This lack of senior developers also leads to the chronic amnesia our industry has, that includes reinventing the wheel every 10 years (same wheel different color). The more the things change the more they stay the same.
I'm on my late 20 and I went to my alma-mater library once a week in the past. I grabbed old books (Liskov, Bentley, Kernighan, some of the software engineering books, etc) a few times and I was surprised to see methods or techniques that people use these days and considered cutting edge were being discussed in a THIN, DUSTY, paper with YELLOW-ish color books.
There are plenty wisdoms in those old books yet these days people were gung-ho with the latest greatest.
I'm not saying that the latest greatest aren't advancing our thought. It's just that there are more noises than actual wisdoms.
The noise I contribute to the general low average age of our industry. Lots of strong opinions combined with inexperience leads to flame wars.
More people would be better off doing their 10 000 hours of practice before raising their voice in an opinion.
That said I always thought the guild model of apprenticeship would be a better model for developing our craft. But I'm not sure how one would go about setting something like that up.
It's a problem in a lot of engineering fields. Even when older engineers stay around, the knowledge often isn't transmitted. Sometimes it's because you have gaps, where a company grew a lot, then plateaued with no new hires for a while, then grew again later. That gives you old and young cohorts. Not to be an old fogey, because the young engineers bring a lot to the table too, especially familiarity with newer technologies. But a lot of knowledge slips through the cracks. I'd imagine it would just slip faster when people retire in their 30s.
Might be worse in tech because you can reinvent the wheel, with less physical infrastructure that has to stay around. Or maybe that makes it better. I started writing up something about some of my experiences with institutional amnesia causing problems when you still have the old wheel around, since old factories don't go away easily. But it ballooned into a much longer story so I threw it up somewhere anonymous and submitted separately: http://news.ycombinator.com/item?id=3311015
I think the circular nature of tech is a good chunk of why older developers are seen as "stuck in their ways" or unwilling to learn.
At a certain point, you realize that people in tech are constantly promoting the holy grail. So you buy into it, learn it, and then see that it has different, or even very similar weaknesses to the old tech. Also, you begin to realize that domain experience is a very powerful thing.
So, I would rather be happy, operating in the 'zone' with the platform I know, doing 20% more work to hack around sub-optimal language feature, than spend time in distress, trying to wrap my head around a completely different platform, all because someone on my team is obsessed with the new flavor.
