I don't think integrating this feature into GLPI Core is a good idea: or it should also work with any objects (eg. with users, to map their belongings).

On the other side, having a v0.90 compliant archires plugin would be great :) (but it might not be the right place to ask for this).

What are projects?

-1

Disclaimer: I like the fact that additional features for GLPI are possible via plugins: this makes the GLPI core less bloated.

Why integrate a plugin into the GLPI Core? If the plugin does the job there is no need to make it part of GLPI.
If you think the plugin's features should be integrated into GLPI's core: please explain why it should be, which advantages it would brings, why the plugin-solution is not working (or badly).

I like the fact that additional features for GLPI are possible via plugins: this makes the GLPI core less bloated.

But I think adding "Generic object" and "Custom fields" features in GLPI core could improve integration.

This integration would require plugins to support theses objects: this is good for user experience but require more work for plugin's developers.

Please take this into account when choosing to implement this idea.

Creating an "external" KB entry (that points to a given URL) is a good and easy-to-implement solution.

A "Location" already has latitude/longitude coordinates.

But I am also looking for a way to precisely locate locations on a self-created map. It would be useful for finding printers in a large open-space or indicating (to a newcomer) where a specific room is: theses examples won't work with "latitude/longitude coordinates", but "pixels coordinates" on a uploaded JPEG/PNG/SVG image would.

If it means adding OAuth support: I'm "+1" on this one.

Although entities are generally not changed often, it would be useful for some setups managing many entities.

This would require plugins such as FusionInventory to be adapted.

But it would not work on Linux (there is no such information available).

I thinks such ideas should go to plugin's GitHub.