Not because it would be actually possible to hide it but because code has/will become so complex that it wouldnt really make sense anymore. Take, for example, sites like Facebook or Gmail: even the developers tools barely help understanding them, let alone the raw "view source".
This (human understanding) will not get better with an intermediate language like WebAssembly (which should not be considered an obstacle though). And this does not take into account a possibly even more drastic change in technology and the web (or whatever the web would be called then).
It's already been possible to 'hide source' for years on the web, either in the form of compiled swfs for Flash files or, these days runtime code or output spat from server side complexity.
Without a fundamental re-write or philosophical change on the internet though, there's little to fear that 'the majority' of sites would go this route in any imaginable near to mid-term future.
Conversely, anyone paranoid or arrogant enough to do so could do, shooting themselves in the foot a little, now.
Your main concern should be, i'd've thought, messy and compacted code that uses a mashup of irrelevant and bloated frameworks, because that way lies bad habits and obfuscated education...
HTML as is though? i'd not concern myself.
In the developing world, mobile as the regular computing environment is a fact of life.
For most of the big tech companies (Facebook, Uber, Google's search arm, etc.) mobile is probably where they encounter their users the most.