tomay / rebioma

Automatically exported from code.google.com/p/rebioma
0 stars 0 forks source link

Reducing the number of required headers #401

Closed GoogleCodeExporter closed 9 years ago

GoogleCodeExporter commented 9 years ago
From practises I propose to remove :
- "geodeticdatum" is by default "WGS84"
- "nomenclaturalcode" : they don't worry about it, by default "ICZN" or "ICBN"
- "YearCollected" and "Year" must be accepted
- "coordinateuncertainityinmeters" is not yet used in the modelling and by 
default its "30" if it's from GPS

Theses generate a lot of the changes if accepted.

Finally, all we need are, 
Genus,SpecificEpithet,DecimalLongitude,DecimalLatitude,Year  and "recordedBy" 
and "locality" for the georeferencing validation

What do you think ?

Original issue reported on code.google.com by ajmrakot...@gmail.com on 3 Aug 2011 at 11:32

GoogleCodeExporter commented 9 years ago
I think it is a very bad idea to remove the geodeticdatum as a required field 
if the coordinates are required. You really shouldn't assume anything about 
coordinates if you don't know this information. If you do know it, it should be 
filled in as WGS84. If you don't, it should be filled in as unknown. These are 
very different from each other, because you are otherwise making a 
falsification that is later undetectable - indistinguishable from coordinates 
where the datum is actually known. 

NomenclaturalCode is there as a required field to make sure that you an 
distinguish taxa that might have the same scientificname, but in two different 
codes. This may not even be possible in Madagascar, and it certainly isn't in 
the data so far. So, though it probably would be fine to not require it in your 
case, but it would not be fine globally.

A more general question though, why remove the requirements, which are in place 
to assure data quality and usability, especially if it creates more work to 
remove them? That seems antithetical to REBIOMA's main distinguishing 
characteristic - data quality.

Original comment by gtuco.bt...@gmail.com on 3 Aug 2011 at 2:48

GoogleCodeExporter commented 9 years ago
Thanks John ! Very good remarks, so may be add some fields to raise the quality 
because
users don't know that they can fill each field by NULL or "unknown". For 
example, apart from geodaticdatum, as you are expert on georeferencing, the way 
to be sure of the coordinates is to compare them from georeferencing method 
based on locality description. My idea behind this was to use the GBIF 
spreadsheet templates as a template too for REBIOMA Data Portal. Because this 
will be used as a data management online before the MadBIF IPT. 

Original comment by ajmrakot...@gmail.com on 4 Aug 2011 at 11:12

GoogleCodeExporter commented 9 years ago

Original comment by ajmrakot...@gmail.com on 21 Oct 2011 at 8:29