The next GMT Community Zoom Meeting will be on Thursday, December 16, 2021 at UTC 1900! All are welcome! Here’s a preliminary agenda for the meeting - feel free to suggest more agenda items by commenting in this thread.
Preliminary agenda:
Status update on GMT development
DCW updates (discuss regions and collections of countries).
Some comments about the dcw collections to talk in the meeting:
Introduction
In dcw collections there are two types of records:
List of countries (list of isocodes)
Bounding box regions (W/E/S/N values)
General Question/doubts
Should be both types of records be in the same file (dcw-collections.txt)? Or should be in separate files?
The file is distributed with dcw? What would happened is someone doesn’t install it? It makes sense that can’t access the list of countries, but what about the regions?
The name of file can be a bit confusing for some users because its includes regions which are not related with dcw.
Questions about the contents of DCW-Collections:
So far, I included UN collections of countries (M49) and seas/oceans regions (IHO).
Should be a limit in the amount of records? Which one?
Is there any other official list that we want to include?
I think we don’t have to stick to official regions or collections because we could skip some widely use name that would help the users. So I think we could include collections of countries (e.g. Baltic countries, Balcans, Scandinavia, Less Antilles) as long as there is a Wikipedia page explaining the term.
In the same way, we should include name of geographic regions that can’t be properly describe as a list of countries. This would include continents (like Africa or Europe). Maybe also we could include the major lakes (e.g. Caspian Sea) and other physiographic regions (like peninsulas, islands?, deserts, mountain chain, basins, tectonic plates?, etc). It could be difficult to define the limits of this regions. It would be great to find some shp (or similar) file that have already done it.
@Joaquim in the meeting you mentioned that you search algorithms to define the countries from the lines of the international borders. Could you share what you have found? I would like to see them.
Federico, I did search for such programs but found nothing. And now I’m not so sure that it would a good idea. For one side it would cut the file size by a factor of < 2 but on the other hand to do plots it would imply a large memory use (and time) required when reconstructing the polygons from the pieces.
And, what about if we create the polygons first? So, for example we could create a new DCW version from the GSHHG. Of course, we won’t be saving the space but both data sets will be coherent between them. Would this be a good idea?
The memory wouldn’t probably be a big issue (not that much) but it would be slower. How much slower, I don’t know. Only with testing but for that we would need to have code for it and this not trivial.
Having a single data origin for DCW and GSHHG would be ideal (but the acronyms are already saying the data sources are different) but we wont want to get into the trouble of recreating DCW after the GSHHG polygons. Specially when we want to replace the current GSHHG.