Well, I think what it takes to be successful is a function of what the market expects. I believe for a while the market expected AAA quality and the tools to make that deliverable from an individual were lacking. I'm not sure from the late nineties to about 2010, anyone but the most talented could deliver to expectations while working alone.
> it's good that more people can make games, and it's easier to learn how to do it. It's bad that this bred a mass influx of shovelware and people who don't want to learn.
I agree. I think the shovelware is expected, and to some degree (at this point) necessary as a corrective measure on the industry. It's a useful indicator that it's not necessarily as easy as it looks, and offers examples of what not to do.
Also, it's sort of like blogs. Many more people distribute their writing now than before, and that has resulted in a log of crap flooding the internet. Then again, I'm sure at least a few of the people I read regularly either wouldn't write publicly or I wouldn't know about them if internet publishing hadn't lowered the barrier to entry so much.
As for people not wanting to learn, I think a portion of that is just people that haven't learned yet. Some people will put out crap and give up. Others will try to learn what worked and what didn't and try again. That may be crap too, but they are on the right path to eventually putting out something that isn't crap. I think, given the amount of crap we currently see, that's a reason to be optimistic. ;)
It was possible to be a single-person game developer since forever. The first games were made by individuals, and at no point there was a time where a single individual coudln't make a decent game. Of course, that requires knowing what battles to fight - you won't make an AAA shooter yourself, but you _might_ make a fun shooter, or you might make a small AAA-quality level alone.
Where the barrier to entry seems to be lowering is in technical skills per se. Each new generation of tools enables people with less knowledge to take on more ambitious projects. 15 years ago, building a 3D world required some understanding of 3D graphics, today you can slap something in Unity from asset store, not stopping at any point to understand what's going on.
I'm not entirely sure whether I think this is good or bad. So let me put it this way: it's good that more people can make games, and it's easier to learn how to do it. It's bad that this bred a mass influx of shovelware and people who don't want to learn.