A similar option is to boot a kernel using the built-in EFI stub (CONFIG_EFI_STUB=y), embed the initramfs if you use one (CONFIG_INITRAMFS_SOURCE), and set the default command line with CONFIG_CMDLINE if configuring it via UEFI is too unpleasant. I boot like this.
Advantage: less complex than adding another layer (and dependency) with the gummiboot stub.
Disadvantage: relinking the kernel is more expensive than rebundling with objcopy, if you change initramfs with the same kernel often.
Disadvantage: it's not an option if you use a distribution kernel binary instead of building your own.
I think MathML might not have failed so hard if it weren't so ludicrously verbose and time-consuming to type compared to TeX notation.
XML apologists generally propose special editors to make the experience less unpleasant, but that's not what working users of mathematics want.
TeX notation provides a way for us to type formulae in plain text form, anywhere that ascii is accepted, as fast as we can type normal text. That's the bare minimum required for any representation to be relevant to mathematicians and physicists.
I don't know what a more 'web flavoured' notation satisfying this constraint would look like. (Are web standards people even capable of brevity?)
Advantage: less complex than adding another layer (and dependency) with the gummiboot stub.
Disadvantage: relinking the kernel is more expensive than rebundling with objcopy, if you change initramfs with the same kernel often.
Disadvantage: it's not an option if you use a distribution kernel binary instead of building your own.