Pierre Krieger

To clarify, I'm primarily looking for answers to these questions because I'm implementing child tries in smoldot. Adding the answers to the spec would be a bonus. Having informal answers...

cc @cheme @arkpar (not sure who to ping)

Thanks for the answer! Another question, in reaction to this: is it correct that `ext_default_child_storage_clear_prefix(child_trie, [])` and `ext_default_child_storage_storage_kill(child_trie)` do exactly the same thing, then? Or is there a difference in...

> (unless ext_storage_root is not call which does not make sense for a runtime) It would still be nice to clarify what happens if `ext_storage_root` isn't called, for the sake...

> Alternatively we could add additional default feature which will gate functionality not related to raw image representation. I would prefer this solution. Extracting crates and types into an external...

It also needs a bit of reorganization. Just moving the sections doesn't really produce a result that makes sense.

I mean reorganize the networking chapter. Don't you think it's very confusing? It is very clearly visible that each sub-section was written separately without any coherence by different authors who...

The trait `GlObject` is meant to provide you with IDs for objects. However since framebuffer objects can be invalidated at any time if the user calls `rebuild_context`, the trait can't...

> is there a hacky way to still get the ID for SimpleFramebuffer? No, but it's something that could be added. > And is there a way to save the...

> In the meantime, can I guess its ID, in other words, are framebuffer IDs generated deterministically? Not at all. > In other words, why wouldn't it work if I...