should you use localization assets instead of making your own if you are just swapping strings?

There a few localization assets that seems pretty complex, and there's a new built-in localization tool that is being developed atm. My question is, should I even bother using any of that if all I need are string replacements for buttons and such? Do they offer any benefit for someone who just needs strings replaced in textmeshpro fields?

 

There's something to say for the simplicity of writing your own purpose-built code. But make it modular so you can easily swap it out it if you find later that you need to do more and want to use an off the shelf solution. For example, maybe you need to replace strings in more than just TMP fields, or you need to localize images/icons, or you need to use different fonts for different languages, or your translators want to import and export Excel files and you don't want to deal with writing an importer.

 

Very interesting question. I did my own localisation which felt pretty easy from start but when I was done it ended up with a fair bit of complexity. Any preferable localisation solution to be recommended?

 

Unity is developing a built-in Localization package. I don't know the latest status of it.

Inter-Illusion's i2 Localization is quite popular.

 

"string replacements for buttons and such" can end up more complex than it sounds, since these strings will be different lengths. You might end up actually needing different font sizes, different fonts themselves for uncommon characters or character accents, a resizing of buttons, etc. I'd assume that some or all localization packages implement this kind of thing. If you go the route of making all your text fields large enough so all localizations easily fit, they probably won't look right in any language, since you'll end up with oversized text fields and buttons for the text displayed in most languages.