Post #2579238

2024-10-11 01:45 UTC

this is a half-formed opinion! but: older tech would force migrations of data into newer formats, and damn anyone who didn't. I genuinely don't have a good understanding of how to evaluate the current status quo, which doesn't force migrations but does force new tools to support old formats

Replies (2)

@MikeMahoney218@fosstodon.org 2024-10-11 01:54

to some extent: I wonder if tools would be better if we were still in a place where we could enforce data standards on providers maybe we can't do that anymore! but that definitely feels like a different era than past tools, which could say "you're doing my data standard or you're left behind"

Open ##2579240
@mdsumner@rstats.me 2024-10-11 03:53

@MikeMahoney218@fosstodon.org netcdf dominated for way too long, the biggest issue I see is that it's a lowest common denominator format, simple enough for matlab but not smart enough when smarts is really needed and it put a hex on xarray and Zarr which adopted the same coordinate model, but they're fixing xarray and it looks really promising! I think they're right about Zarr, but still it's not a good choice when GeoTIFF *suffices*

Open ##2579241