I am exploring a project that essentially consists of placing custom, touchable (or clickable) icons over
either a Google-style geo-map
or a map (e.g. of a building) that was scanned in.
The current application is an "iOS-only mess," and there is strong interest in being able to deploy this app ... as "just one app" ... to native-iOS, native-Android, and HTML5. I think it's also important that this app not simply be "JavaScript + HTML + CSS," but a genuinely native application. For instance, multi-finger touch to zoom in and out is a requirement.
The fundamental business consideration for evaluating this approach is to reduce the number of parallel code-bases that must be maintained today "to do the same thing," and of course to bring haXe's strong-typing to JS-target development. I'm also quite interested in "future-proofing" the approach that is taken.
I've been away from haXe for a couple of years but have had good experiences with it in the past. I'm therefore interested in hearing what others would consider to be "the state-of-the-art-today approach" for dealing with such a requirement using haXe – and if, in fact, haXe would be the tool-of-choice. (Hey, I can ask ... and if not, what would you choose today and why?)
The project would commence almost immediately, therefore "real soon now" won't cut the mustard. Any existing examples with source-code that I could refer to [legally] on the net would be a big plus.