Hi Everyone,

GMX-V (Global Information Management Metrics eXchange - Volume) is a draft LISA OSCAR standard for word and character counts (http://www.lisa.org/standards/gmx/). This should bring some sanity to word and character counts. GMX-V is predicated on a canonical form that is based on XLIFF.

We are almost at the end of the internal GMX-V review before it goes out for final public comment. We have had a final discussion concerning inline elements that imply some form of white space, but may not be necessarily preceded of followed by any spaces e.g.

<p>This is a<br/>that is not preceded nor followed by spaces.</p>

This instance would be rendered in XLIFF as:

<source>This is a<x ctype="x-html-br"/>that is not preceded nor followed by spaces.</source>

In the proposed GMX-V standard it is recommended that the program that is creating the XLIFF file precedes the any inline elements with spacial characteristics with a space if there is not one present:

<source>This is a <x ctype="x-html-br"/>that is not preceded nor followed by spaces.</source>

This is specific to XLIFF files created for GMX-V purposes.

I was wondering if it would be desirable/beneficial to consider this requirement within the XLIFF specification itself. Certainly having a space would assist any segmentation algorithms etc. The onus would be on the merge program to remove any such spaces that were inserted. It is only the extraction and merge processes that can do this as they have the required intimate knowledge of the source environment.

Please find attached the latest version of the GMX-V draft specification (version 0.7).

Best Regards,

AZ