I'd be perfectly fine with getting rid of the half standard with a two-tier marathon qualifying standard that was relatively softer than it is today, but I don't necessarily have a problem with a half standard, either.
Look, most of the 64min guys are going to go out in 66-67min and blow up. I don't think that's so much an effect of marathon inexperience, but of the nature of the Olympic Trials. It's a day to "swing for the fences" as it were. Go out there, run with the group, take your shot-- you'll probably fail, but failure can be useful. In our modern "if you're not first, you're last!" point of view, saying "it's ok to fail if you learn something useful" is almost blasphemy. I disagree. Ron Clarke once observed something similar in Australia--that youngsters are encouraged not to try if they can't win, and that's much worse than failing.
An acquaintance of mine who was a low 28min 10k runner for the US back in the early 2000s once told me he saw little point in going to the Trials and running a modest PR for 10th. True to his word, in his last Trials, he ran in the lead group until he couldn't anymore. It probably cost him a 2:12, which would have been a PR, but PRs aren't really the point of the Trials. He was older, this was his last realistic shot--unlike, say, a much younger guy, who might be looking at this race as experience for the next go-around.
Another point I've noticed is that Americans seem to put a lot of emphasis on finishing, then getting faster. That is, for Americans, the "ideal" 2:08 marathoner might debut at 2:11, then run 2:10, 2:09, and so on. I recall reading an interview where a Kenyan athlete (this was years ago on the now-defunct MensRacing.com site) thought this curious. He observed it was more common in Kenya for an athlete who wished to run 2:08 to make it maybe to 25k at that pace in his first attempt, then 30k, 35k, and so on. In America, an athlete with 3 DNFs then a 2:08 would probably be looked on with suspicion and derision. You can imagine the Letsrun comments now- "oh so Bill Shorter couldn't even finish a marathon last year, now he's a 2:08 guy? Blah blah blah doping."
Regarding specific marathon standards, I admit that it makes little sense to me for the Trials A standard to also be the Olympic A standard. The idea that the Trials is ONLY to select 3 athletes to represent the US at the Games is also strange to me, since I think the history of the event backs up the idea that the Trials is much more than that. As stated earlier in the thread, the head honchos at USATF seem to disagree, and would prefer a road race with maybe 20 athletes with a realistic shot. It seems to me a tremendous waste of resources to go through the trouble of staging a marathon for 20 people, as well as in opposition to the idea that the more milk one has, the more cream will rise to the top.
All this being said, I think the more-or-less sensible thing to do would be to have a reasonably tough A standard where qualifiers' expenses got covered, then a relatively "easy" B standard aimed at promoting depth and encouraging runners to run more marathons.
A standard, expenses paid: 2:15 (tougher than is ideal for expenses covered, I think, but let's throw a bone to the Mens' LDR committee)
B standard, $75 entry fee to defray costs: 2:24
I know people will disagree with an entry fee for the slower guys, but $75 is not a bad deal compared to the cost of many marathons, and, let's face it, if you're running sub-2:24, you're likely getting some entry fees already covered by your local marathons anyway and $75 isn't going to put you on the streets.
This two-tiered system represents, I think, a fair compromise between the two camps of "TRIALS ARE ONLY TO SELECT THREE!!!11" and "historically and practically, the Trials have always been about promotion and development, too."