I don't think the "next big thing" here will be HTML5, though. Unity is _ridiculously_ easy and has led to large swaths of indie games on Steam - people don't quite understand just how easy it is to make games in Unity these days. You can pass without understanding coding at all to a large extent, and reminds me a _lot_ of working in Flash.
If you don't believe me, look up Sam Hogan on YouTube - or search YouTube for "unity game in 24h". I'm not saying these are all good - but it reminds me a _lot_ of how Flash enabled rapid development of game ideas.
The reality is that making good games is extremely hard, and has always been.
Unity is not the first game framework, there have been many along the years. Flash being one of them (while not even designed for that originally!).
Some of them are easier, some are more featureful, some have better graphics, some are specialized into a genre etc.
While some genres need almost no code, others are full of pages of tricky game logic. And that is without counting into account the engine code.
Further, Unity had nothing to do with Steam being flooded. The actual reason is that
1- Valve pivoted into being a distribution platform,
2- they stopped curating what goes in,
3- gaming exploded and became way more mainstream (along with streaming, e-sports and programming, all which helped a lot), and
4- people started making money selling courses, degrees and content on how to make games.Your comment is great without that first bit.
Out of curiosity, how do you deal then with veiled ads etc. in this day and age?
I know that's not a perfect answer, but it's pretty much all we have, and the approach does have one major advantage: it doesn't need to care how veiled an ad is, or if it's an ad at all.
But seriously, those are just a fraction of the amazingness I've experienced. I don't even think the 2D platformer Portal clone is in that list - that was amazing.