What Do LOVE2D Developers Actually Want?

[sphyrth]

A convenient official way to release in html5 and mobile devices. What we currently have are either finicky or not yet officially supported by the main page.

[dusoft]

A convenient official way to release in html5 and mobile devices. What we currently have are either finicky or not yet officially supported by the main page.

That would require official HTML5 support that AFAIK is not on the table.

[BrotSagtMist]

Me having no idea how html5 is different than previous versions:
Is that even technically possible to have lua run on html5 without the help of js?

[ivan]

If you want an "official" love2d port that runs inside the browser you have to do the legwork.
Everybody wants super advanced features, but love2d is missing the basic stuff like the ability to lock files, loading 3rd party modules, better security and sandboxing, big numbers, SSL support and decent distribution.
It is easy to scrap something together, but making the basic features work consistently between platforms is a lot of work. That is one of the reasons why the Lua language itself doesn't include these features.

[MrFariator]

better security and sandboxing

Something like that doesn't feel like something love2d itself can solve PC-side, because any rogue agent could always develop their own program (without the involvement of love2d, or by removing any sandboxing features from love2d) to target whatever save files or data they choose to focus on. To prevent something like that I imagine OS vendors would have to get involved. That, and how do you actually deal with conflicting identities? Something like Java's com.company.project convention?

I can see the appeal of SSL support, though.