VEP-007 update

[Baptiste Cecconi]

Hi Markus, 

I think this will do for now. 
We may have to iterate with a new VEP later if we need to refine the #documentation vs #metadata discussion.

I would thus vote for going forward now.

Cheers,
Baptiste

> Le 19 oct. 2021 à 11:01, Markus Demleitner <msdemlei at ari.uni-heidelberg.de> a écrit :
> 
> Dear Colleagues,
> 
> As noted in
> http://mail.ivoa.net/pipermail/semantics/2021-September/002846.html,
> VEP-007 (#documentation in datalink/core) exposed a problem with
> #documentation's definition.  It currently explicitly requires
> human-readability:
> 
>  Extra information on the item in human-readable text form, ranging
>  from processing logs to weather reports to technical documents on
>  instruments to related publications.
> 
> It turns out that "human-readability" is hard to define.  Is a FITS
> header human-readable?  Or a PDF label?  Or... a PDF?  Frankly, if
> all I have is a text editor, I'll go for the FITS header rather than
> the PDF.  On the other hand, there are few things that are
> human-unreadable if given the appropriate tools.
> 
> Hence, I'd say the "human-readable" ought to be taken out of the
> definition as not reproducible between indviduals, and also not
> useful in practice; I'm quite sure that I'd be cross with the
> publishers if they had a detached PDF label but wouldn't show it
> under #documentation.
> 
> I have amended VEP-007 accordingly
> (https://volute.g-vo.org/svn/trunk/projects/semantics/veps/VEP-007.txt)
> to change documentation's definition to:
> 
>  Structured or unstructured metadata helping to understand, 
>  interpret, or work with #this.  Such information can range from
>  processing logs to weather reports to technical documents on
>  instruments to related publications.
> 
> The part on "structured or unstructured" caters (or so I hope)
> towards François' concerns from
> http://mail.ivoa.net/pipermail/semantics/2021-September/002844.html,
> where once we have further terms talking about explicitly
> computer-readable ("structured") metadata and we find we have to
> enable common behaviour among them, the introduction of, say,
> *#structured-metadata will not come as a surprise.
> 
> Can everyone live with this?  Do you perhaps have suggestions for how
> to make this more precise?
> 
> Thanks,
> 
>             Markus