Integrating reading-level calculators

[bburns]

For various reasons that I won't go into here, we need to generate reading levels for our publications. This has to be done at the map level as there needs to be enough text to get a "reliable" score. I see that it's possible to integrate external tools via Tools >External Tools > Configure...

Has anyone integrated an external reading-level calculator using this approach or some other?

Thanks in advance.

[Radu]

Hi Bill,

Interesting, honestly did not know that such tools existed.
The problem would be that the external tool would have access to the DITA Map but it would need access to a resolved DITA Map, with all conrefs and topicrefs expanded.
How about if on the DITA Map root element you set the chunk="to-content" attribute and then publish the XHTML so that a single very large HTML document is obtained? After this you can open the HTML in a web browser and copy the content from it...
Or at some point Jarno Elovirta created a plugin which outputs plain text from a DITA Map com.github.dita-ot-plugins.plaintext:

https://github.com/jelovirt/dita-ot-plugins

So I think it would be feasible to creating a DITA Open Toolkit plugin which first obtains all the text from the DITA Map and then calls an external process to score it. For example I created some time ago a DITA OT plugin which generates a single merged DITA file from the original DITA project:

https://github.com/oxygenxml/dita-merged

Applying an XSLT stylesheet which only extracts the text from that file and then calling an external process to score the text should work.

What scoring tool were you planning to use?

Regards,
Radu

[bburns]

We're using Flesch-Kincaid, but we've looked recently at other indexes that seem to provide more reliable results. We have a process that runs on both maps and topics, but really only the map reading level is useful. Just trying to determine how we might implement this in Oxygen to work as it did in our previous editor.

[Radu]

Hi Bill,

Maybe you could give me more details about how this used to work in your previous editing solution.

Regards,
Radu