionality was also a drawcard.

On the usability side, the interface has several types of graphical views, allowing one to tailor its use to almost any level of user, and at the time I did not find another XML editor which was as sophisticated and flexible as this. I haven't found one since, though I haven't looked very hard at other products in the last two years.

How do your employees feel about using Serna?

They had not edited XML data before I introduced it via Serna, except in Wordperfect 12's XML authoring component, so they do not have a lot to compare it with, but I think they would prefer Serna to any other XML-authoring product. (When we used Wordperfect it kept crashing and the Corel support team could never solve this.)

I have not had any complaints about the use of Serna, and users do all like the results (e.g. automatic report-generation from the XML data they edit). Apart from myself most users are non-technical, however, and new temporary assistant users are able to get started with minimal training and have very few queries, which speaks for itself.

You mentioned "automatic report-generation". Could you describe how it is done?

We use Serna for task lists and workflow tracking, often by embedding related markup inside dictionary entries which we are editing or reviewing. Since this is in XML, we can use XSLT to generate statistical information and document summaries automatically.

These summaries, if they are short, can be placed at the top of a document full of dictionary entries (for example listing entries according to various status categories such as "do we want to publish this entry? Yes, No, Maybe; who is working on this and at which stage is the work at; start date, at what stage of editorial review" etc).

If the desired report is to be very detailed or must integrate information from many XML documents, we usually put this on our Web server. Again XSLT is used, and these can provide quick workflow views, as well as links to specific content, which are very popular with our staff.

How different is Serna from the tools your company previously used?

It is technically far more advanced and user-friendly than Wordperfect, the only other XML-capable authoring program we have used for dictionary authoring.

Did you customize Serna? Was it difficult?

My customisation has been in developing custom XSL stylesheets for our datasets, almost all of which use custom XML markup.

The XSLT development was not difficult once I taught myself the basics of XSLT.

What do you see as the main benefit of using Serna?

It allowed us to get started editing a large XML dataset quickly with minimal training and development. Development and data integration can be iterative -- you don't need to do it all at once. Data integrity is essential for us because our records are highly structured in sometimes subtle ways, and the validation features (including the possibility of adding user feedback via XSL) allowed us to work safely, even when introducing unattended trainees to new tasks. And there is scope for much more custom development.

How does Serna affect the workflow?

The use of XML affects our workflow by making tasks more structured and efficient in the long run, and of course Serna facilitates the use of XML by making it easy for people to use it.

How has Serna impacted on the business?

In short, it has allowed us to adopt industry standards painlessly and prepare the way for a more efficient workflow than was possible prior to the arrival of Serna.

What advice would you give to another company about when or why they should use Serna?

Unless you are sure you will be using only the predefined stylesheets included in Serna, you will need one person who is prepared to develop XSL stylesheets and manage technical aspects of configuration and so on.

Beyond this, I think it is a suitable product with which to introduce XML to your users, as well as a suitable product to stick with for the long term as your company's needs and skills evolve. If you can find alternative XML authoring programs, use them if you think they are better suited to your needs, but note that many of the flashier products actually limit one's ability to adopt XML standards fully, which defeats the purpose of using XML in the first place.

Also test and compare the level of support you will get from other companies. My experience of Syntext is that its staff are responsive, thorough, very well-trained and willing to accommodate the needs of even a small client such as ourselves.

• Create quality XML content in familiar environment on any platform
• Accelerate content creation with global collaboration
• Boost quality and consistency by reusing content
• Publish XML content to multiple channels in a variety of languages

Learn more about Syntext Serna XML editor benefits.

"Aside from the affordable pricing, positive feedback from our users and good support from Syntext, Serna made our transition to XML painless for users." - Tim van Niekerk, Editor - Systems Developer, Dictionary Unit for South African English, South Africa.

Serna Enterprise XML editor has been used successfully by customers worldwide in all kinds of industries:

• Airlines
• Aerospace&Defence
• Electronics
• Health Care
• Media
• Publishing
• and others