Re: [xml-dev] RE: Percentage of XML documents exclusively processedby ma

[Lech Rzedzicki <xchaotic@gmail.com>]

If that percentage is to be used as a metric for optimising future
approaches to data exchanges then it obviously needs to include other
metrics and prioritise them.
In my view computers and XML are here to make life easier for humans
and that should take precedence in most cases.
Given an area that I am intimately familiar with that is producing
EPUB from XML, some of the raw XML may never be looked at in it's raw
form, but it is tremendously helpful to have it in human-readable form
when things go wrong. It may be that 1 in a 100 (obviously you never
know which one) titles has something wrong in them, perhaps a single,
badly formatted unicode character and one go armed with nothing but a
notepad.exe and fix the title to go on sale just before Christmas. In
my domain that's usually worth more than processing efficiencies.

Also, like others have said, it is impossible to tell what does it
mean that the XML is read by humans, in a somewhat literal
understanding, slightly stylised OOXML is being read by millions of
casual MS Office users every single day.