[GRASS-stats] spgrass7 support for strds?
Roger Bivand
Roger.Bivand at nhh.no
Thu Feb 19 12:26:06 PST 2015
I've committed "rgrass7" for the next package, leaving the others as they
are. I'll try to submit by the end of the month.
Roger
On Thu, 12 Feb 2015, Rainer M Krug wrote:
> Just for clarification - the rgrass solution would mean:
>
> GRASS for GRASS 5
> spgrass6 for GRASS 6
> rgrass for GRASS 7
>
> If this is correct, then it is very confusing.
>
> To re-iterate:
>
> I like the rgrass, but
>
> rgrass6 and rgrass7
>
> would make things clearer.
>
> Rainer
>
>
> Edzer Pebesma <edzer.pebesma at uni-muenster.de> writes:
>
>> I'm in favour of rgrass.
>>
>> On 02/12/2015 12:13 PM, Ahmadou Dicko wrote:
>>> I will also cast my vote in favor of 'rgrass' because of the consistency
>>> with other library and gis linked to R (RSAGA, rgeos, rgdal, etc.).
>>> We can just then make sure that it works with both version of GRASS (6
>>> and 7).
>>>
>>> I also started a project (idea) a long time ago with Rainer to have
>>> something similar to pygrass[1] in R through Rcpp
>>> (https://github.com/RGRASS/grassRLink) in order to access the low level
>>> GRASS API.
>>>
>>> I didn't have time to finish it but I will have more time this summer
>>> and hope to work on it again.
>>> I think it can be a good idea of GSOC for either GRASS or R.
>>>
>>>
>>>
>>>
>>>
>>>
>>> [1]: http://grasswiki.osgeo.org/wiki/Python/pygrass
>>>
>>>
>>>
>>>
>>>
>>>
>>> On Thu, Feb 12, 2015 at 10:51 AM, Rainer M Krug <Rainer at krugs.de
>>> <mailto:Rainer at krugs.de>> wrote:
>>>
>>> Roger Bivand <Roger.Bivand at nhh.no <mailto:Roger.Bivand at nhh.no>> writes:
>>>
>>> > Question further down ...
>>> >
>>> > On Wed, 11 Feb 2015, Edzer Pebesma wrote:
>>> >
>>> >>
>>> >>
>>> >> On 02/11/2015 03:42 PM, Rainer M Krug wrote:
>>> >>> Edzer Pebesma <edzer.pebesma at uni-muenster.de
>>> <mailto:edzer.pebesma at uni-muenster.de>> writes:
>>> >>>
>>> >>>> On 02/11/2015 02:52 PM, Rainer M Krug wrote:
>>> >>>>> Roger Bivand <Roger.Bivand at nhh.no
>>> <mailto:Roger.Bivand at nhh.no>> writes:
>>> >>>>>
>>> >>>>>> Since I'm uncertain whether the maintainer of spacetime is on
>>> this
>>> >>>>>> list, I'm CC-ing him here. It might make sense to discuss this -
>>> >>>>>> perhaps spgrass7 in the R-forge spgrass project is a logical
>>> place to
>>> >>>>>> support these functions?
>>> >>>>>
>>> >>>>> It is a brilliant idea to support spatio-temporal data in the
>>> spgrass7
>>> >>>>> framework.
>>> >>>>>
>>> >>>>> I agree that the functions should be in spgrass7 and that they
>>> should
>>> >>>>> have the same interface structure as the other
>>> read/write[RAST/VECT]
>>> >>>>> functions.
>>> >>>>>
>>> >>>>> I never used the package spacetime or spatio-temporal datasets
>>> in GRASS,
>>> >>>>> but maybe these functions could be re-used (or at least parts
>>> of it)?
>>> >>>>
>>> >>>> yes; and use the TGRASS (Cc: Soeren, the author), framework. Also,
>>> >>>> raster stacks with time z should be of interest (Cc: Robert).
>>> >>>
>>> >>> Please forgive my ignorance - but is there actually an accepted (as
>>> >>> SpatialGridDataFrame et al are accepted) object framework for these
>>> >>> spatio-temporal datasets in R? It sounds as if there is not at
>>> the moment?
>>> >>
>>> >> When is something accepted? raster and spacetime have both been
>>> on CRAN
>>> >> for some years, and have both been considered useful by people
>>> >> independent from the developers. There is a SpatioTemporal task view
>>> >> that tries to list everything possibly relevant.
>>> >>
>>> >>>
>>> >>>>
>>> >>>>>
>>> >>>>> But this should not hold back the release of spgrass7 in it's
>>> current
>>> >>>>> form.
>>> >>>>
>>> >>>> I suggest to rename this package to "grass", supporting the current
>>> >>>> grass release.
>>> >>>>
>>> >>>> Grass took 7 years to move 6 to 7; I discussed this with Markus
>>> last
>>> >>>> night, they might as well give up this crazy major version
>>> release cycle
>>> >>>> idea, with all kind of restrictions they impose on themselves.
>>> >>>
>>> >>> This might cause confusion. And as long as GRASS has major
>>> versions, I
>>> >>> think we should stick with this naming scheme.
>>> >>
>>> >> We don't do this with rgdal, rgeos, RSAGA, RArcInfo, and so on.
>>> But many
>>> >> of these have an "r" in front. This seems obsolete as they are R
>>> >> packages, but may be helpful in sentences like "We used grass and the
>>> >> r-package rgrass for carrying out multidimensional scaling on the
>>> MODIS
>>> >> imagery ...", which, without the leading r, would have the word
>>> "grass"
>>> >> twice in it referring to different things.
>>> >>
>>> >> spgrass has sp in it for historic reasons - we thought about
>>> prefixing
>>> >> everything with sp in the early sp days. Now sp is default, and one
>>> >> could think about a better name.
>>> >
>>> > OK, I think we keep (archived) "GRASS" for the interface to GRASS 5,
>>> > "spgrass6" for GRASS 6. As of now, we've called the development
>>> > version "spgrass7", but arguably "rgrass" could be preferred. As GRASS
>>> > 7 is due on Sunday, could I ask for a quick show of hands on
>>> > "spgrass7" or "rgrass" as the name for the package to be submitted to
>>> > CRAN (tomorrow)? We can continue to consider moving spacetime code to
>>> > this package, but we should have the new package on CRAN matching the
>>> > release of GRASS 7.
>>>
>>> The r in front makes sense - but if one looks at the changes in the
>>> interface between GRASS 6 and GRASS 7 I would keep the 7 - so I would
>>> suggest
>>>
>>> rgrass7
>>>
>>> and possibly also release the (modified) spgrass6 as rgrass6, and leave
>>> spgrass6 there for compatibility, and give a warning upon load that the
>>> package has been superseded by rgrass6.
>>>
>>> This also opens the possibility of releasing a package called
>>>
>>> rgrass
>>>
>>> which then, depending on the GRASS version loaded / requested loads the
>>> corresponding rgrassX package.
>>>
>>> Rainer
>>>
>>>
>>>
>>> >
>>> > Roger
>>> >
>>> >>
>>> >>>
>>> >>> Cheers,
>>> >>>
>>> >>> Rainer
>>> >>>
>>> >>>>
>>> >>>>>
>>> >>>>> Cheers,
>>> >>>>>
>>> >>>>> Rainer
>>> >>>>>
>>> >>>>>>
>>> >>>>>> Roger
>>> >>>>>>
>>> >>>>>>
>>> >>>>>> On Wed, 11 Feb 2015, Veronica Andreo wrote:
>>> >>>>>>
>>> >>>>>>> Hi
>>> >>>>>>>
>>> >>>>>>> Since you are discussing the new spgrass for GRASS 7 (and since GRASS 7
>>> >>>>>>> includes these great temporal modules), I was wondering if there's any
>>> >>>>>>> chance to include support for reading/writing spatio-temporal data sets.
>>> >>>>>>>
>>> >>>>>>> At present, there are two functions in spacetime package (read.tgrass and
>>> >>>>>>> write.tgrass) that allows reading and writing strds, but you need to export
>>> >>>>>>> it from GRASS first, do the R processing you need, and then, import it back
>>> >>>>>>> into GRASS. To have something a bit more straightforward would be awesome!
>>> >>>>>>>
>>> >>>>>>> I don't know how complicated this could be, that's why I dare to ask :)
>>> >>>>>>>
>>> >>>>>>> Keep the great job you are doing!!! And thanks a lot!!!
>>> >>>>>>>
>>> >>>>>>> Cheers,
>>> >>>>>>> Vero
>>> >>>>>>>
>>> >>>>>
>>> >>>>>
>>> >>>>>
>>> >>>>> _______________________________________________
>>> >>>>> grass-stats mailing list
>>> >>>>> grass-stats at lists.osgeo.org <mailto:grass-stats at lists.osgeo.org>
>>> >>>>> http://lists.osgeo.org/mailman/listinfo/grass-stats
>>> >>>>>
>>> >>>
>>> >>
>>> >>
>>>
>>> --
>>> Rainer M. Krug, PhD (Conservation Ecology, SUN), MSc (Conservation
>>> Biology, UCT), Dipl. Phys. (Germany)
>>>
>>> Centre of Excellence for Invasion Biology
>>> Stellenbosch University
>>> South Africa
>>>
>>> Tel : +33 - (0)9 53 10 27 44
>>> <tel:%2B33%20-%20%280%299%2053%2010%2027%2044>
>>> Cell: +33 - (0)6 85 62 59 98
>>> <tel:%2B33%20-%20%280%296%2085%2062%2059%2098>
>>> Fax : +33 - (0)9 58 10 27 44
>>> <tel:%2B33%20-%20%280%299%2058%2010%2027%2044>
>>>
>>> Fax (D): +49 - (0)3 21 21 25 22 44
>>> <tel:%2B49%20-%20%280%293%2021%2021%2025%2022%2044>
>>>
>>> email: Rainer at krugs.de <mailto:Rainer at krugs.de>
>>>
>>> Skype: RMkrug
>>>
>>> PGP: 0x0F52F982
>>>
>>> _______________________________________________
>>> grass-stats mailing list
>>> grass-stats at lists.osgeo.org <mailto:grass-stats at lists.osgeo.org>
>>> http://lists.osgeo.org/mailman/listinfo/grass-stats
>>>
>>>
>>>
>>>
>>> --
>>> Ahmadou H. DICKO
>>> statistician economist (Ingénieur Statisticien Économiste)
>>> PhD candidate in Climate change economics
>>> Faculty of economics and managment - Cheikh Anta Diop University
>>> West African Science Service Center on Climate Change and Adaptated Land
>>> Use (WASCAL)
>>> Center for Development Research (ZEF) - University of Bonn
>>> email : ahmadou.dicko at ucad.edu.sn <mailto:ahmadou.dicko at ucad.edu.sn>
>>> twitter : @dickoah
>>> github : github/dickoa <https://github.com/dickoa>
>>> tel : +221 33 827 55 16
>>> portable: +221 77 123 81 69
>
>
--
Roger Bivand
Department of Economics, Norwegian School of Economics,
Helleveien 30, N-5045 Bergen, Norway.
voice: +47 55 95 93 55; fax +47 55 95 91 00
e-mail: Roger.Bivand at nhh.no
More information about the grass-stats
mailing list