Re: The Multiple types of coverage requirement

Here is a classification by grid complexity, hope it helps somehow:

- just array, mapping is 1:1
    -> GridCoverage (GML 3.2.1)
- linear mapping from n-D array to n-D grid, g = a*x+b for a,b in R^n
    -> RectifiedGridCoverage
- linear mapping with g = A*x+b for A in R^n*n
    -> ReferenceableGridCoverage, byVector
- linear mapping from n-D array (x_n) to m-D grid (g_m), m>n: g_m = A * x_n + b
for A in R^n*m
    ->ReferenceableGridCoverage, byArray
- nonaffine transformations, ex: Sensor geometries

(anybody disagreeing?)

One could classify along dimension axes (horizontal, vertical, time, ...) but
the above one I believe is most helpful for the purpose on hand, and it also
very much determines implementation complexity and dataset sizes (in particular:
byArray approx. doubles data set size).

-Peter



On 05/31/15 09:35, Kerry.Taylor@csiro.au wrote:
> I am quite ok with the UCR expressed as it is, but I note that our use cases
> do not actually require very many of these multiple types and I suggest that
> we should look more closely at * which* of the multiple types we really need
> as we go progress. I would like to suggest that we make a note of this in the
> UCR, attached to this requirement, to prevent  the "multiple" being
> interpreted as"all that have ever been thought of" . 
>
> note- i am prepared to back down on the request for the note, as  I think
> argued almost contradictorily in the case of "multiple" applied to
> "multilingual" in last week's meeting. Although in that language case I saw no
> harm other than realism in attacking every conceivable language, but in the
> coverage case I think there is a risk of harm in confusion if we take  on too
> many, which is worse.
>
> Kerry
>
>
>
> On 29 May 2015, at 11:47 pm, "Peter Baumann" <p.baumann@jacobs-university.de
> <mailto:p.baumann@jacobs-university.de>> wrote:
>
>> Frans-
>>
>> here a slate (not comprehensive, but likely covering all considered by W3C
>> currently):
>> - gridded coverages:
>>     - by dimension: 1-D through 4-D (climate people also consider 5-D)
>>     - by grid type:
>>         - regular grid (equidistant spacing ("resolution"), such as ortho
>> imagery)
>>         - irregular grids (grid lines have individual spacing per axis, such
>> as timeseries often have)
>>         - warped grids (grid points sit anywhere in space, but still
>> topologically isomorphic to a grid)
>>         - sensor grids (geo position of grid points determined by sensor
>> model, usually some involved non-linear algorithm)
>> - non-gridded coverages:
>>     - point clouds
>>     - (rest likely not of interest here)
>>
>> cheers,
>> Peter
>>
>>
>> On 05/29/15 14:36, Frans Knibbe wrote:
>>> Hello Alejandro,
>>>
>>> I am looking at the Multiple types of coverage
>>> <http://w3c.github.io/sdw/UseCases/SDWUseCasesAndRequirements.html#MultipleTypesOfCoverage>
>>> requirement now: "It should be possible to represent many different types of
>>> coverage."
>>>
>>> Does this mean some kind of standard classification of coverage types is
>>> required, so the coverage type can be indicated in the metadata for example?
>>>
>>> Or does this mean that there should be standard encodings for different
>>> coverage types? 
>>>
>>> Greetings,
>>> Frans
>>>
>>> -- 
>>> Frans Knibbe
>>> Geodan
>>> President Kennedylaan 1
>>> 1079 MB Amsterdam (NL)
>>>
>>> T +31 (0)20 - 5711 347
>>> E frans.knibbe@geodan.nl <mailto:frans.knibbe@geodan.nl>
>>> www.geodan.nl <http://www.geodan.nl>
>>> disclaimer <http://www.geodan.nl/disclaimer>
>>>
>>
>> -- 
>> Dr. Peter Baumann
>>  - Professor of Computer Science, Jacobs University Bremen
>>    www.faculty.jacobs-university.de/pbaumann
>>    mail: p.baumann@jacobs-university.de
>>    tel: +49-421-200-3178, fax: +49-421-200-493178
>>  - Executive Director, rasdaman GmbH Bremen (HRB 26793)
>>    www.rasdaman.com, mail: baumann@rasdaman.com
>>    tel: 0800-rasdaman, fax: 0800-rasdafax, mobile: +49-173-5837882
>> "Si forte in alienas manus oberraverit hec peregrina epistola incertis ventis dimissa, sed Deo commendata, precamur ut ei reddatur cui soli destinata, nec preripiat quisquam non sibi parata." (mail disclaimer, AD 1083)
>>
>>

-- 
Dr. Peter Baumann
 - Professor of Computer Science, Jacobs University Bremen
   www.faculty.jacobs-university.de/pbaumann
   mail: p.baumann@jacobs-university.de
   tel: +49-421-200-3178, fax: +49-421-200-493178
 - Executive Director, rasdaman GmbH Bremen (HRB 26793)
   www.rasdaman.com, mail: baumann@rasdaman.com
   tel: 0800-rasdaman, fax: 0800-rasdafax, mobile: +49-173-5837882
"Si forte in alienas manus oberraverit hec peregrina epistola incertis ventis dimissa, sed Deo commendata, precamur ut ei reddatur cui soli destinata, nec preripiat quisquam non sibi parata." (mail disclaimer, AD 1083)