You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Currently the way that map-tiles are rendered is that they are either retrieved from an online API, or obtained from a separate local-hosted tile server, which ends up being a good amount of work involved . We can look into how these tiles could be instead directly hosted on GCOM instead.
Advantages:
Less in-field setup work required, potentially less debugging in the case that the separate tile-hosting source doesn't work.
One less process to run on the localhost machine, less complexity, greater chance of being able to bundle FE and BE together in a docker container or something like that.
Drawbacks:
Could be potentially very complex to setup. These map tiles are served in very special formats through layers of JSON metadata.
Is the benefit-cost ratio worth it? Currently I'm thinking that we can spend a bit of time looking into this self-hosting solution for a couple weeks, but time is definitely better spent looking at something else if there is no good progress after e.g. a month
Alternatives:
Maybe we could bundle GCOM BE, FE and the tile-hosting server all into one docker image for ease of setup.
The text was updated successfully, but these errors were encountered:
Currently the way that map-tiles are rendered is that they are either retrieved from an online API, or obtained from a separate local-hosted tile server, which ends up being a good amount of work involved . We can look into how these tiles could be instead directly hosted on GCOM instead.
Advantages:
Drawbacks:
Alternatives:
The text was updated successfully, but these errors were encountered: