Blogs

I have just added some new API code to the bulkpub module to allow an application
to query which fields exist for a content type, and to set field values when
calling the newPage() API. I am using this in an application I am developing that
uploads xhtml derived from genealogy GEDCOM data. In the future I plan to use it
to attach fields to DITA topics that could include:

The Bulkpub module provides the necessary APIs to do bulk publishing to a Drupal 7 website. This blog explains how to write a script to drive the publishing process.

In order to use the Drupal 7 Bulkpub module to upload content, you need to write a script to drive the process. To give you an idea how to do this, here is the rough outline of the script we used to upload the Bulkpub documentation itself to this site using the Bulkpub APIs.

I see many job postings that ask for a person with skills writing XML transforms. The posts almost always end up assuming that such a person needs to be an XSLT expert. This need not always be the case! While it it is true that XSLT can be used to transform XML, there are reasons why you might want to consider doing XML transform using another programming language. My favorite language for transforms is Python.

Here is why I often choose to script in Python or PHP instead of using XSLT:

I have used my wsdl2dita Python script to generate DITA source files from the WSDL file describing CMIS (Content Management Interoperability Standard), which is sponsored by OASIS (Organization for the Advancement of Structured Information Standards), the same organization that sponsors the DITA standard.

Using the DITA Open Toolkit I generated a PDF file, which is posted on this site's download page. For those curious, you can browse through this PDF (also available on the Downloads page) to get an overview of the web services and operations that make up the CMIS standard.

Here are some of the key WordPress and Drupal features that we use in our DITA-based information mashups:

WordPress (xmldocs.info):

One nice thing about publishing documentation on a Drupal or WordPress site is that it is easy to instrument the site to find out which which topics are being read and searched for by the users. Here is what we have seen in the last few month's activity on our content-rich model sites.

We have seen daily search requests from people looking for samples and examples. Both the garage and grocery shopping topics are popular.

People are interested in the basics of getting started with DITA. The various planning, tutorial, and "how-to" topics are also popular.

WordPress 3.1.2 and Drupal 7 (our initial model information mashup publishing platforms) have the following common characteristics:

We would expect information mashups to be popular with organizations that have one or more of the following needs and characteristics:

We believe our WordPress- and Drupal-baed information mashups illustrate the key benefits of a combined structured/unstructured information solution:

In choosing the initial publishing platforms for our model information mashups, we had the following objectives:

DITAmashups are information mashups in which the structured topics are DITA based; that is, the structured project is made up of XML/DITA source files.

We believe that the DITA structured information standard and the tools that support the standard are an excellent fit for structured information projects, but not for unstructured information projects.

The diagram below provides more information about the differences between DITA-based and unstructured information.

Information mashups are collections of structured and unstructured content combined in a single, web-based location so they can be browsed and searched as a single body of information.

Posting the information collections in a single location makes them more useful to users; creating and publishing them separately allows the content owners and originating authors more flexibility in creating and managing the content.

I confess I always used to hate DITA short descriptions: to me most of the ones I produced or read seemed like either "motherhood and apple pie" statements with no information value or statements that might serve to introduce a topic, but didn't work well in isolation (for example, as hovertext).

I now have a solution I'm happy with: The short descriptions in the DITAinformationcenter have been programmatically produced from the keyword tags associated with each topic, and they appear with the topic and sub-topic titles in mini-tocs, as well as in hovertext over links.