Believe it or not, but when I wrote 'some confidence in Bitcoin' I meant literally Bitcoin and not any particular exchange or rehabilitation project.
Right now, Bitcoin does need as much good public relations as it can get, and a quick and full compensation for the small-fish victims of the MTGox disaster would constitute good PR indeed.
A large group of angry customers who get back their funds means many less people who have (somewhat understandable) motivation to actively criticize and oppose the whole Bitcoin project.
Besides, the idea of compensating smaller creditors with some preference over bigger creditors appeals to my sense of justice. (Despite the fact that I do not belong to the former group).
The idea of prioritizing small clients is not bad by itself, but that could perhaps be proposed to the liquidation court. AFAIK, Japanese liquidation law already establishes a "socially just" priority scheme: first any unpaid salaries and benefits, then actual creditors (banks, suppliers, landlords, contractors, utilities, etc.), and lastly investors. I don't know where MtGOX clients would fit in this priority ranking.
The Sunplot proposal obviously cannot change the priorities of employees and actual creditors; the court could not possibly agree to that. It apparently inserts the new owners and managers, as well as the expert investigator (Mr. Karpelès?) into this ranking,
before the clients. Isn't that so?
IANAL, but IIUC rehabilitation requires that you actually become solvent; You can't just unilaterally cram down your debts to everybody, even if the majority of creditors agree. An example of how this has actually worked would be Intrade, where the founder died trying to climb Mount Everest and the new management inherited a company without the money they should have had. You can follow what happened here:
http://www.intrade.com/v4/home/latest-news/The key point about this is that they need to get creditors to agree to give up their claims on enough of the debt to make the company solvent. In Intrade's case they were missing less than half of the money, and they had a good chance of getting the rest back through litigation (which they ultimately did). Since it's a voluntarily agreement between the company and each individual creditor, they can apply whatever terms they want to, as long as each creditor is prepared to accept them. Apparently Intrade didn't bother asking people owed less than $1000 to agree to forbearance.
The problem here is that if you've only got the order of 20% to 25% of what you owe, it only takes a small proportion of creditors to say "No, I insist that you pay me in full" to torpedo the whole thing. Presumably that (plus the general mendaciousness and shadiness of the people involved) is probably why the court hasn't been taking the idea very seriously.