Re: [web-annotation] The textDirection and processingLanguage properties are not needed

@azaroth42 well ...  this is not the message that I wanted to send ...
> .. "I don't like the way you wrote that" 

The message is "It is not understandable who needs this and when, and 
what for".
This is completely unclear, and there are many people that already 
indicated that the two fields are not solving the reported i18n 
This ground  i18n problem is not described in the standard, and I 
think that the people that identified the problem, and the ones that 
discussed and proposed the solutions are the ones that are in the best
 position to describe the problem in details.... not the ones that 
indicate this as an improper solution!

If I remember correctly, the motivation for the processingLanguage, 
was saying:
"The NLP algorithms needs exactly one language in order to work 
propertly, therefore the existing dc:language is not a solution".

In other words we are trying to solve in Annotations, the problems of 
NLP algorithms. That's just wrong from modeling point of view! And 
apart of that, why should be one resource that uses mixed languages be
 processed exactly by one NLP algorithm? What is the argument agaist 
using 2 NLP algorithms to process the text that mixes two languages?

The main reason I remember for the introduction of processingLanguage,
 was the cardinality. While dc:language has multiple values, the 
processingLanguage was set to have cardinality 0...1. 
I think this is a very week motivation, and partialy wrong. Apart of 
that it is not clear at all which is the "correct" value for this 
field, who should set it, and what should be the consequence of its 
presence in the annotation.
So .. maybe it is the time to split the issue in smaller ones to be 
able to discuss the details.

GitHub Notification of comment by gsergiu
Please view or discuss this issue at
 using your GitHub account

Received on Friday, 5 August 2016 08:21:39 UTC