I am 65 and trying to retire, but I have one client now plus my previous employer wanting me to help out. I could probably get as much consulting work as I want, at a rate significantly higher than I could get when I was 50. I doubt anyone would hire me for full-time work, but I don't really want that. And I'm not programming for hire - I do that for fun. I expect more leverage than that. I figure if someone wants me to write code, I'm not charging enough. You need to find ways to sell your experience.

We could easily gain back much more than the "lost" 34% by cutting our meat consumption. If you are really concerned about your personal impact on the planet, you can do more by cutting out the meat - all of it - than by buying cfl bulbs or a high-mileage car or whatever. There's a nice little book published back in the early 70s called "Diet for a Small Planet". It's more relevant now than it was back then.

I waited until my kids were grown up and out of the house. But seriously, the biggest problem I often have is knowing when to quit for the day and when to take a day off. You can easily get into a 7X12 or more work situation if you are not careful. If the weather is good, I try to get outdoors for an hour or so every day. If it's really good, it may be four hours, which I then have to make up in the evening.

I am an author and an editor of a journal that could use a higher impact factor to get noticed. But I have never been "encouraged" to add a reference that was not clearly missing (there have been one or two of those, due to inadequate research on my part), and as an editor I have never asked for additional references except in cases where there was clearly prior work that the authors should have been aware of and should have cited, usually because the missing references actually showed the results the authors were claiming as new contributions. So I think this is a case of extreme self-selection, and perhaps a particular field or journal where some practices need to be examined. I just don't see it in Computer Science, Economics, or related fields where I read and publish.

At Minnesota, where I teach, and where I did my Masters and Ph.D. theses, students and faculty own copyright to their original work, including scholarly work (papers, theses, etc.) and original course materials. See http://policy.umn.edu/Policies/Research/COPYRIGHT.html for details. My understanding is that this arrangement is extremely common in the U.S. I am a strong advocate of open source and creative commons, but in this case I would encourage you to simply copyright your thesis. That does not mean others cannot use it, it just means that they must attribute the work to you, and cannot claim it as their own.

I have been in the pool once in almost 40 years. But lawyers don't seem to like professionals on juries, and they really don't like professors or scientists. Too skeptical, I suppose. So I was dismissed without even being questioned for every case.

How about Dijkstra's Algorithm?

Yes, indeed, there is a huge untapped frontier in software, both for making discoveries (programs that find and fix their own bugs, for example), and for doing interesting research in other areas. One place to look is computational economics - building complex market scenarios and figuring out how they work. As far as I know, nobody did that before the big mess in California's energy market in 2000. See the Trading Agent Competition or Leigh Tesfatsion's summary of Agent-Based Computational Economics.