Let’s talk about text baby!

Let’s talk about text baby!

A question I get all of the time is ‘can we do [S1000D] XML with free text editors?’ So in short, the answer is yes, of course, you can – but very quickly I will ask “but why would you want to?”. I can fly to New York from the UK, in a nice comfy seat, but I could row in a dingy too if I wanted. One is far more efficient, safe and sensible if a little more costly, the other requires much more effort energy and of course riskier.

Also Available as a PODCAST – Listen on the TDW App

In this article, I am talking S1000D & XML based data modules, but the same rules and considerations can apply to SGML. However, SGML is slowly, and I mean slowly, dying out, and no serious, self-respecting document strategy will use SGML in the 21st century.

Now full transparency – I use text editors all the time, they have a role to play in any structured information process. However, there are severe limitations and real reasons why you would not want to base your XML strategy on open text editors. They have a place in most technical authors toolkits but as part of a kit, not the entire bag!

A quick Google search will show you the endless list of freely available text editors on the market – Wikipedia even has a dedicated page around text editors. I have three go-to tools that I use when it comes to raw SGML/XML source viewing or editing. TextPad, NotePad++ and Dreamweaver (yes Dreamweaver) – Adobe Dreamweaver is mainly used for web development and is not a free tool, but it is part of our subscription with Adobe, and it is a potent tool beyond structured editing and is an excellent tool for a technical author/web content developer – blogs for another day!

However, focusing on free tools only, TextPad and NotePad++ are my two main tools. I can open up XML and view the source quickly and navigate the content for quick review. I can make minor adjustments and save the XML straight back out again without the need to load up dedicated structured editors and their associated applications. This approach is a serious time saver when needing to make minor adjustments or do a quick content search/replace or check.

The problem with text editors is that they are dumb, they (in the main) do not perform any validation or verification of the structure that you may have just edited. There is no guidance to content editing, and they just let you write. A minor omission or random character will throw errors when we move back to our heavy hitting tools. The open text editors will only save the file as ‘XML’ and move on, where industrial-strength tools and applications will perform a validation check and warn you of any errors that may exist within the file.

Even the very best of us have accidentally deleted or included an extra chevron or inconsistent tag, used the incorrect ISO character code or inadvertently adjusted a reference destination – all calling our XML source integrity into question, and we enter a fault-finding cycle with our structure. We no longer are writing our content; we’re fixing the errors we have introduced – beyond textual content!

S1000D is no different (after all it is just a fancy name for XML) – I have seen projects attempt to produce data modules in freely and openly available text editors, technical authors are spending more time trying to ‘build’ the XML rather than focus on what goes between the tags. I have seen everything from structurally invalid content right through to unusable material that will need rework when it hits a professional, structured working environment or sent to someone who wants to use it.

A story for you:

I was called into a significant project to discuss the delivery of S1000D data modules and was asked to demonstrate how the [S1000D] content can deploy. Rather than theorise, I suggested that we took some of the clients existing data modules imported and published them. The technical author at the time turned white when he saw that his modules would not import into the CSDB nor would a structured editor open the tool without throwing a wobbly and a raft of warnings that the content was structurally invalid.

The technical author had some serious questions to answer and is no longer with the organisation. The data modules were eventually reworked, imported and published – but at what cost?

In summary text, editors have a role to play, but they are not a substitute for industrial strength structured editing tools. Structured editing tools make the authors life easier so that they can focus on the content as opposed to the tagging that is happening in the background.

That does not mean that technical authors do not need to concern themselves with tagging, we often need to look at the tags to see what is going on, but this is not the norm, the norm is to focus on content and let the appropriate tools write the necessary tags for us. Text editors are a valuable and often free addition to our content creation toolkit. 

What tools do you use? I would love to hear what tools you use for your content creation.