Even Rouault
It has been ages since the last time I blogged. Here it is "Those concepts in the geospatial field that have cost us millions of $$$":
https://erouault.blogspot.com/2024/09/those-concepts-in-geospatial-field-that.html
Those concepts in the geospatial field that have cost us millions of $$$

Every domain has its baggage of concepts, which at first sight don't appear to be that terrible, but which are in practice. Let's start with...

Sep 26, 2024 at 8:30pmWeb
Show threadHoward ButlerSep 26, 2024
@EvenRouault I'm just here for the rant-mode Even...
Show threadEven RouaultSep 26, 2024
@hobu Turn on the subtitles. The rant is never far away ;-)
Show threadThomas Szczurek-GayantSep 26, 2024
@EvenRouault you're burning out recently. Please dont 😅 take some time for yourself
Show threadGiorgio ComaiSep 26, 2024
@EvenRouault "the only solution I see to solve the issue it is to ask all population to relocate on the line of longitude=latitude, and exclude any mapping outside of it." 🤣🤣🤣
Show threadNyall DawsonSep 26, 2024
@EvenRouault empty geometries are insane 🤬
Show threadSean GilliesSep 27, 2024

@nyalld @EvenRouault And *typed* empties are doubly insane.

It's like distinguishing between 32 bit and 64 bit NaN.

Show threadEven RouaultSep 27, 2024
@sgillies @nyalld I've always dreamt about being able to use the degrees of freedom left by all those possible bits in a NaN representation to actually store useful information in them. That's a bit like quantum physics with Vacuum energy or this old saying "[Redacted entity] created not-a-number, saw this was (not) good, and from it/them created numbers"
Show threadEdzer PebesmaSep 28, 2024
@EvenRouault @sgillies @nyalld R's NA for floating point numbers (distinct from NaN) actually does that; IEEE 754 however doesn't mandate its propagation (but current processors seem to do so); see for a longer discussion the NA section in https://blog.r-project.org/2020/11/02/will-r-work-on-apple-silicon/
Will R Work on Apple Silicon? - The R Blog

Show threadHDSep 27, 2024
@EvenRouault i wonder how much fossil fuel is burned to store all those 15 decimal precision geojsons and other file formats that come like that as default...
Show threadEven RouaultSep 27, 2024
@hareldan There's actually a GDAL RFC for that (https://gdal.org/en/latest/development/rfc/rfc99_geometry_coordinate_precision.html). It just needs to be used...
RFC 99: Geometry coordinate precision — GDAL documentation

Show threadMatija NalisOct 4, 2024
@hareldan
re: "store all those 15 decimal precision geojsons" obXKCD: https://xkcd.com/2170/
@EvenRouault
Coordinate Precision

xkcd
Show threadRaúl Nanclares 🍜Sep 27, 2024

@EvenRouault
This is the stuff nightmares are made off... geo "standards".

There's also R sf dataframes that allow geometries with different CRS yay! We really need geometry collections with multiple CRS, because that's what we usually use /s

Show threadAndré StumpfSep 27, 2024
@EvenRouault This is gold, thanks for sharing!
Show threadAndré StumpfSep 27, 2024
@EvenRouault Can we add GPKG's coordinate precision which allows to map the position of an atom to the rant list? https://github.com/opengeospatial/geopackage/issues/365
Support for TWKB or other compact geometry encoding · Issue #365 · opengeospatial/geopackage

The vector data in geopackage is today encoded as WKB, a well-defined and well known way of formatting spatial geometries. However, since all coordinates are encoded as two 64bit double values, the...

GitHub
Show threadEven RouaultSep 27, 2024
@andrestumpf There are people actually requiring that level of precision, so we shouldn't exclude them. Well, maybe not at the angstrom level, but tens of millimeters is something geodesists routinely want to achieve. At least some I've interacted with... That said the main issue of TWKB or friends in GeoPackage is that would make it a non-interoperable extension...
Show threadJavier Jimenez ShawSep 28, 2024

@EvenRouault

"the only solution I see to solve the issue it is to ask all population to relocate on the line of longitude=latitude, and exclude any mapping outside of it."

Genius.

Show threadEven RouaultSep 28, 2024
@jjimenezshaw The initial version of it was to relocate everybody at Null Island, but turns out to be impractical given that the buoy has disappeared, and due to the lack of clear datum there, only WGS84 coordinates could be realistically used, so we would need to communicate an epoch, and that would confuse people. Yes, Null Island (lambda, phi)=(0,0), but when exactly ?!?
Show threadJavier Jimenez ShawSep 28, 2024
@EvenRouault you can see that as a feature, not a bug.
The uncertainty of WGS84 in space... and time, could provide about 10 square metres at 0,0 That should be enough for 8 billion people. It would increase with time... to fit the new population.